UNIVERSITY OF SURREY

School of Management

Financial disclosure and market reaction in a crisis situation

by

A. Voeiker

Submitted in part fulfilment of the requirements for the degree of Doctor of Business Administration

May 2011

ABSTRACT

The research question of this study is whether or not companies can improve the

relationship with their investors by voluntarily extending the scope of their risk disclosure. The novel aspect of this research is that it analyses the relationship between companies and investors in a unique crisis situation.

This thesis explains and tests the relationship between listed European insurance

groups that disclose risk information in their annual financial statements and the

reaction of the capital market during the recent sub-prime loan crisis. It covers the

disclosure periods 2007 and 2008 of thirteen insurance groups that file their financial

reports according to International Financial Reporting Standards (IFRS). Based on literature it is assumed that during a crisis situation investors estimate the value of the

company they have invested in and decide whether or not they sell their shares or hold

on to them. Research studies indicate that investors demand company-specific information about the investment risk, measurement of financial instruments and forecasts that are up to date and specific to the crisis situation.

To analyse the relationship between listed companies and investors the thesis

measures their disclosure performance and the capital market response towards it. A

self administered risk disclosure performance index is used that is based on

accounting disclosure requirements and specified to the crisis. It measures the

discretionary disclosure of companies about mortgage portfolio-based financial

instruments. These instruments caused financial distress of banks that spread to a

global economic crisis. The thesis measures the reaction of the capital market towards

the disclosed risk information by applying the event study methodology at the peak of the crisis situation.

The statistical analysis of the relationship indicated that voluntary risk disclosure is not

well recognized by investors as reported in the literature. It found out that investors had

recognized those listed European insurance groups that have been most severely

affected by the financial crisis. Investors came to that conclusion even before the

researched companies disclosed risk information about their financial instruments in

annual group financial statements.

The statistical research has been extended using structured-interviews with experts that represent significant users of financial information. This qualitative approach seeks to verify and explain the statistical results.

The results of the interviews with professional users of financial information such as

financial analysts, managers of rating agencies and bankers have shown that they do

not depend on risk information from annual financial statements of respective

companies. They seek to get risk information in the event of a crisis by directly

addressing the management of insurance groups. This enables them to gain

comments on the impact of upcoming problems in the financial market from these

managers.

These results imply that theories about the relationship between disclosing companies

and the capital market are different in crisis situations when psychological factors

influence the behaviour of both companies and the capital market. These factors are difficult to separate and need further research.

Practical implications of this research study arise for regulators and standard setters. Regulators should be aware that communication between management and

professional users of financial information exists that exclude private investors during a

crisis situation and deprives them from company-specific and recent information. This

information disadvantage of private investors is amplified by the difficulty to understand the business model of the insurance industry. Therefore detailed and understandable information that explains the aftermath of the financial crisis on the insurance company

should be made freely and promptly available. Regulators should require the insurance

industry to do so because their discretionary risk disclosure is not sufficient to enable

private investors to make snap-shot financial decisions.

Based on the evidence that insurance groups have hesitated to disclose their risk

positions during a crisis, it is questionable whether they will comply with upcoming financial standards that require the disclosure of sensitive data about the fair market

values of assets and liabilities. These data can be considered equally sensitive to listed

insurance groups as information about their risk position on financial instruments. Such

hesitation may be critical in other crisis situations when discretionary disclosure is

necessary from an investor point. Accounting requirements cannot add the necessary

information because they cannot be modified quickly enough to force companies to

disclose risk information that is critical in each individual crisis.

Regulators and accounting standard-setters should not rely on listed European

insurance groups disclosing sufficient risk information during a crisis voluntarily. They

should enhance incentives for insurance groups to disclose information about their risk

position to create a level playing field between institutional investors with easier access

to crisis relevant information and private investors that lack such information in crisis

situations. Investors should reward discretionary risk disclosure of companies by

investing in them.

DECLARATION OF ORIGINALITY

I declare that my thesis entitled Financial disclosure and market reaction in crisis situation fort he degree of Doctor of Business Administration (DBA) of the University of Surrey, embodies the results of an original research programme undertaken by me. I have included specific references to any other work, by me or other sources, whether published or not.

Ansgar Voelker

Date:

TABLE OF CONTENTS

Page 1. Introduction 9 2. Relevance and Definitions 15

2.1. Insurance companies and the financial crisis 17 2.2. Investors and the financial crisis 20 2.3. Relevance of risk disclosure for future accounting standards 23

3. Literature Review 27

3.1. Communication between companies and the capital market 30 3.1.1. Theories on communication ............................................................... 30 3.1.2. Empirical research on communication ............................................... 31 3.1.3. Forms and legal requirements of risk disclosure ............................... 38 3.1.4. Implications for the research methodology ........................................ 51

3.2. Motivation of companies to disclose risk 53

3.2.1. The principal agent theory ................................................................. 53 3.2.2. Corporate governance affects risk disclosure on corporate level...... 55 3.2.3. Non organizational factors ................................................................. 57 3.2.4. Implication for the research methodology .......................................... 61

3.3. Reaction of the capital market towards disclosure 62

3.3.1. Participants in the capital market ....................................................... 63

3.3.2. Psychological factors that influence the capital market ..................... 67

3.3.3. Implication for the research methodology .......................................... 69 4. Hypotheses

4.1. Development of Hypotheses 4.2. Selection of Variables

72 72

77 4.3. Measurement of Variables 81

4.3.1. Self-administered risk disclosure performance measurment ............ 81 4.3.2. Measurement of the capital market reaction variable (CAR) ............. 88

4.3.3. Change in modified equity ratio 2007 to 2008 ................................... 99

4.3.4. Percentage of institutional investors of stated equity ...................... 101

4.4. Research Methods for Data Processing 102

4.5. Data and Sample Selection 104

5. Research Methodology 107

5.1. Epistemology 108

5.2. Assumptions 111 5.3. Quantitative Methods 114

5.3.1. Testing for Normality ........................................................................ 115

5.3.2. Setting of the confidence interval ..................................................... 115

5.3.3. Correlation analysis between two variables ..................................... 117

5.3.4. Regression between multiple variables ........................................... 118 5.4. Qualitative Methods 119 5.5. Confidentiality and Ethical Considerations 124

6. Results and Discussion 126 6.1. Statistical Results and Discussion 126

6.1.1. Results of the test of normality ......................................................... 126

6.1.2. Correlations between the variables ................................................. 128 6.1.3. Multivariate regression ..................................................................... 131

6.2. Qualitative results and discussion 132 6.2.1. Experts on the implied theory .......................................................... 133 6.2.2. Experts on the research methodology ............................................. 135 6.2.3. Experts on the results of the statistical results ................................. 140

6.2.4. Experts on the future implication of research results ...................... . 142 7. Conclusions and Implications 145

7.1. Main findings 145 7.2. Implications of the findings 147 7.3. Limitation of the research 148 7.4. Policy orientated suggestions 149

Appendices 168

ü

LIST OF TABLES

Page

Table 1: Confidence intervals 112

Table 2: Test of normality 124

Table 3: Univariate analysis measured by Pearson correlation 128

111

LIST OF APPENDICES

Appendix A: Checklist 2007 for measuring risk disclosure performance

Appendix Aa: Checklist 2007 Pre-crisis Aegon

Appendix Ab: Checklist 2007 Pre-crisis Allianz

Appendix Ac: Checklist 2007 Pre-crisis Aviva

Appendix Ad: Checklist 2007 Pre-crisis Axa

Appendix Ae: Checklist 2007 Pre-crisis Generali

Appendix Af: Checklist 2007 Pre-crisis Hannover Re

Appendix Ag: Checklist 2007 Pre-crisis ING

Appendix Ah: Checklist 2007 Pre-crisis MunichRe

Appendix Ai: Checklist 2007 Pre-crisis CNP Assurance

Appendix Aj: Checklist 2007 Pre-crisis Legal & General

Appendix Ak: Checklist 2007 Pre-crisis Standard Life

Appendix Al: Checklist 2007 Pre-crisis Old Mutual

Appendix Am: Checklist 2007 Pre-crisis Zurich

Appendix B: Checklist 2008 for measuring risk disclosure performance

Appendix Ba: Checklist 2008 Pre-crisis Aegon

Appendix Bb: Checklist 2008 Pre-crisis Allianz

Appendix Bc: Checklist 2008 Pre-crisis Aviva

iv

Page

165

168

171

174

177

180

183

186

189

192

195

198

201

204

211

215

Appendix Bd: Checklist 2008 Pre-crisis Axa 219

Appendix Be: Checklist 2008 Pre-crisis Generali 223

Appendix Bf: Checklist 2008 Pre-crisis Hannover Re 227

Appendix Bg: Checklist 2008 Pre-crisis ING 231

Appendix Bh: Checklist 2008 Pre-crisis MunichRe 235

Appendix Bi: Checklist 2008 Pre-crisis CNP Assurance 239

Appendix Bj: Checklist 2008 Pre-crisis Legal & General 243

Appendix Bk: Checklist 2008 Pre-crisis Standard Life 246

Appendix BI: Checklist 2008 Pre-crisis Old Mutual 250

Appendix Bm: Checklist 2008 Pre-crisis Zurich 254

Appendix C: Scoring documents for measuring risk disclosure performance 258

-criteria applied in the checklists 2007 and 2008

Appendix D: Variables, Formulas, Results and data of the event study 296

method

Appendix E: Document for measuring the modified equity ratio 312

Appendix F: Clustering listed European insurance groups 2007 313

Appendix G: Clustering listed European insurance groups 2008 314

Appendix H: Guidelines for structured interviews with experts 315

Appendix {: Interviews with experts in the European insurance industry

Appendix la: Interview Anonymous (2010) 317

Appendix lb: Interview Noack (2010) 321

V

Appendix Ic: Interview Seel (2010)

Appendix Id: Interview Anonymous (2011 a)

Appendix le: Interview Anonymous (2011 b)

325

329

332

vi

ACKNOWLEDGMENTS

I would like to thank my wife, my parents and my family for their support. During the four year program I much appreciated the guidance and help from my supervisors Professor Dr. Chen and Professor Dr. Gilbert. I enjoyed discussing topics with Mr. Otte (DeutscheBank), Mr. Funck (GoldmanSachs), Ms. Rieger, Ms. Varney, Dr. Weber and Dr. Edelkoetter.

It has been a great, challenging but somewhat lonely time.

vii

ABBREVIATIONS

AICPA American Institute of Certified Public Accountants

AIMR Association for Investment Management and Research

CIFAR Center for International Financial Analysis Research

DAX Deutscher Aktien Index (German index of listed companies)

GRS German Reporting Standards

EC European Commission

EFRAG European Financial Reporting Advisory Group

FAF Financial Analysts Federation

FASB Financial Accounting Standard Board GASB German Accounting Standards Board HGB Handelsgesetzbuch (German Commercial Code)

IASB International Accounting Standard Board

IDW Institut der Wirtschaftsprüfer (German Institute of Auditors)

IFRS International Financial Reporting Standards

MDAX Mid-Cap-Deutscher Aktien Index (German Index of listed mid-cap companies) PwC PricewaterhouseCoopers RechVersV Verordnung über die Rechnungslegung von Versicherungsunternehmen

(Accounting Regulation for German Insurance Companies)

WpHG Wertpapierhandelsgesetz (Securities Trade Act)

vii'

Chapter 9

1. Introduction The subject matter of this research is the relationship between listed European

insurance groups that disclose accounting information and the reaction of the capital market towards this information.

Under normal business circumstances companies disclose information about their business activities and the results they have achieved during the last financial period. Although the frequency of financial reports has increased with quarterly financial

statements their function as a detailed analysis of the previously achieved financial

result has not changed. With a background as an accountant in the insurance industry

the researcher wanted to find out to what extent financial statements provide forward looking information which can be used by investors to take current financial decisions.

Therefore it is the aim of this research study to analyse the content of disclosed information as well as the structure and efficiency of the relationship between disclosing insurance groups and the capital market in the light of the recent financial

crisis situation.

Is the capital market able to determine the market value of listed European insurance

groups based on the information provided by them in their annual group financial

statements? Accounting requirements applied in these financial reports are supposed to provide information according to IAS 1 par. 9 so that investors can predict 'the

entity's future cash flows, their timing and certainty' (IASB, 2003) and be able to

summarize them in a present market value. But is it right that the share price changes with the extent of disclosed information (Bealy et al, 2006) especially in a crisis

situation when investors make snapshot decisions to hold or sell the share of a troubled company?

From the perspective of the company preparing financial statements, when and to

what extent do insurance companies disclose information about the risks they own in

the form of structured financial instruments based on mortgage portfolios which have

been felt to increase the financial distress during the financial crisis that lasted from

2007 to 2010 (Scharpf and Schaber, 2009)? Previous empirically proven research has

9

shown that discretionary disclosure - defined as voluntary disclosure that exceeds the

requirements set by accounting standards (Core, 2001) - reduces cost of capital for the disclosing company (Welker, 1995, Sengupta, 1998, Healy et al, 1999, Botosan and Plumlee, 2002). However, all these studies used a discontinued US disclosure index

and were performed under the assumption of normal market conditions such as a functional market and rationally acting shareholders. All these assumptions do not meet the conditions of the recent financial crisis which has the underlying characteristic that fair values of certain mortgage portfolio based financial instruments dropped

significantly and remained low for more than one financial period of listed companies. The drop of fair value in these financial instruments affected the insurance industry,

which is a pool of investments by its business model, most severely and uniformly whereas the effect on banks depended on their degree of activities in the securitization of mortgages that differed according to their location and business activities. By focusing on the risk disclosure performance of listed European insurance groups this

research closes a research gap because financial institutions are often excluded from disclosure research studies as a result of their specific accounting requirements.

By focusing on a crisis scenario the objective of the research is to eliminate other confounding factors that influence the relationship between disclosing companies and the capital market reaction expressed in their share market value. This research approach will deal with criticism by Healy and Palepu (2001) who hold that relationship models between disclosure and capital market are too simple and contain too many restrictions and assumptions.

Another way to focus on a crisis situation is current research by Baek et at. (2002) that

studied how companies disclosed their bad risk and how the capital market reacted in the 1997 Korean finance crisis. These researchers found out that Korean firms with higher disclosure quality suffered less in their share value during the crisis. The

research question is whether or not this reaction of the capital market has continued in

the recent global financial crisis. However Baek et at. (2002) used simple proxies to

measure disclosure performance. Another objective of this research is to measure the

disclosure performance with self-administered risk disclosure performance indices that

can be considered scientific standard as applied by Kajüter and Winkler (2003),

Bungartz (2003), Vielmeyer (2004) and Khana et al. (2004) instead of proxies.

10

Therefore this disclosure research is performed using more specific disclosure

measurement methods and recognizing better risk disclosure performance by listed

companies due to the application of a risk-related accounting standard 'financial

instruments: disclosure' IFRS 7, (IASB, 2005).

The recent sub-prime loan crisis offers a unique opportunity to analyse the effect of disclosure on investment / divestment decisions that financial market participants make during a period of time which is characterized by a uniquely non-transparent risk

situation of financial companies (Baetge et al., 2008). Baetge et al. (2008) blame the

spread of a regional crisis to a world-wide crisis on the chosen tentative step-by-step disclosure by banks of their financial distress. This lack of transparency leads to a

world-wide crisis of trust between banks themselves and between banks and investors.

Although insurance companies had not been directly affected by the sub-prime loan

crisis, in the aftermath of the crisis they nonetheless had to cope with investors'

anxieties and indirect effects on the valuation of their investments as the crisis

continued. Therefore it is to be examined whether or not risk disclosure during the

crisis has become a dominant factor that influenced the capital market.

Can companies improve investors' trust in them by voluntarily extending the scope of their financial disclosure? Disclosure is regulated by German national and international

accounting rules for listed companies. Originally only German accounting rules

required extended corporate information about risks in a separate management report but international accounting rules have also adopted this approach recently (Kraft and Nolte, 2005). Do German national and international regulations provide a sufficient level of disclosure in order to compare risk disclosure of companies from different

national backgrounds? Are there economic reasons that motivate companies to

exceed their mandatory disclosure requirements?

When disclosing information about their own risk position insurance companies fear

that this would be viewed critically by the financial market and their customers, i. e. as a

sign of weakness in their business model (Meyer et al., 2004). This understanding has

been influenced for the better or worse by the sub-prime loan crisis. When it peaked

during 2008 investors were unsure about the extent to which financial institutions were

involved in that crisis. Without sufficient information about the risk exposure of the

companies in which they owned shares or about the financial consequences for each

11

company the share market failed to provide fresh capital so that banks and insurance

companies in need of capital had to first look for sovereign wealth funds instead

(Anonymous, 2008a) and then for governmental aid that few of them received.

Although the relationship between disclosure and the capital market can be considered

a mature research field according to Verrecchia (2001) the research objective to

analyse the empirical effect of disclosure on the share performance of companies has

rarely been explored due to declining data-bases, changing accounting standards, industry-specific regulations and excessive cost or improper computer tools to analyse

qualitative data in financial statements (Daske and Gebhard, 2006). That is why this

thesis analyses risk disclosure requirements and searches for incentives to exceed them, both for the company as information provider and for the investor as the user. To

broaden the previous, predominantly economic approach, psychological motivations

are embedded into this research to explain investor behaviour during times of crisis. The novel aspect of this thesis is to apply existing theoretical and empirical studies to

the recent sub-prime loan crisis situation.

Existing studies of investors' behaviour in crisis situations are either rare or restricted to

regional markets (Baek et al, 2002) or certain institutions like US-Banks (Jain and Gupta, 1987). This thesis intends to close this research gap of risk disclosure and its

effects on the capital market by applying different research tools to a sample group of listed European insurance companies during the recent market crisis. There are, however, a few observations to be made in respect of the scope of this work. Due to

frequently changing disclosure requirements, this study is restricted to a relatively short time period with stable disclosure requirements and it is focused on a special industry

that has been uniformly affected by the sub-prime loan crisis. These restrictions limit

the sample size of analyzed companies in order to compare their risk disclosure

performance with scientific diligence.

The researcher acknowledges the validity of the content and of the measuring instruments as defined by Drucker-Godard et al. (2001) in this study. Through various

sources of literature the variables that influence the risk disclosure behaviour of listed

European insurance groups and the reaction towards it by the capital market have

been indentified. The measurement process of the variables has been made

transparent by disclosing every measuring step in the appendices of this study. The

12

results of this research study can be generalized based on the restrictions that only listed companies of a certain industry have been analyzed. In regard to the willingness

of companies to disclose 'sensitive' information the insurance industry can be

considered as a good proxy for companies from other industries because according to

Meyer et al. (2004) insurance groups fear that disclosure of risks can be judged

critically by investors and customers as a sign of weakness in their business model. This enables the researcher to identify tendencies of management to detain ̀ sensitive' information. These traits can be considered by regulators and standard setters in their

attempt to provide all investors with relevant information.

Another aspect of validity is that predictions based on one research instrument need to be confirmed by another. The statistical analysis of data from a limited sample size can lead to misleading results that should not be generalized. This problem has been

addressed in this research study by holding structured interviews with experts in the field of the European insurance industry. The intention was to verify statistical results in

order to improve the research results by the research concept of triangulation. By using two methods to analyse the research question the consistency of findings generated by the one method is validated by the other one (Patton, 2002).

The overall objective of this thesis is to provide insight into the relationship between

risk disclosure contained in the annual group financial statements of 2007 and 2008

within the European insurance industry, their share performance in contrast to an index

at the peak of the crisis and the effect the financial crisis had on the insurance group

using their modified equity ratio as a proxy.

This research objective has lead to the following structure in this thesis: The relevance

of risk disclosure for insurance companies, investors and future accounting standards

as well as the definition of terms are set out in chapter 2 'Relevance and Definitions'.

Chapter 3 'Literature Review' analyzes the theoretical aspects and empirical studies in

the literature on the relationship between disclosing insurance companies and the

capital market. Chapter 4 'Hypotheses' contains the research question and the

hypotheses that derive from it. The literature review shows that certain aspects should be considered in selecting and applying research instruments. Chapter 5 'Research

Methodology' sets out the research methods used to analyse the relationship. This

chapter presents the epistemological standpoint of the researcher and assumptions

13

made in this thesis. Additionally the quantification of the variables risk disclosure

performance, the capital market reaction, the effect of the crisis on companies and the

degree of institutional investors are described. This chapter also explains the use of

qualitative research methods. In chapter 6 'Results and Discussion' the results of the

quantitative and qualitative research instruments are presented. Chapter 7

'Conclusions and Implications' sets out the findings, limitations of this thesis and includes impulses for disclosing companies, standard setters and regulators as well as investors.

14

Chapter 2

2. Relevance and Definitions The purpose of this chapter is to show the relevance of the recent financial crisis for

the insurance industry by answering the two following questions: What has been the

key effect of this crisis and why has the insurance industry been affected more than

other ones by this crisis? The next step analyzes the type of information required by

investors during a crisis. This is based on influential studies and leads to the definition

of terms used throughout this thesis. The results of this chapter are used to analyze how the effect of the crisis is disclosed in the financial statements in subchapter 3.1.3.

Based on this the risk disclosure performance of insurance groups is measured in

subchapter 4.3.1.

According to Reinhart and Rogoff (2009) a crisis is characterized by a major decline in

economic performance for an extended period of time. The origin of the financial crisis that is the topic of this thesis consisted of deferrals of payments in mortgage loans of financially weak private individuals in the United States of America during the years 2007 and 2008 (Scharpf and Schaber, 2009). This led to financial troubles of US

mortgage financial institutes. Also affecting other US-banks the crisis extended to the

share market. The US share market lost wealth of about $8 trillion between October

2007, when the share market reached an all-time high, and October 2008

(Brunnermeier, 2009).

Thus, the key effect of this crisis has been a sudden decline of the fair values of investments that lasted longer than one financial period. Falling prices of bank shares

and other equity investments increased the risk of insurance companies as significant institutional investors. According to Scharpf and Schaber (2009) investments that were first and most directly hit were Asset Backed Securities (ABS), Commercial Mortgage

Backed Securities (CMBS), Residential Mortgage Backed Securities (RMBS) and

Collateral Debt Obligations (CDO) and other financial instruments directly connected to

the falling US-housing market. Corporate as well as sovereign bonds were hit indirectly

by their lower fair value. Lower market prices of investments lead to immediate losses

in the profit and loss statements as well as reduced equity positions in the balance

sheets of listed European insurance groups depending on their categorization of

15

investments according to the international accounting standard IAS 39. The default of

bond issuers hit listed European insurance groups because they are required by

national regulations to hold a part of their investment portfolio in corporate and

governmental bonds. This situation is characterized as credit risk. 'Default' is

recognized in the financial statements as a loss that is subtracted from equity in the

same financial period. Additionally markets for certain financial products did not function any more so they were unable to value structured financial instruments any

more (Brunnermeier, 2009). The nature of these instruments and the insecurity about their price or about the assumptions that were made in order to calculate their fair

value by using mathematical models spread the crisis globally and affected the global

economy.

Why is this financial crisis particularly relevant for the insurance industry? Insurance

companies are unique because the handling of risk over a long period of time is the

main economic purpose of this industry. It shares risks like investment risks with other financial institutions, however the distinctive criterion of the industry is covering insurance related risk and matching it with investment risk (Großer, 2001). In order to

provide insurance coverage insurance companies manage the premiums they receive from customers and the claims for which they have to indemnify their customers by

matching the timing and level of cash inflows from premiums and the cash outflows from claim events (Rejda, 1998). They invest cash inflows deriving from insurance

premiums in short and predominantly long term financial instruments. After the

occurrence of claims based on an insured event these investments are sold to create

cash outflows available to settle the claim of the insurance customers. Because most likely there is an extended time period between the cash inflow of insurance premiums

and the cash outflow to settle claims, insurance companies are pools of capital investments (Farny, 2000).

The following subchapters link the effects of the crisis such as a prolonged decline in

the fair value of investments, insecurity about measurement of these investments and

the future development of companies to the insurance industry. Subchapter 2.1.

describes three characteristics of insurance companies in order to answer the question

why the insurance industry has been affected by the crisis. Subchapter 2.2. analyses

what kind of risk information is important to investors during a crisis situation.

16

Subchapter 2.3. distinguishes certain terms used throughout the thesis and puts this

research in a wider perspective of upcoming accounting regulations that are regarded

as a paradigm change in the European insurance industry.

2.1. Insurance companies and the financial crisis According to Schmalenbach (1953) accounting pursues the aim of an income

calculation on an accrual basis that recognises the financial period to which economic

effects belong over the timing of payments. Another objective is to present all assets

and liabilities at a certain point of time to provide a true and fair view on the economical

situation of the company (Cropper et al, 1949). To accomplish both objectives

accounting regulations on the recognition, measurement and disclosure of economic

circumstances were adopted. Especially the measurement and disclosure as part of insurance companies' accounting has been influenced by the crisis which is

characterized by a sudden decline of the fair values of investments.

Insurance companies share the following three characteristics that derive from their

business model, that are considered in their accounting and that are linked to the

financial crisis: They are pools of capital investments, they are guarantors of financial

performance and they have a long term orientation in their business activities.

Starting with the first characteristic the primary focus in the accounting of insurance

companies are investments that are funded by premiums paid insurance customers. They are bought, measured over the period during which they are owned by the

insurance company and then sold to provide the necessary cash flow to settle insurance claims. The measurement of investments can be either prudent or

economically focused. Accounting requirements differ in their way to measure investments. To demonstrate these differences the thesis lists German and

international accounting requirements. Whereas for example German national

accounting regulations allow the valuation at historical cost even when the market

value of investments increases above their costs according to § 253 and § 255

German Commercial Code (HGB), international accounting regulations such as

International Financial Reporting Standards (IFRS) prefer that market related increases

be shown either in the profit and loss accounts or in equity according to IAS 39 par. 46.

These preferences are documented by the restrictions that are attached to other 17

categories for investments with differing valuations. In the opposite case of falling

market prices national German accounting regulations tend to disclose lower prices than historical cost of investments only when the fall in fair values is prolonged

according to § 253 (2) HGB. International accounting regulations on the other hand are faster in writing down the book value of investments to their lower market value (IAS 39

par. 58)

The risk of a lower book value of investments is epitomized in the market value. The

market value measurement of investments has a significant impact on the accounting of insurance companies. The extent of the impact differs because it can affect yearly results as being part of the measuring aspect of accounting or only be a supplemental information without earning effects being a disclosure aspect of accounting. This

accounting treatment differs by German national and international regulations. Whereas national German regulations refer to historical costs and present fair market values in the notes international accounting regulations such as AS 32 par. 11 and AS 39 par. 9 refer to the market value with effects on the annual results.

Therefore, the market value of investments can be part of the measurement process of

assets in the balance sheet or it can be an additional disclosure outside the balance

sheet and without effects on equity or profit and loss account. Confusingly, both

accounting ways are named fair value accounting in the literature (Barleu and Haddad,

2003, p. 387). It is therefore important to distinguish between fair value accounting with

effects on equity and profit and loss accounts and accounting without them. Therefore,

this thesis presents special definitions in chapter 2.3.

The measurement of the market value of investments is the responsibility of the

company, and it can differ according to methods chosen and assumptions made by the

insurance companies. One assumption made by the standard setters in requiring the

measurement of assets with their fair market value has been the existence of a functioning market that could provide a price at any given point of time for every asset. Thus, the risk of incoherent market value measuring increases during a crisis situation

when companies tend to change their assumptions, be it as part of a change in

accounting policies or due to the lack of functioning markets and the resulting lack of

prices provided. This phenomenon of changing assumptions has arisen during the

18

financial crisis. It caused the IASB as an accounting standard setter to publish

guidance on this topic at the height of the financial crisis (IASB, 2009a).

Aside from market valuation of investments also obligations which insurance

companies have against their customers can be measured at market value. Although

these obligations are not traded on active markets so far, their market value can be

calculated based on mathematical models. This is the idea of an ongoing discussion

about the accounting of insurance contracts which will be discussed in chapter 2.3.

The second characteristic of insurance companies is their guarantor status. E. g., in life

insurance contracts companies guarantee future death or surrender benefits to their

customers. These amounts are calculated based on the assumption of a certain investment rate that can be met by the average return on investments of the life

insurance company. As a guarantee this interest rate is locked in as part of the life

insurance contract. The life insurance company recognize as a guarantee provided

and records it as a liability for future policy benefits (Mayr, 1999). If insurance groups

with such guarantees are confronted with lower investment income they have to fund

the difference between higher guarantees and lower generated investment income

with their own money by lowering their equity ratio. These guarantees - common

especially in German life insurance contracts - amplify the crisis effect on the equity

ratio for these insurance groups even more. In a financial crisis these guarantees add financial stress to insurance companies because the guarantees are steady expenses that are not compensated by fluctuating investment income. The investment income

decreases because of depreciation and impairments on investments due to their lower

fair market values that reduce income of the period. Also, investments are harder to

sell during a crisis so that higher expenses cannot easily be compensated by gains that are realized by their sale. This leads to liquidity problems of insurance companies

and requires them to use their equity to compensate for the expenses deriving from

financial guarantees. With a prolonged reduction of equity in a crisis the companies

can face insolvency.

The third characteristic of insurance companies is their long term orientation. They can be held liable for claims from their insured customers even after years. That

responsibility affects their financial strategy and their time preference in choosing

investments (Farny, 2000). An accounting implication of this characteristic is that future

risks are recognized in the accounting for investments of insurance companies as a

19

discounted receivable amount. This time-based measurement is an accounting treatment that can be influenced by a financial crisis. The long-term perspective

exaggerates negative effects on the present value measurement because the

assumptions that are implied in the fair value of long term investments are based on

recent financial conditions. Therefore in a crisis situation the fair value of long term

investments decreases more rapidly than short term investments.

Insurance companies are characterized by their investments measured at market value, a guaranteed return of life insurance contracts and their future long term

orientation. As a result the insurance industry is more sensitive to a rapid decline in the

market value of investments during crisis situations than other industries. The next chapter presents three approaches to find out what kind of information is relevant to investors during a crisis situation.

2.2. Investors and the financial crisis The risk situation of insurance companies has been significantly changed by the

financial crisis with its sudden decline of fair values of investments. This change had to

be recognized in their financial reporting because due to its characteristics the risk

position of insurance companies worsens with a prolonged decline in investment value.

What kind of information is demanded by investors during a crisis situation? Three

approaches can be used to answer these questions: A regulatory approach analyzes how regulations are altered in response to the credit

crisis. Such changes indicate what information has not been disclosed by the

companies. New regulations demonstrate what information were relevant and required

as the crisis continued.

The long term investor approach searches for types and formats of information that are

relevant for investors. This approach is based on influential surveys among investors

that identified perpetual risk information requirements of financial market participants.

The situational approach seeks to find situational information that is required by

investors. This thesis tries to find key words from media that provide this type of

information and are relevant and special to the crisis from an investor point of view.

20

The regulatory approach has to focus on international regulations because of their

flexibility. Unlike German national regulations that have not been altered international

regulations changed as a reaction to the credit crisis in March 2009 with the

amendment of IFRS 7 and in May 2009 with the standardization of fair value

measurement (IASB, 2009a and IASB, 2009b). The amendments required additional

disclosure of companies about their methods of measuring risk and explicit statements

about the group's liquidity risk. Such additional information had to be disclosed since

2008 in the notes of consolidated financial statements. These corporate specific data

can be considered important for investors because they were additionally required by

regulators in the financial crisis.

Although these data were situational influential studies by AICPA (1994) and by

EFRAG (2009) document that the information requirements of investors are almost

constant over time. This long term investor approach is based on the assumption that

information demand increases in importance to investors during crisis situations. The AICPA study - later in literature named the Jenkins Report - resulted in

recommendations to improve US business reporting. It is based on a survey that

collected the opinions of investors and creditors as primary users of financial

statements. The users recommend to disclose financial and non-financial measures to

explain long term developments of companies and rapidly changing situations. As a

result of these studies it became apparent that users have a high interest in being able

to identify trends, the substance of transactions and the significance of risks. This risk

information demand can be seen as a prerequisite for active investors. Provided with

relevant situational risk information they want to be enabled to make investment or

divestment decisions in the financial market. As a result they demand risk disclosure

on uncertainties over the valuation of assets and the economic substance of complex

capital instruments (Elliott, 2007).

The EFRAG study (2009) is based on a survey of investors, creditors and financial

analysts. Whereas the Jenkins report was restricted to the USA this study covered a

sample'of European countries. The survey was carried out at the end of 2006 before

the recent financial crisis. The assumptions of the Jenkins report that investors and

creditors use risk information in financial statements to make their 'hold/buy/sell'

decisions have been empirically confirmed by the EFRAG study. Similar to the Jenkins

report the surveyed user groups of the EFRAG study want more risk information about

21

risk management and additional prospective financial information. Empirical support for

this demand was provided by another corporate reporting survey conducted in 2007

(PwC, 2007); this survey specifies that investors request risk information about the

future economic perspective of a company in respect to management forecasts. Along

with this forecast the PwC survey found out that users of financial statements demand

risk information about the current investment risk associated with the type and the

valuation of investments. Summarizing the long term investor approach it can be

empirically supported that investors demand for certain types of risk information is

constant and increases during a crisis. The relevant risk information concerns

prospective developments of the fair value of financial instruments that companies are invested in.

The risk situation of every company is different. How can this thesis compare the risk disclosure among financial statements issued by various listed insurance companies? The situational research approach that analyzes a group of companies must focus on

risks that are common to all companies. Research by Chen (2007) supports this

approach by using bank reports to provide key words that are associated with investment risk. Using this approach risk information from different companies can be

compared and analyzed in regard to the frequency of these keywords. This approach is supported by findings of Coval et al. (2009) on the recent financial crisis. They found

out that structured financial products which are the heart of the financial crisis are

affected by systematic risk factors to an exceptional degree. They found out that risk information in the form of corporate explanations on systematic risk factors are

extensive in financial statements and connected to key words. The situational

approach states that risk information of companies with the same risk position is similar by containing common key words. This characteristic can be used in the research

methodology to identify and compare risk information.

Based on the aforementioned approaches there is a shared understanding that risk

information does matter to investors. The information should cover trends and other

up-to-date developments to companies and the risk that is associated with them. This

demand can be specified to the recent crisis situation. Risk that derives from certain

type of financial instruments and the effect of a sharp drop in the fair value of these

instruments are relevant to investors. This risk has an effect on the valuation of assets

22

and liabilities in financial statements. Additionally this effect must be described in the

notes of consolidated financial statements that are analysed in this thesis. The

accounting effect occurs because no prices for certain financial instruments were

provided by the financial markets. To substitute the prices companies relied on

mathematical models for the valuation of their investments. Consequently investors

need to be aware of the risk that the valuation of investments might affect the financial

results of the insurance companies. Coval et al. (2009) point out a characteristic of the

recent financial crisis that a small error in the assumptions to measure structured financial instruments can lead to a great misinterpretation of their probability of default.

According to IAS 1 par. 118 (IASB, 2003) `it is important for an entity to inform users of the measurement basis or bases used in the financial statements because the basis

on which an entity prepares the financial statements significantly affects users

analysis'. Therefore, the measurement assumptions of financial instruments are important factors that need to be disclosed to investors in order to enable them to

evaluate the likelihood of a decrease of the investments' fair value and the value of the

company.

2.3. Relevance of risk disclosure for future accounting

standards This subchapter defines risk disclosure specific to the crisis situation analyzed in this

thesis. The relevance of risk disclosure in the context of upcoming International

Financial Reporting Standards (IFRS) is demonstrated.

In the accounting literature there is no definition of crisis that relates to the overall economic situation. From an accounting point of view a crisis is specific to the

individual situation of the company and the resulting risk situation is defined by Rejda

(1998) as a significant deviation from the expected outcome of a prediction. Risk

therefore has a company-specific aspect because expectations vary from company to

company. To broaden this definition and to ensure a comparability of risk disclosure by

companies an extended risk understanding needs to be considered than that one of

accounting rules. This includes risk information that is required and that companies

provide voluntarily in excess of rules. In order to be comparable risk disclosure can

23

only be analyzed from companies within the same industry, with similar business

interests and a comparable business structure.

In the following the relevance of empirically researched risk disclosure for the

implementation of future accounting standards is described. To distinguish between

risk disclosure as the research subject and risk accounting as a subject of future

regulation by international accounting standard setters these terms are subdivided by following definition:

Risk disclosure in this research study is defined as all information disclosed by insurance companies, whether required by German national or international

regulations or voluntarily disclosed, that reveal the financial situation of investments

and their measuring process as well as the future view on corporate positions of risk and chances. Risk disclosure takes place predominantly in a qualitative form. When

quantitative risk is disclosed it does not have to affect the balance sheet or the income

statement in order to meet this definition. The definition of risk disclosure for the

purpose of this thesis differs from the one by Vielmeyer (2003). His general definition

of risk disclosure based on required accounting information should be extended and specified in so far as risk disclosure contains risk information that companies give above the required level as well, and it should include information that is specific to the

changing risk situations of a financial crisis situation.

Risk accounting can be defined as all information disclosed by insurance companies, required or discussed by German national or international regulators, that reveal a historic, actual or future view on corporate positions of risks and opportunities. This

definition meets market valuation of balance sheet items, quantitative as well as verbal

explanations on certain types of risk in management reports and business forecasts

embedded in financial statements. Risk accounting includes the aspect of valuation of

obligations arising out of insurance contracts that is discussed in upcoming international regulations on insurance contracts (IFRS 4 Phase II). Risk accounting therefore has an effect on the income statement and on the balance sheet. A subpart

of risk accounting is fair value accounting. It includes the market valuation of investments and obligations with effects on equity and profit and loss accounts

according to Penman (2007) as an one line risk indicator. However, it is not required by

24

today's accounting standards but included in standards that are to be applied in the future.

Results of this thesis might be useful to support companies that extend the disclosure

of their risk positions because it may be rewarded by the capital market. Such

voluntary risk disclosure might convince standard setters to pursue fair value

accounting as an one line risk indicator (Penman, 2007) which results in higher

transparency on risk instead of historic cost accounting that hides the effects of risk.

Can existing research studies help to predict how the disclosure of additional risk information, as required by fair value accounting, affects listed companies? A study by Pape and Schlecker (2009) analyzed the effect of a crisis situation on credit default

rates as an insolvency proxy depending on the accounting regime of the companies. Based on a 1998 crisis it found out that companies with fair value accounting are not considered to have a higher possibility of insolvency, although they disclose risk information to a higher extent than companies following historic cost accounting. Empirical research by Chamberlain and Magliolo (1996) supports this idea by finding

out that fair value accounting and the accompanying disclosure of risk does not

negatively affect share prices of insurance companies. Although no positive causal relationship between the form of accounting and the companies' capital market performance could be found by Chamberlain and Magliolo (1996) the statement by the FASB (2001) that the importance of disclosure is expected to increase in a changing business environment affected by the credit crisis is a starting point of this thesis.

The result of this chapter 2 is as follows: Due to its characteristics the insurance industry as a pool of investments and a guarantee provider with a long term investment

horizon is especially sensitive to changes in the fair value of investments that are at the

heart of the researched financial crisis. Investors in insurance companies demand risk information that can be compared between companies. Therefore, risk information

required by German national and international accounting standards as well as

voluntarily extended by insurance companies are recognized and measured in this

thesis. The results of this thesis on the disclosure of sensitive risk disclosure can be

used to predict behaviour of insurance companies in their future application of full fair

25

value accounting. They can provide clues for regulators about how to spread sensitive

risk information that is relevant to investors in a crisis situation. The following literature review analyses theoretical and empirical studies on the

communication and relationship between companies and the capital market. It thereby demonstrates different ways to measure corporate risk disclosure in order to find implications for the research methodology.

26

Chapter 3

3. Literature Review This research analyses how companies disclose information. This is done because

Lenz and Diehm (2010) state the prognostic value assumption that investors can

predict the financial effect a crisis has on a company based on its risk disclosure.

There is a consensus in management literature how companies in crisis situations should act concerning disclosure. They should provide accurate information as quickly as possible to stakeholders according to Dillenschneider and Hyde (1985) as well as Sellnow and Ulmer (1995). Such disclosure behaviour of companies increases the

credibility of organizations (Williams and Dolnik, 2001). Empirical research by Arpan

and Roskos-Ewoldsen (2005) shows that the credibility of an organisation influences

the behaviour of people that receive negative information about the company during a

crisis situation. In such a crisis situation an organization shall therefore "break the

news about its own crisis before it is discovered by the media" or investors (Arpan and Roskos-Ewoldsen, 2005, p. 425).

According to this thesis insurance groups should have disclosed every risk associated with their financial instruments that are based on mortgage portfolios already at the beginning of the financial crisis in 2007. They should have used every possibility to include new risk related information in their group financial statements 2008 as knowledge about the extent and its effect on each insurance group evolved.

The reason to research risk disclosure in crisis situations is the ongoing discussion

about fair value accounting in the insurance industry as a requirement to disclose

sensitive information in financial statements. The fair market value of insurance assets

and insurance obligations would disclose a true and fair view on the financial strength

of the insurance company at year or quarterly end. Such disclosure would reveal

operational and financial risk by fair value measurement and would empower investors

to consider this risk in their investment/divestment decisions. However, since fair value

measurement has only been partially implemented and might be fully applied only in

future international accounting regulations, the capital market effects of other forms of

27

risk disclosure - in this thesis risk disclosure about mortgage portfolio based financial

instruments during a crisis - are analyzed in the following literature review.

Disclosure research can be subdivided in theoretical and empirical studies. Both types

of studies have been undertaken to describe the quality of disclosure (AICPA, 1994

and PWC, 2006), to measure and compare it (Armeloh, 1998 and Vielmeyer, 2003)

and to create and verify economic models that assume a relationship between

disclosure of companies and the capital market reaction (Verrecchia, 1998, Botosan,

1997,2002 and Sengupta, 1998). It has to be acknowledged that literature agrees to

the statement that the capital market is affected by multiple factors and not only by the

publication of financial statements and other reports. Researchers critizise that single

researchers expect too much from disclosure to explain capital market reactions (Dobler, 2008) and that models used to explain this relationship are too simple with too

many restrictions and assumptions (Healy and Palepu, 2001).

This is recognized in the following literature review. This part of the thesis focuses on

exploring cause and effects between risk disclosure and capital market reactions and thus to improve the research design. Based on the definition that accounting is a social

system that consists of norms and determinants of behaviour (Harrison and McKinnon,

1986) the influencing factors of this relationship are analyzed as proven significant and later included as variables in the research design.

As the research objective is to find and define the relationship between disclosure and the behaviour of capital market participants and analyse it during a time of market distress, the literature is classified in a way that clarifies the interdependencies

between regulations, companies and investors. This structure is recommended for

research on causal analysis (Mbengue, Vandangeion-Derumez, 1999) and has been

used in one of the leading research contributions to the topic by Vielmeyer (2004). It is

a three step structure that has been introduced by Verrecchia (2001) and adapted to

the research topic of company's disclosure. This given structure is extended by one

additional step, because in a crisis situation additional factors influence the behaviour

of the capital market participants that are not recognized in the study of Verrecchia

(2001). These factors have not been coherently analyzed in previous disclosure

research because they assumed a constant economic environment instead of a crisis

28

situation. In this additional step the psychological and economic theories about the behaviour of capital market participants are analyzed. As a result of this extension the literature review is subdivided into four steps:

In a first step the theoretical explanations of communication between companies and the capital market that build the relationship between them are described. Followed by

a review of empirical studies on this topic a distinction is made between required and

voluntary risk disclosure (Core, 2001, p. 443). In addition literature on corporate disclosure measurement and attempts to specify it to situational risk instances are documented in chapter 3.1.

In a second step the incentives for the company to disclose information are analyzed. Why do companies invest in the cost of better disclosure and have managers and their

companies a common motivation to disclose information? Theoretical and empirical

studies on these aspects are analyzed in chapter 3.2.

Why do financial market participants demand disclosure is the third step and how does

this demand alter in times is the forth - additional - step both described in chapter 3.3.

The behaviour of financial market participants can be explained with economic models that assume a rational investor. This assumption is critical during a market crisis

situation. A more adequate way to explain the behaviour of market participants is to

take psychological aspects of behaviour into account. In substitution or combination

with rationality it can be assumed that investors are influenced by two opposing

emotional factors in their financial decisions under such circumstances. With trust

through timely and extended risk disclosure companies can build up a reputation that

might lead the investor to hold on to its shares during distress in the market (Vater et

al, 2008, p. 2606). The opposite psychological factor of trust is fear that might induce

investors to be influenced by the behaviour of other investors, who sell their shares. This effect caused by opinion leaders like analysts might lead to a mass flight out of

shares, a reaction undesired by the disclosing company. This crisis specific adaption of the general approach by Verrecchia (2001) is specific to this thesis.

As a result of the literature review the research design should be predetermined with the unit of analysis, all relevant variables that affect investment / divestment behaviour

29

of capital market participants and that affect the risk disclosure of listed insurance

companies and the measuring procedures of the variables. As far as the information

being relevant, measurable, publicly available and adjusted to the recent crisis

situation they are included in this thesis and described in chapter 5 "Research

Methodology".

3.1. Communication between companies and the capital

market This chapter analyzes the communication between companies and the capital market. Theoretical approaches in chapter 3.1.1. and empirical research studies in 3.1.2. are structured in order to analyse how they can be used to specify the research design.

In order to analyse to what extent companies can exceed the expectations of the

capital market with their disclosure it must be known what level of disclosure is

required by regulations. After analysing the legal accounting requirements on German

national and international level it is examined how insurance companies can fulfil or exceed these requirements in chapter 3.1.3. These steps are prerequisites to

determine the group of analysis which can be defined as insurance companies that are

similar enough to be compared in this thesis.

Subchapter 3.1.4. examines how risk disclosure that occurs in different forms like

earning announcements, or forecasts and in different parts of the financial statements

can be confined to this crisis situation.

3.1.1. Theories on communication According to Healy and Palepu (2001) a relationship exists between investors and disclosing companies. How this relationship functions and what cause and effect

models can be drawn from it is in the centre of disclosure theory. To analyse causal

relationships researchers focus on the addressee of disclosure information and on the

disclosing companies. They disregard standard setters who stipulate form and content

of the communication (Vielmeyer, 2004). This is considered as the influence of

standard setters in changing accounting standards in chapter 3.1.2.

30

Disclosure theory intends to explain and predict accounting practice. The structure to

solve this task is universal, the theoretical underpinning differs and the conclusion

about the application of these theories in research is surprising. Theories try to explain the overall relationship (Watts and Zimmermann, 1978 and 1986) (Blackwell and Girshik, 1954) whereas empirical studies focus on certain

decisions investors or companies can make within that relationship. Nevertheless, the

predictive power of theories is restricted due to the variety of assumptions and

restrictions connected to them. The general idea is that additional disclosure is

appreciated by investors based on the assumption that every additional information

can be analyzed and integrated in the decision making process of the investor

(Blackwell and Girshik, 1954). Lenz and Diehm (2010) specify this theory to crisis

situation. They state that crisis relevant information disclosed by companies helps the

investor to predict the effect such crisis has on the fair value of the company he is

invested in. This theory is summarized as the prognostic value assumption. Another

significant assumption of all theoretical models is that every communication between

investor and company is based on contractual forms of relationship and therefore

regulated (Watts and Zimmermann, 1978,1986). But this does not hold true to crisis

situations with information demands of investors changing on a daily basis and

companies using other, often unregulated forms to communicate to them.

Thus, the conclusion by Vielmeyer (2004) can be supported that the question whether

or not additional disclosure is beneficial for investors cannot be conclusively answered

and should be tested empirically. Such empirical research might provide useful insights

and allow crisis related adaptations of the research methodology.

3.1.2. Empirical research on communication Empirical studies on disclosure research vary in their focus. They explain the

regulatory situation as well as predict and empirically verify behaviour of investors or

companies vis-ä-vis various aspects of accounting and risk. Therefore, the focus of this

research study is on empirical research that covers capital market reactions to

regulatory changes as well as on empirical research between the communication of

companies and capital market reaction towards it.

31

Various empirical studies analyse capital market reaction to changes in accounting

rules. Studies by Gonedes and Dopuch (1974), Leftwich (1981), Barth (1994) and Barth et al. (1992) tried to forecast share market performance of companies which are

affected by regulations that require fair value measurement of assets to different

degrees. Their results can be summarized that no share price reaction takes place

unless the adoption of new accounting regulations cause a change in real cash flow.

Other studies by Barth (1994) and Cornett et al. (1996) contradict these findings when

researching the share market effect of US-Banks that were required to disclose

additional fair value information in their financial statements. They found out that

upcoming regulations that require additional disclosure on fair value and risk

connected to financial instruments have statistically significant effects on the banks'

share performance.

Thus upcoming accounting requirements shall be eliminated as an irritating factor from

the research design. Therefore this research study shall only be performed during a

period of time when accounting regulations did not change in regard to financial

instruments such as during the period from January 2007 to March 2008.

Empirical studies on the communication between companies and the capital market

can be distinguished according to their research method in three categories (Armeloh, 1998), i. e. descriptive, comparative and marked related studies.

The first category of empirical research is descriptive. Studies that describe what kind

of information investors demand (AICPA, 1994, PwC, 2007 and EFRAG 2009) or what information companies provide after certain regulations have been changed are

performed by auditing companies (PwC 2006 and Ernst & Young 2009) or academics (Freiling and Wehrle, 2008 and Krüger et al. 2008). In respect to risk disclosure these

studies explore to what degree new international accounting regulations such as IFRS

4 and IFRS 7 are applied by listed companies.

Empirical studies that measure and compare disclosure are in the second category. Central to these studies is the necessity of an indexation that enables researchers to

measure and compare the disclosure performance of companies. Most of these

studies apply indices that are based on classes specified in accounting regulations.

32

s

These classes are weighted and integrated in an index. This index is applied to

measure disclosure and summarizes the results in a single number that is then used to

rank companies according to their disclosure performance.

The way the disclosure classes are weighted can be used to distinguish these

empirical studies of the second category. Empirical studies on risk disclosure

predominantly use indices that are based on subjectively weighted criteria (Bungart, 2003, Kraft and Nolte, 2005, Kirchhoff, 2007). Only two empirical disclosure studies by Armeloh (1998) and Vielmeyer (2004) are based on objectively weighted indices. In

order to achieve it these two researchers collected the preference of users of financial

statements about the decision usefulness of disclosure classes via surveys. This

required quantitative research using questionnaires that were sent to analysts, auditors (Armeloh, 1998) and additionally to preparers of financial statements (Vielmeyer, 2004)

in the process of their individual research projects. A sub-group of these empirical

studies uses disclosure indices with objectively weighted classes provided by

institutions. An index by AIMR frequently used in the literature until it has been

discontinued was a ready available disclosure index with a country focus on USA. The

advantage of this disclosure index has been that it was easy to use for statistical

correlations with variables the represent capital market reaction such as cost of capital (Sengupta, 1998) or credit default swaps (Healy et al, 1999).

The third category of empirical research studies extends the measurement of disclosure quality and analyzes its effect on the financial market. Whereas empirical studies of the second category are predominantly work of German researchers caused by the lack of ready to use indices by institutions such as AIMR in the USA empirical

studies on the market effects of disclosure are performed by American researchers in

particular. Studies of the third category use the widely published but discontinued index

by AIMR which is objectively weightend (Welker, 1995, Healy et al, 1999, Sengupta,

1998) and in more recent empirical studies subjectively weightend indices (Botosan,

1997,2002), (Bloomfield and Wills 2000).

After the categorization of empirical research studies two aspects turn out to be followed on in order to compare risk disclosure performance in this thesis. The disclosure regulations that are crisis relevant must be described in the first category.

33

T

This is done in chapter 4.3.2. to create risk disclosure criteria. The other prerequisite for further empirical research is to measure and compare risk disclosure performance.

To apply it is a major difficulty and a limitation to research studies in this research field

according to Healy and Palepu (2001). Two methods have been used in empirical

research studies to measure financial transparency: External indices and self

administered indices.

External indices have been created by financial analysts or rating agencies. US-based

research by Sengupta (1998), Lang and Lundholm (1996) and Zarowin and Gelb

(2002) used the published disclosure quality index of the Financial Analysts Federation

(FAF), a branch of the Association for Investment Management and Research (AIMR)

later renamed CFA-Institute. However AIMR discontinued its disclosure ranking in

1997, after ranking fiscal year 1995 (Core, 2001). No replacement of this US-oriented

external disclosure index is in sight. Even a recent study by Shaw (2003) relies on

these historic data, using FAF disclosure quality scores of the years 1985 - 1989. The

Standard & Poor's Transparency and Disclosure study United States (S&P, 2002)

used to analyze companies from Asia-Pacific and Europe by Khanna et al. (2004) was

not continued after being published for the first time. The index of the Center for

International Financial Analysis Research (CIFAR) has been used by Hope (2003) but

has been discontinued as well. Faced with the lack of external indices Hussainey et at.

(2003) suggest a self constructed index approach for future researchers in lack of

AIMR data for further studies.

Self administered indices are developed by the researcher and requires to weight and

integrate the disclosure criteria. In order to operationalize these type of indices they are

processed into a check-list that is applied to analyse and compare published financial

statements. Yet to apply these auditor type check-lists to individual financial

statements is labour-intensive so that annual transparencies studies have been either

discontinued (Daske and Gebhardt, 2006), or applied only to small samples (Core,

2001).

Self administered indices of overall disclosure performance have been used by

Botosan (1997), Armeloh (1998) and Miller (2002). Armeloh (1998) presents an

extended and widely followed approach to create such a self administered overall

34

disclosure performance index for German companies. Based on the opinion of auditors

and analysts he developed a disclosure check-list which he used to evaluate the

qualitative information disclosed in management reports of German companies included in the national share index, Deutscher Aktien Index (DAX) during the 1995

fiscal year. But the content of that past index cannot be used today because

accounting requirements have significantly changed from German national to

international accounting standards for listed companies in Germany within a decade.

His research work is the basis of the only external disclosure index published over last

two decades and still continued in Germany. The German periodical "manager-

magazin" has sponsored the work of 40 researchers who annually evaluate qualitative data in the financial statements of DAX-and MDAX listed companies using a check-list

approach since 1994 (D6hle, Manager Magazin 2004,2005,2006. ). The check-lists that are used by these researchers are published ahead of their annual examination

process. The check-list is adapted to several industries every year, including the

insurance industry. An alternative academic approach to administer a disclosure index

by the German periodical "Focus-Money" was discontinued in 2004 (Daske and Gebhardt, 2006).

Self administered indices that focus on risk disclosure have been applied by Kajüter

(2001), Kajüter and Winkler (2003), Bungartz (2003), Vielmeyer (2004), Kajüter and Winkler (2004), Gleißner et at. (2005), Berger and Gleißner (2007) in studies that focus

on risk disclosure in Germany and Botosan (1997), Botosan and Plumlee (2002) as

well as Khana et at. (2004) in international studies. Although such a variety of studies

exists only recent studies by Kraft and Nolte (2005), Freiling and Wehrle (2008) as well

as Filipiuk (2008) include or focus on the risk disclosure of insurance companies.

Due to the lack of self-evident criteria to categorize risk disclosure and no dominant

qualitative studies to weighten risk disclosure categories researchers suggest a check list that leads to an index. Luo et al. (2006) suggest to include criteria by auditors and

review relevant literature to include those in the check list in addition to the legal

requirements. Clustering these criteria in risk disclosure classes will result in a self

constructed index. Hussainey et al. (2003) suggest to use key words from analysts

reports in order to find categories of risk disclosure quality in self administered indices.

35

Lacking research to weighten risk disclosure criteria subjective ways are needed to fill

this gap.

This, nevertheless, would induce bias. To reduce bias Cheng and Courtenay (2006)

recommend statistical calculations to test self constructed indices for internal

consistency. This could be performed by comparing the results of self administered indices with external indices or alternatively by the verification of subjective

assumptions by external experts. The most promising attempt to create risk disclosure

criteria is suggested by Kötzle and Grünung (2009) who consider risk disclosure as a

response to external regulatory pressure and market incentives. They suggest to

discuss risk disclosure performance on the basis of regulations.

Empirical research on how companies actually comply with legal regulations are

subdivided according to two time periods. Before 2006 only listed German companies

were required by German national regulations to disclose risk related data. After 2006

the introduction of an international accounting standard (IFRS 7) and its uniform

application since 2007 made the requirements uniformly applicable to European listed

companies and therefore enlarged their risk disclosure.

Research on disclosure in financial statements before 2006 became standardized with the research by Armeloh (1998). Although this research study was neither insurance

specific nor focused on risk disclosure it became the standard on how disclosure

quality should be measured in the future. However this standard could not be applied directly to risk disclosure. The first risk disclosure related empirical study by Bungartz

(2003) created one self administered index that included corporate governance information that Bungartz applied in his empirical work. His work is critizised by Dobler

(2004) for insufficiently explaining his subjectivly weighted criteria that were integrated

in the index.

Vielmeyer (2004) improved the methodological approach and created "RDScores" (risk disclosure scores) to rate individual risk disclosure performance. Vielmeyer defines risk disclosure criteria according to the requirements of the German accounting regulation GRS 5-20 that set the risk disclosure of German insurance companies in the time before 2006. He weighted these criteria on the opinions of analysts, auditors and

36

preparers in order to achieve objectively weightend criteria. Similarities to "Dscores" by

Botosan (1997) and Botosan and Plumlee (2002) as scoring modules that include a longitudinal research approach on risk disclosure based on a ten year historical

performance review, forecasted infos and management report information provided in

the annual reports of companies. But the resulting risk disclosure index was based only

on companies from the machine-manufacturing industry (Vielmeyer, 2003).

Vielmeyer's research influenced further work on risk disclosure in Germany by Kajüter

and Winkler (2003) and was applied to the German insurance industry by Kraft and Nolte (2005). The later empirical studies of this period show weaknesses of measuring

disclosure performance using rather simple criteria such as the number number of key-

words or pages of the risk report. Additionally these empirical studies were all

restricted to Germany.

Research on measuring and comparing risk disclosure after 2006 envisaged IFRS 7a

new international accounting standard, which requires specified risk information from

listed groups in their consolidated financial statements. Instead of analysing the

disclosure in a country specific way the new regulation allowed research on companies from different countries that all require IFRS accounting.

These studies could not find a common opinion about the improvement this standard

provided to risk disclosure. Krüger et al. (2008) state that risk disclosure is even less

informative in regard to forecast information with the application of IFRS 7. However

this research result was based on financial statements of smaller listed companies. It is

most probable that these companies have not applied IFRS 7 coherently at the time

the study took place. The reason for this incoherence could be that IFRS 7 has been

new at the time of the study. First time adoption of the new accounting standards could be worsen by lack of appropriate information on the side of disclosing companies.

Krüger et al. 2008 findings are supported by research from Ernst & Young (2009) on

companies which comply with IFRS 7 in terms of risk disclosure that reveals that the

disclosure requirements are only partly met in 2006. On the other hand research by

Freiling and Wehrle (2008) about the ten biggest insurance companies worldwide, that

apply IFRS in their 2007 financial statements comes to the conclusion that due to the

introduction of IFRS 7 and the insurance specific IFRS 4 risk information is more useful to financial decision makers and can be employed by them in their financial decisions.

37

Summarizing it can be stated that risk disclosure improved due to IFRS 7 in 2007 and from there on. As European listed companies are required to apply IFRS the group of

analysis researched in this study can include listed European insurance groups in the

period 2007 and after that apply IFRS 7 uniformly in their annual group financial

statements. A self administered index can be based on the categories of IFRS 7. But

to weighten these categories can only partially be based on the survey supported

check-list by manager magazin.

As a result of this chapter it can be stated that theory cannot fully explain the

communication between companies and the capital market. Empirical studies explain

only single aspects of the communication. Important conclusions can be drawn from

these studies in regard to the measuring procedure of risk disclosure. Risk disclosure

must me measured and compared using IFRS 7 accounting requirements in order to

create a self-administered index, due to the lack of a risk disclosure index provided by institutions. Existing insurance specific check lists that are based on German national

accounting requirements should be used in addition to IFRS 7 requirements.

3.1.3. Forms and legal requirements of risk disclosure

The purpose of this chapter is to find and operationalise categories of risk disclosure

that are relevant to the researched financial crisis. In order to achieve this, risk

disclosure is categorized and compared in the form of German national and international accounting requirements, respectively GRS 5-20 and IFRS 7. The goal of

this chapter is to find risk disclosure requirements that are useful to investors based on

their preference analyzed in chapter 2.2. They can be used to build a risk disclosure

performance checklist that is the basis of a self administered risk disclosure index as

an integral part of the research methodology.

Disclosure of information by companies can take place in formalized or unstructured

ways. Formalized ways include annual / quarterly financial statements and ad hoc

statements that are disclosed to the public and are available via internet. These

statements related to German national regulations such as HGB, RechVersV and WpHG, are supplemented by German national regulations such as GRS 5-20 on risk

38

reporting for insurance companies and GRS 15 on management reports and extended by international financial reporting standards specified IFRS 4 and IFRS 7.

Unstructured ways of communication are calls to key shareholders, road shows by the

management, investor days, statements during the annual shareholders' meeting and talks to rating agencies. These forms of communication are not regulated or like the

shareholders' meeting only regulated concerning the sequence.

Risk related information can also be disclosed in formalized or unstructured ways. Management is required to describe prospective risk situations such as transactions

that might have an unexpectedly high negative result and thus might threaten the

existence of the company to their shareholders at the annual meeting (Kiethe, 2003).

But such kind of unstructured risk related information about conversations to key

shareholders are not disclosed. Information from the annual shareholders' meeting are

open to the public but not documented. Ad hoc statements are disclosed and documented. Conference calls investors and experts take part in and can ask

questions to the management have not been uniformly documented in the past. They

tend to be documented temporarily on the internet platforms of the companies. Nevertheless, content and form are individual to each insurance group.

The content of the balance sheet, profit and loss accounts, cash flow statement and

equity statement as the major parts of the financial statements that disclose risk is

formalized. But risk is preponderant future oriented and consequently not reported in

such historic, back looking statements. According to the results of chapter 2.3. not all information disclosed is relevant to investors in a crisis situation. Especially,

information on investment risk in its three forms of market risk, credit risk and liquidity

risk as well as measuring risk of crisis relevant mortgage portfolio based financial

instruments and statements about the forecasted performance of companies can be

considered relevant to investors. Therefore, the requirements to disclose these five

information elements are analyzed.

In the following international accounting requirements are stated on the previously defined areas of risk disclosure. The purpose of this comparison is to select relevant criteria that can be included in a check-list in order to measure the risk disclosure

39

performance of listed European insurance groups in their consolidated group financial

statements during the financial crisis. The legal basis of this comprison are international requirements [AS 36, IAS 39 and IFRS 4, IFRS 7. As far as they do not

sufficiently specify the areas of risk disclosure German national requirements RechVersV, HGB, GRS 5-20, GRS 15 are used in a supplementary way.

The comparison starts with the market risk. It occurs when the fair value of assets in

the balance sheets drops. As a consequence listed European insurance groups are

required to disclose the fair values in their financial statements. Depending on German

national and international accounting regulations the disclosure of the assets fair value

vary in their degree.

Whereas § 54 RechVersV requires the disclosure of the fair value for all investments of

insurance companies, the international accounting standard IAS 39.45 requires the fair

value of the categories "available for sale" and "designated at fair value" to be

published. Assets in the categories "held to maturity" and "loans and receivables" are

not disclosed with their fair values neither in the balance sheet nor in the notes. German national requirements to disclose the fair value of all investments in the notes

exceed the fair value disclosure of international ones.

Another important risk information for investors is the likelihood of a drop in fair values

of investments. This information is based on scenarios that the company predicts, the

consequences for the investments held by the company and are summarized in a

quantified risk information in the company's notes. German national and international

accounting requirements differ in how the quantified risk information must be disclosed.

IFRS 7.39D together with 7.40 requires a sensitivity analysis that tells investors how

the company is exposed to the market at the end of the financial year according to

Rockel et al. (2005). To do so international accounting standards require that the effect

of a drop in fair values on the profit and loss accounts as well as on equity must be

disclosed. Additionally the assumptions and method used by the management to

prepare such a sensitivity analysis must be presented to the investor in the notes. GRS

5-20.29f as the German requirements does not specify the method used to measure

and summarize the market risk.

40

Yet the method used must be a recognized risk-monitoring method such as sensitivity

analysis or stress test. Both types of regulations allow alternatives: If the risk cannot be

quantified by management GRS 5-20.31 requires the disclosure of the profit and loss

accounts and equity effect of a given scenario. The German regulation is a 20%

reduction of fair values on shares and non-fixed interest securities and a1% shift in the

interest curve on the market value of fixed interest securities and bonds. International

regulations on the other hand do not require the disclosure of a given scenario but

demand reason about why the management has been unable to apply a sensitivity

analysis and why the analysis does not make sense according to IFRS 7.42.

The market risk that investments are exposed to must be disclosed in a similar way according to national and international regulations. Both kinds of regulations require a

quantitative risk information that is based on scenarios. These scenarios quantify the

market risk of a company's investment. Nevertheless, different methods on scenarios

and the exemption clause in IFRS 7.42 make it harder to compare corporate risk disclosure on market risk according to Rippel (2006). Therefore international

accounting requirements are used as categories in the check-list. The exemption

clause should not be considered acceptable in that check-list.

The regulations on credit risk as part of investment risk are analyzed next. During a prolonged financial crisis bond issuers default on their payments. This credit risk significantly affects the financial performance of insurance companies that have to

record these defaults in their financial statements through depreciation of their investments.

German national and international accounting regulations share the duty to disclose

credit risk disclosure as of GRS 5-20.32 and IFRS 7.34a respectively. However the

German national requirements allow as an alternative when the management is not

able to provide a quantification to disclose external disclosure categories of fixed

interest securities and bonds. It cannot be concluded that IFRS 7.34a together with B7

do require a quantification of the credit risk, so that qualitative about external ratings of investments can be deemed as comparable in regard to the information usefulness of investors. International accounting regulations surpass German national ones when

41

they demand the maximum credit exposure from financial guarantees granted in insurance contracts in IFRS 7.36 according to Rockel et al. (2005).

Comparing both types of regulation it can be concluded that German national

requirements are less specific about the form. The information usefulness of the

information differs however, when a qualitative risk disclosure as an alternative to the

quantitative disclosure is chosen under German national regulations. This informative

difference is enhanced by the requirement of IFRS 7 to state the quantitative figure of

maximum credit exposure without risk reducing transactions being considered. Because of its focus on quantitative risk information international regulations are far

more strictly than the German national ones (Rippel, 2006).

Liquidity risk as another subgroup of investment risk affects especially insurance

groups with a life insurance segment that guarantee a certain interest to their

customers and cannot receive this level of return from their investments when a financial crisis decreases their value and returns. German national and international

requirements on liquidity risk are specific to the insurance industry.

According to GRS 5-20.26c information about the cash inflow of investments and the

matching of cash outflows from obligations based on insurance contracts must be

provided. Similar to this German national regulation these cash inflow and outflow information must be provided according to the insurance industry specific IFRS 4.38. The specification of IFRS 4 IG 61 for a sensitivity analysis of equity and profit and loss

accounts on the effect of a change in those factors that influence the liquidity of an insurance company most is only a recommendation.

The comparable level of liquidity risk disclosure has been amended in March 2009 with

an IFRS extension (IASB, 2009a). As a reaction to the financial crisis a maturity

analysis of all financial liabilities and assets as well as a description how this analysis has been calculated has been added by a modified IFRS 7.39. The transparency of

outflows by the settlement of insurance based obligations has been improved by the

required disclosure of the earliest demand point of time outstanding insurance based

guarantees. Guarantees as off-balance items have not been included in the liquidity

risk disclosure based on the previous IFRS 7.39. Additionally the IFRS extension

42

required that all significant deviation from earlier cash-outflow plans that have been

caused by the crisis must be disclosed according to the modified Appendix B of the

Application Guidance to IFRS 7.

Both German national and international regulations require quantitative information

about how insurance companies tune future cash in- and cash outflows. The biggest

improvement of decision usefulness to investors are the amendments to international

regulations (IASB, 2009) that require additional information on how the crisis changed the liquidity risk. Yet, the amendments are requested for financial statements form

2009 onwards and are not be considered in this research which is based on the 2007

and 2008 consolidated group financial statements.

This financial crisis affects markets that measure investments but became

dysfunctional as an effect of the financial crisis. Regulations on the disclosure of this

measuring risk of investment products held by insurance companies vary. Transparency on how the value of these products is measured is significant to

investors in times of financial distress to make well informed financial decisions. The

requirements of German national and international regulations before the IFRS

extension have been similar requiring qualitative information about the measuring

methods for financial investments and certain significant financial obligations according to § 56 (6) RechVersV and § 268 (5) S. 3 HGB as well as IAS 1.108 together with IFRS 4 IG 11. The extension of IFRS in response to the financial crisis and the

possibility to reclassify financial assets to other categories required the disclosure of

profit and loss account as well as equity effects caused by this change. This

requirement for disclosing measuring risk is a positive indication that a listed European

insurance group utilized the reclassification opportunity provided by the IFRS

extension.

The way investments are measured must be described according to German national

and international regulations. This requirement is critical for investments that are not

priced by markets because the financial crisis led to significant variances on investments that had to be measured with mathematical methods based on data form

inactive markets. Therefore the information usefulness of international regulation in

regard to measuring risk is significantly higher if reasons for insecurity of measuring

43

models are disclosed. The disclosure of the effect of recategorization can be used as

an indication for insurance groups facing financial distress during the crisis.

A last element of this comparison is the future oriented information. This is relevant in a

crisis situation, because it relates to the characteristic of the long term perspective of insurance companies. This risk disclosure can have different forms in group financial

statements prepared under different accounting rules.

A separate risk report that includes risk related and forward looking statements is

embedded as an individual statement of financial statements in Germany. In the UK

and the US risk related data are disclosed as supplemental information to the balance

sheet and profit and loss accounts as part of the notes. The same holds true under international accounting standards. This difference in form can be attributed to the

standard setting institutions. The German management report (Lagebericht) the

operating and financial review (OFR) in the UK and the management report named MD&A in the USA are similar in their function but differ in detail (Hartmann, 2006).

Their regulation density is high and they are set up and enforced by national bodies.

The German management reports stick out in their management forecast

requirements. The management report became an instrument with special legal status. It extends the historic information, processed in the financial statements to include future events by delivering a forecast (Lange, 2001). The forecast requirement was

specified and extended by modifications of the German Commercial Code in 1998

(KonTraG) and 2004 (BiIReG).

These requirements based on German regulations do not have the legal status of law

but are set by a private standard setter acknowledged by the Finance minister, the

German Accounting Standard Board. The management report is a unique source of information specified by the regulations GRS 5-20 and GRS 15 (GASB, 2001). The

mandatory part of the management report requires companies to disclose risk

connected to their business activities within the next two years (Gräfer and Scheld

2005). The German regulations specify risk disclosure categories and adapt them to

banks and insurance companies. The required disclosure can be either based on

44

individual risk measuring methods created by the company or by more robust set up

as-if-predictions.

International regulations on the other hand came in late and do not require a separate report. The IASB released the IFRS 4 Insurance contracts in 2004 and IFRS 7 Financial instruments - Disclosure in 2006. The legal status of the risk disclosure

requirements is disputed. The disclosure specifications are not part of the legally

binding standard, but part of the implementation guide.

The project of the IASB to draw up a standard on management commentary as a

separate report has developed. An exposure draft as a step ahead of a final

accounting standard has been published (IASB, 2009c). The reason for being late in

comparison to German national requirements might be that the IASB has to cope with different national backgrounds on risk disclosure and lacks a common enforcement institution. The IASB followed another approach instead and requires companies to

disclose risk related information about their financial instruments as supplemental information in the notes instead of a cohesive report. Because they can become a dominant risk factor for a company, IASB issued IFRS 7 disclosure of financial

instruments in 2006 to be applied from 2007 onwards and the industry specific IFRS 4

Insurance contracts in 2004 to be applied from 2005 onwards. Later standard requires the disclosure of insurance specific risks. Together they build the regulative framework

for listed insurance companies on risk disclosure.

Both standards and the resulting risk disclosure must be applied by listed German

insurance companies and are enforced by their auditors. European listed companies have been required by EU regulation No. 1606/2002 to account according to IFRS

since 2002.

It can be summarized that the form of risk disclosure required by German national and international regulations varies. In order to compare the risk disclosure performance of European companies a level playing field has to be established. German insurance

group had to comply with more demanding risk disclosure requirements especially in

the past. Other listed European insurance groups can close this disclosure gap by

45

voluntarily publishing relevant information in their group financial statements according to IFRS accounting rules.

Forecasts allow investors insights on opinion of the management about the effects the

crisis has on the listed European insurance group and tells them what plans the

managements came up with to pursue their plans. Contrasting the German national

and international regulations the first mentioned ones require the disclosure of

expected developments, negative trends, significant assumptions made by the

management. A very demanding requirement of national German requirement GRS

15.84 is that the management must forecast significant risk for the following two years. The risk shall be quantified in the management report according to GRS 15.89.

Penman (2007) regards this information as a single risk indicator with superior decision

usefulness for investors. However by a recommendation of the German regulatory body this requirement has been lowered in response of the financial crisis when the

uncertainties are too high. Additionally auditors (PwC, 2009) and regulators (GRS,

2009) diluted this requirement for the then ongoing financial crisis. This

recommendation did not let the companies off their duty to forecast in a qualitatively

way. This requirement has been critizised after the crisis of being unrealistic in times of

crisis by Schwab (2010). However this requirement has been upheld by a German

district court for public companies that are listed in the German share index (Schwab,

2010). In contrast to the German national requirements to quantify a forecast for a

specified period of time in the future no such one exists according to international

accounting standards.

German national and international requirements require similar levels of disclosure

about current risk, both requirements allow a qualitative description of risk. When GRS

5.26 recommends to describe interdependencies between individual risk, IFRS 7.34c

requires the management to describe the concentration and extent of risk that arises from financial instruments with similar structures. In regard to current risk international

requirements are more specific by focussing on financial instruments whereas German

national requirements stress the disclosure of interdependencies without a clear focus

on specific assets. In regard to the crisis situation IFRS 7 seems to be a requirement

46

that provides investors with more specific risk disclosure although it is only qualitative

and thus not preferred by them according to AICPA (1994) and EFRAG (2009).

Another similarity between the requirements can be found regarding risk that threatens

the existence of the company. Such risk shall be disclosed according to German

requirements as of GRS 5.24. [AS 1.23 requires the disclosure of liquidity risk that may

threaten the existence of the company, as far as they effect the going concern

assumption that is vital to the measurement of assets and liabilities in the balance

sheet. The German national requirement is specified in the time frame when it states

that the forecast period of existence threatening risk is generally one year. The severe

risk that may threaten the existence of companies is regulated in a similar way in

German national and international accounting standards.

The form of disclosure is only regulated by German national accounting requirements. GRS 5.30 together with GRS 15.91 advise that information about risk shall be reported in a different section of the financial statements than the forecast statements of the

management. This regulation tries to separate risk and forecast information in order to

prohibit a mix of information that would reduce the usefulness for investors. No such

requirements exist in international regulations as all qualitative information is disclosed

in the notes.

The risk reporting profile varies according to German national and international

regulations in regard to the place of disclosure within the financial statements, detail

and forecast period. According to German national requirements risk information is

centered in the notes and in the management report as a special risk section of the

financial statements. In contrast international regulations demand an unstructured disclosure of risk related information in the notes due to the lack of a special

management commentary that includes all risk information. This makes collecting data

from financial statements of listed European insurance groups more difficult.

Regulations differ in detail and commitment in so far as risk disclosure is

recommended or mandatory. The advantage that international regulations are adapted to the crisis situation is an advantage of the case based approach over the principal

47

based approach of German national regulations. International regulations can respond to crisis situation faster.

Amendments to the IFRS 7 within 2009 (IASB, 2009a) prove the flexibility of the case based approach of international regulations. The principal based approach of German

national regulations is more rigid when risk information must be quantified and has the

unique requirement to make a forecast statement over a period of maximum two

financial years. According to Kaiser (2005) this attests the ability of the management to

identify changes and risk that are considered significant by investors, because this

ability guarantees the survival and future prosperity of the company.

This requirement goes far beyond the common requirement under international

regulations to disclose risk that threatens the immediate future of the insurance

companies. Quantitative information about future risk is not required in this financial

crisis according to PricewaterhouseCoopers, an auditing company (PWC, 2009). It is a

situational and crisis related risk disclosure that provides investors with the

management's assessment of the future impact of the situation. The decision

usefulness of this key risk disclosure to investors is not provided by international

regulations.

Analysis of German national and international risk disclosure requirements showed that both types of regulations cover the three investment risk categories, the measuring

risk and the forecast requirements that are assumed to be relevant for investors in a financial crisis situation.

The required detail and perspective of risk disclosure varies among German and international requirements. Most significant is the forecast requirement of financial

results and risk situation according to German regulations that is missing in

international regulations. This prediction must have a perspective of one year ahead. It

adapts risk disclosure to the individual situation of the company and must consider the

current risk environment created by the financial crisis as well as strategies of the

management to cope with the crisis. The lack of such a forecast according to international requirements makes is hard to compare the group financial statements of

48

Non-German insurance companies that do not have to comply with these national

regulations.

It might be an advantage of all listed German insurance companies that have to

comply in their group financial statements with German national and international

accounting rules. Differences exist regarding the detail of required disclosure. Some

international requirements are more detailed and require a quantified risk information

instead of a qualitative one.

Can different risk disclosure requirements result in a common level upon which the risk disclosure of European insurance companies can be compared? Because German

national and international requirements differ in detail it is impossible to define one set

of rules as the required level of risk disclosure. Therefore it is beneficial to assume that

these insurance groups compete for a certain level of disclosure by providing more information than necessary to meet the accounting rules.

Core (2001) introduced this assumption to disclosure research by stating that voluntary disclosure can be defined as disclosure that exceeds regulatory requirements. Therefore the highest level of both German national and international regulations can be set as the required level and all risk information that exceed that level can be

defined as voluntary disclosure. This stricter definition recognises exceeding disclosures practices. This theoretical approach might face some difficulties in

measuring the level of voluntary risk disclosure, because the disclosure level among the analyzed insurance groups might be higher than relevant accounting requirements.

As a result information that does not meet the peer group level must be negatively

recognized when risk disclosure is compared among European companies. German

national and international regulations offer alternative accounting treatment if an information cannot be disclosed. The alternatives can be less specific qualitative disclosure (GRS 5-20 par. 32) or the statements of reasons why the required disclosure cannot be made. Because of these alternative treatments it is possible to

recognize information that researched companies must not or cannot disclose. Thus, a

statement that a company must not or cannot disclose that information can be

distinguished from the behaviour a company that just omits this information.

49

As a result of this literature review it can be stated that German and international risk

disclosure requirements have levelled in and present a common basis of requirements

since 2007. They differ slightly in regard to the analyzed categories of risk that are

crisis relevant and differ significantly in their forecast requirements. However the

statement of Dobler (2005) can be supported that risk disclosure remains subjective,

although comparable regulation exist. But risk disclosure must be made comparable in

order to be analyzed with a self administered disclosure index. To perform this the

preferences of investors must be recognized. To make risk disclosure measurable and

comparable according to Cheng and Courtney (2006, p. 267) a "checklist-approach"

shall be applied. This is achieved by defining the highest level of German national or

international disclosure level as the reference point. Additionally the preferences of investors shall be recognized to weighten the criteria in order to measure risk disclosure.

In previous studies researchers construed disclosure indices assuming that the

amount of risk disclosure is a proxy for disclosure quality (Botosan, 1997, Lang and Lundholm, 2000, Hussainey et al, 2003, Evelt et al., 2009). Improved methods are

used by Schleicher et al. (2007) who measure disclosure quality as the number of forward looking profit statements in annual report narratives. This method, as well, is

based on the assumption that quantity equals quality approach and should be

extended.

Investors preferences based on the Jenkins report (AICPA, 1984) and a more recent

report (EFRAG, 2009) regard quantified forward looking risk disclosure as of the

highest informative quality, specifications of the risk factor is of medium informative

quality and mentioning risky financial instruments as least informative. Both studies

prove these preferences to be stable over time.

Another method to measure the quality of disclosed information is analyzed by Beattie

et al. (2004) who suggest the time orientation, financial orientation and quantitative

orientation of information disclosed must be recognized in measuring a disclosure

profile. This approach extends methods that assume disclosure quality is solemnly

connected with the quantity of key words disclosed. However this superior method to

50

measure disclosure quality cannot be applied to crisis situations, because risk disclosure within annual group financial statements is highly regulated and cannot be fully varied. As an example, the time orientation of the researched risk disclosure

cannot be chosen freely because financial statements are legally required to be

published at certain points of times after the end of the financial year. This restricts the

possibility to measure the timing of information over a short crisis related period.

It can be summarized that the highest level of German national and international risk disclosure requirements can be used as categories within the check-list that creates

the self administered index. The degree of fulfilment of these categories is measured

according to information preferences of investors.

3.1.4. Implications for the research methodology The literature review on the interaction between companies and the capital market has

implications on the researched period of time, the unit of analysis and the measuring of

risk disclosure performance.

Theories do not provide a simple cause and effect model that would explain what

extent and quality of risk disclosure investors' demand. As a result an optimum of risk disclosure cannot be defined for the recent credit crisis. Theory indicates that during

crisis situations other factors that influence investors beside risk disclosure become

less significant. So crisis situations are appropriate to analyse the effect risk disclosure

can have on investors in the capital market. But the restriction has to be recognized

that accounting standards should be stable and uniformly applied by insurance

companies in order to result in a comparable risk disclosure performance. This limits

the research period of time to a period between 2007 when IFRS 7 was uniformly

applied in Europe (Ernst & Young, 2009) and 2008 when the moratorium that

postponed significant changes in international accounting regulations by the IASB

phased out (Hayn and Prasse, 2009).

Empirical studies by Armeloh (1998) Bungart (2003), Vielmeyer (2004) Kraft and Nolte

(2005) as well as Kirchhoff (2007) apply self administered disclosure indices that are based on classes specified in accounting regulations. This procedure requires constant

accounting standards that are uniformly applied by the companies that are analyzed. 51

The insurance companies must apply IFRS 4 and IFRS 7 in order to approximately

meet the risk disclosure requirements of German insurance companies in their

management reports by additional risk disclosure in their notes. Differences between

risk disclosure requirements can be levelled out by non-German insurance companies by providing voluntary risk disclosure. Therefore insurance companies from Germany

and other European countries that are required to apply IFRS can be compared as long as they have a similar risk portfolio that derives of their business structure and

geographical business activities.

Studies by Chen (2006) and Coval et al. (2009) have significant implications on how

risk disclosure performance is being measured by focussing on key words to find

relevant information in the narratives of consolidated financial statements. Such key

words add additional power to a self administered risk disclosure index when it is

adjusted to the individual crisis situation.

This index should be operationalized by an audit-like check list. In order to achieve this

the five classes of risk disclosure detailed by German national or international

accounting standards that cover risk disclosure are taken as a founding structure and the extent and quality of risk disclosure by each listed European insurance group

should be measured based on the risk information disclosed in the group financial

statements. The restriction on group financial statements recognises that IFRS 4 and IFRS 7 are only mandatory in these financial reports and not in quarterly statements.

In tune with constant investor preferences quantitative risk disclosure including

forecast information is being regarded as highly informative, specified qualitative disclosure including negative statements about the lack of a certain risk is regarded as

medium informative and unspecified qualitative disclosure is regarded as least

informative. The classes must be weighted and then summarized to be an index. The

annually published and insurance related index used in the German periodical "manager-magazin" that derived from research by Armeloh (1998) shall be applied on the relevant disclosure risk classes because investors information preference are fairly

constant over time (AICPA, 1994 and EFRAG, 2009).

52

3.2. Motivation of companies to disclose risk The motivation of listed companies to disclose risk in their financial statements is

analyzed on three different levels. The structure of the analysis is the same on every

level and starts with theories or empirical studies and ends with the research results

being applied to the recent crisis situation. The motivation is analyzed in order to

identify factors or derive methods that should be recognized in the methodology of this

research project.

First, economic theories and studies are assessed to explain using the principal agent

theory why companies are motivated to disclose risk in chapter 3.2.1. In this chapter it

is assumed that on this high conceptional level no organizational rules or restrictions

preexist. On the second level in chapter 3.2.2. it is analyzed what factors motivate

managers to disclose or withhold information. This is done on a corporate level

assuming a set of rules and internal corporate limitations defined as corporate

governance being applied. On the third level in chapter 3.2.3. it is analyzed whether

other non organisational factors influence the disclosure behaviour of managers and

companies. Implications for this research methodology are summarized in chapter 3.2.4.

3.2.1. The principal agent theory The relationship between the company's management and its shareholders is

explained in economic research by the principal agent theorem. The relationship

between management and investors can be interpreted as a relationship between the

agent who is entrusted by the investor to manage his company. An information

asymmetry between agent and principal exists, because the agent knows more about

the market and corporate circumstances than the principal. By disclosing information

the information asymmetry between the management as the agent and shareholders

as the principals narrows (Core, 2001) (Lambert, 2001).

It has been recognized in the economic theory that contracts between principal and

agents can enforce the disclosure of information and thus reduce this asymmetry by

obliging managers to report to their principals. Jensen and Meckling (1976) introduced

principal agent and contracting theory to economic research as main concepts of

misalignment of interests between owners and managers of companies. They

53

recognized that this potential conflict, coupled with the inability of owners to write

costless perfect contracts and monitor the managers inherently reduces the value of firms as economic entities (Cheng and Courtenay, 2006, p. 262).

The selfishness of managers by using private information gathered in their function as

managers and withholding them from the public has been researched in regard to

management compensation. According to Healy and Palepu (2001) as well as Bushman et al. (2004) managers influence the timing of reported profits because their

compensation terms - that are fixed in contracts - depend on the point of time when

profitability figures are disclosed in financial statements. Such misuse falsifies the true

and fair view investors should get on companies through financial statements (IASB,

2001b) and negatively affects their financial decisions resulting in economic inefficiencies. Bloomfield and Wilks (2000) support this conclusion with their research

under the assumption of a laboratory financial market that is assumed in economic theory. They found out that greater disclosure quality leads to higher share prices

resulting in an increased market value and higher liquidity of the company.

This empirical research results coincide with the unravelling theorem by Wagenhofer

and Drücker (2007) that managers increase the fair value of their company by

disclosing sensitive information, because investors otherwise assume the worst. These

positive effects being coordinated with the costs of disclosure determine an optimal level of disclosure on this high level of economic theory (Baiman and Verrecchia,

1996).

The general finding of Bloomfield and Wilks (2000) states that positive economic

effects of disclosure are stronger when investors face a risk event. How can this

theoretical background be applied to the researched crisis situation?

A characteristic of this financial crisis was the insecurity of investors about the fair

value of certain financial instruments of the company they owned. Being complex in

their structure these instruments have no listed price or original market value. Their

value derives from calculations based on assumptions set by managers of the

companies caused by the lack of verified information from markets or other external

scources.

54

This information asymmetry between manager and investor is defined as measuring

risk in chapter 2.3. and is specific to this recent financial crisis. Kötzle and Grüning

(2009) in their research study support the conclusion that the information asymmetry

enlarged during this crisis. According to Cheng and Courtenay (2006) asymmetry

cannot be eliminated neither by costless perfect contracts nor by constant monitoring

of managers by the owners of companies. Therefore corporate rules or regulations on

the disclosure of the measurement assumptions on certain financial instruments must

address this crisis specific measuring risk. This can be achieved on the company level

by applying rules or a set of rules such as corporate governance. These rules shall

increase disclosure in order to reduce the information assymmetry.

3.2.2. Corporate governance affects risk disclosure on corporate level

How can the selfishness of managers be tamed? Companies try to limit the negative

effects of their managers' selfishness with corporate restrictions and try to utilise it with

contractual arrangements that motivate managers to disclose information. The limiting

aspect is stressed in the corporate governance definition by Menz and Nelles (2009)

stating that it is the sum of all rules and facts that influence the corporate decision

making process of the management. These restrictions can be internal or external

ones such as banks, auditors and courts (Cheng and Courtenay, 2006). The aspect of

utilizating managers' selfishness extends the scope of this definition. It is included in

the corporate governance definition by Denis and McConnell (2003) which includes

both internal and market based mechanism that induce the self interest of corporate

agents to make decisions that maximize the value of the company to its owners.

Empirical studies focus on the factors that limit selfishness of managers by the

disclosure of relevant information. In the following studies that researched the

relationship between disclosure and other variables are presented such as institutional

ownership of the companies, existence of audit committees, quality of company's

auditors and the independence of the board of directors.

The assumption that institutional ownership controls agency problems could not be

empirically supported by Eng and Mak (2003) who found out that the structure of

ownership and not the single factor of institutional ownership determines the level of

monitoring and thereby the level of disclosure desired. These research results are

55

specified by Chen and Cheng (2007) who found out that beside auditors' quality the degree of institutional shareholders is significant to the quality of disclosure.

Institutional shareholders need a certain degree of significance in the form of a share in

voting capital to prevent managers from increasing or decreasing reported profits towards a level desired by them (Chung et al., 2002). Non voting or an insignificant

share of institutional investors thus does not improve disclosure.

Few empirical studies focus on the market based aspects of corporate governance. They do not focus on single rules or contracts that utilize the selfish motivation of

managers. Instead they analyse correlations between corporate governance indices

and variables of the financial market. Lehn et al. (2007) find statistical support that

good grades on corporate governance indices increase the share value in the long run. The reliance on a corporate governance index of this study shows that individual

corporate solutions to utilize managers' motivation are hard to identify, to measure and to compare. Market based mechanisms that induce the self interest of managers are

specific to each company and difficult to compare.

How does corporate governance affect disclosure in the crisis situation? The terms of

managers' compensation packages are named as one reason of the recent crisis that

lead to overhauling risk due to short-sighted behaviour of managers (Anonymous,

2009b). Rewarding short term success these contracts encourage managers to make deals that improved the short term profit situation of the company and accordingly their

annual bonus without recognizing the long term risk burden these deals placed on the

companies. Only some financial companies changed their compensation contracts as

a result of the crisis to risk adjusted measures that determine the size of bonus

payments (Anonymous, 2009b).

Aside from being inflexible to changing requirements and individual to the company

can corporate governance improve the economic situations in times of crisis? Research on the cause and effect relationship between corporate governance and the

estimate of market participants on the insolvency probability of companies in the crisis

situation showed puzzling results.

56

Menz and Nelles (2009) found no correlation between credit default ratings and

corporate governance index ratings of German companies before and during the

financial crisis. So this empirical research does not support that high corporate

governance performance reduces insolvency risk. This study however excluded insurance companies and sets no results for this research project. It indicates that no

single cause and effect relationship in regard with corporate governance exists and that other factors that influence risk disclosure should be recognized in research

methodology as well. The next chapter analyses factors that are not part of economic theories or corporate governance.

3.2.3. Non organizational factors This chapter analyzes factors that influence corporate disclosure and underlying behaviour of managers in the absence of corporate restrictions. Fields et al. (2001)

recommend to structure research findings according to the sources of pressure that

affects disclosure. Following this approach the sources can be subdivided into external

and internal factors that upheld pressure on managers and on corporations to disclose

risk information. External factors can be supervisory power, group pressure, corporate

size or linkage to foreign markets. Internal factors that motivate to disclose are

predominantly cost related.

External pressure can be put on managers and companies by supervisory agencies

and by factors that influence the performance of the company which they are

responsible for. The pressure to perform increases with the benchmark group of

companies that the manager with his or her individual performance is compared to.

Such an increase in comparable companies can be caused by internal growth that

brings the company above a certain size so that it is recognized and compared to

others. Another way to increase the number of comparable companies is by expanding to other markets especially foreign markets and open the company with such a step to

the scrutiny of additional customers and investors. Supervision, group pressure and

pressure by other markets are analyzed as external sub-factors causing pressure on

managers to disclose risk information.

In regulated industries like the insurance industry pressure is put onto companies by

regulatory bodies (Scheytt et al., 2006). Facing external pressure insurance companies

57

improve their governance transparency defined by Bushman et al. (2004) and

Bushman and Piotroski (2006) as disclosure to German national or international

supervising agencies. Despite the possibility to use their external pressure supervisors

generally do not apply direct pressure but instead use the intrinsic motivation of

insurance companies to reach an adequate level of transparency. Smith and Tombs

(1995) researched that higher transparency lowers the burden of inspections for

individual insurance companies. This however can vary over time and depends on

local situations, because according to Ericson and Doyle (2004) insurance companies

have a constantly changing partnership with the state. Because misdemeanors against

regulators can have legal consequences managers have an incentive to meet

disclosure requirements set by regulators in their statutory accounting for regulating

national agencies. Their incentive to exceed them voluntarily however is limited.

Group pressure to disclose information is thoroughly researched. Based on the

economic theory of information asymmetry competition between companies tend to

reduce private information of managers. This information is available in several

competing companies so that single managers cannot selfishly use them in the market. Verrecchia (2001) summarizes that competition is an incentive for companies to

disclose.

Based on the influential study by Welker (1995) Botosan and Harris (2000) find out that

the size of companies is a significant factor associated with disclosure. According to

theoretical research by Bushman et al. (2004) and Vielmeyer (2004) transparency is

significantly higher when firms are larger. This theoretical finding is empirically

supported by research from Khanna et al. (2004). Smaller companies try to meet the

legal requirements while bigger companies try to excel them using discretionary

disclosure defined as voluntary disclosure in excess of existing regulations, according

to empirical research by Kötzle and Grüning (2009).

Based on this common understanding Kirschenheiter and Jorgensen (2003) found out

that aside group pressure as one factor another factor that influences disclosure is

whether the information disclosed are negatively or positively received by investors.

58

This influences the timing of disclosure and depends on the market structure the

company is embedded in. Dye and Sridhar (1995) researched that disclosure improvements are similarly timed in oligopoly markets by participating companies. These results are disputed by an empirical research study with contradicting results

performed by Pincus and Wasley (1994). Insurance companies using international

financial reporting standards and having a similar business portfolio can be seen as a

group similar to the group analyzed by Dye and Sridhar (1995) characteristics. Similar

disclosure behaviour of these companies has been identified with the application of IFRS 7 that was fully applied in the industry by 2007 although required by 2006 already

(see chapter 3.1.3. ).

These research results lead to the conclusion formerly discussed by Pae (2002) that

group pressure as a factor to improve disclosure limits the selfishness of managers to

alter the extend or timing of disclosed information. This leads to the requirement of a homogeneous group of insurance companies being studied in this research.

The interaction with other markets is another extrinsic pressure factor on companies' disclosure. Cross listings are a relevant form of interaction that influences corporate

accounting and has been thoroughly researched. Khanna et at. (2004) found that

companies that are subject to US-accounting standards especially increase their

disclosure quality. This supports previous research by Lang and Lundholm (1993) and Lang et al. (2003) and can be explained by the exceptional enforcement power of the

SEC as the regulatory watchdog of US-accounting standards which is not achieved

under non-country related accounting standards like IFRS according to Pellens (1997).

Apart from this institutionalized external pressure to extend disclosure Baily et al. (2006) argue that internationality itself is a factor that increases disclosure (see

Khanna et al. 2004) even without pressing institutions. This aspect is discussed as the

'legal bonding hypothesis'. Supporting this theory insurance companies analyzed in

this research study should comply to the same accounting standards in order to be

comparable in their risk disclosure.

Internal pressure to disclose information is put on companies by the economic incentives associated with it. According to Weick (1995) organizations have an intrinsic

59

motivation to inform the market about the way they handle risk by accounting and information systems. Extended disclosure reduces the cost of the company to acquire

additional equity and debt on the capital market. This economical explanation is called transaction cost economics by Williamson (1975) and has been extensively analyzed by Verrecchia (2001). His theoretical explanation has been thoroughly tested in

empirical research studies by Welker (1995), Botosan (1997), Botosan and Plumlee

(2002), Sengupta (1998) and Healy et al. (1999). While these studies that researched US companies and the US financial markets support Verrecchia's theory Kötzle and Grüning (2009) researching German companies state that there is no empirical proof

of an incentive to increase disclosure in order to reduce cost of capital. This statement however is restricted to conditions in Germany with its less developed capital market

culture. Disregarding research of Kötzle and Grüning (2009) all insurance companies

analyzed in this research should be listed companies that avail themselves of

additional disclosure by lower cost to a comparable level.

Beside all the incentives to disclose that are described above there are internal and external costs of disclosure that reduce companies' incentive to disclose risk information according to Verrecchia (2001) and Vielmeyer (2004). The costs of disclosure are not published in the financial statements of companies but internal costs

could be measured. They include cost of information collection (Beierle, 2004) and cost

of processing information and report relevant information in financial statements (Depoers, 2000). Both types of costs are paid for by the disclosing company. The cost level depends on how sophisticated the management reporting systems of the

disclosing listed European insurance group are and how diversified its organizational

structure is that makes it more difficult to gather the relevant information (Depoers,

2000). External costs on the other hand cannot be measured. They include the cost of

unintended use of sensitive information by competitors (Beierle, 2004). Such use can become financially harmful to the disclosing entity according to Depoers (2000). This

effect however cannot be measured. Other external costs are proprietary costs

according to Verrecchia (1986). He defines them as costs for withholding the

disclosure of bad news by companies. Verrecchia (1986) argues that traders expect a

general disclosure of bad news among groups of the same market. When one group does not disclose traders assume the worst for this group that withholds such information. As a result the share value of this listed group decreases. Cost and benefit

60

of disclosure must be recognized in order to find an optimal level of disclosure

(Bushman and Verrecchia, 1996). However Vielmeyer (2004) along with Wagenhofer

and Ewert (2003) disagree that this level can be determined. Thus the costs of

disclosure are neither fully measurable nor published in the financial statements and

cannot be considered as a variable in this research study.

Another limitation is the behaviour of managers. Theoretical research by Lambert

(2001) states that risk disclosure tends to decrease because it is seen by managers as

a negative information that unveils their past mistakes. As a result they try to disguise

their actions by providing biased information to the investors (Surrey, 2006). Lenz and Diehm (2010) challenge this when they state that risk information are hard to verify and

cannot be traced back to the managers.

A similar aspect of disclosure timing is empirically researched. According to Shaw

(2003) managers delay recognition of bad news in the balance sheet until they

prepared the market by enhanced disclosure in the notes. Managers therefore try to

delay the recognition of bad news but they cannot prevent it altogether.

3.2.4. Implication for the research methodology The analysis of the motivation of companies to disclose risk has lead to implications on the type of risk disclosure, the variables and the unit of analysis.

The principal agent theory supports the research topic. Managers know the

assumptions and procedures to measure complex financial instruments. This sensitive

risk information should be disclosed in a crisis situation in order to reduce the

information asymmetry between management and investors. Such risk disclosure

would create transparency that restricts the power of managers in setting the

assumptions and calculation method to value these financial instruments. Without

information about the measurement investors would assume the worst in a crisis

situation according to Wagenhofer et al. (2007). In the setting of a financial crisis the

worst for investors would be that these financial instruments have no positive fair

market value, cannot be sold or only by a fire sale that creates losses. These losses

would reduce the equity position of the company and therefore increase its risk of insolvency of the company.

61

In order to increase the pressure on managers to disclose this information institutional investors with a significant share holding in the company's equity are necessary. They

can be included as a control variable in the research methodology. The proportion institutional investors should have in the equity can be set in analogy of research by Shleifer and Vishny (1986).

A further non-organizational factor that restricts individual interests of managers is

group pressure. As an implication the group of analysis - the insurance companies that are analyzed in this research study - should be share exchange listed companies that are coherent in their business structure.

The idea of an optimal level of disclosure is a theoretical concept that cannot be

applied to this study. Lack of published disclosure cost would disrupt any calculation of cost and benefit that can result in the determination of an optimal disclosure level. It is therefore not considered in the research methodology.

Research by Bagnoli and Watts (2007) analyzes the setting of complex situations in

which disclosure is used to pursue a certain corporate strategy. Although not related to

the research study by topic conclusions about the research methodology can be drawn

from it. The authors suggest the use of event study methods during situations of

market extremity. So this method is used in the research methodology of this thesis to

quantify the reaction of the capital market towards disclosure.

3.3. Reaction of the capital market towards disclosure The Focus of this chapter is the behaviour of capital market participants. Different

types of capital market participants and their decision making process are analyzed. Based on the categorization of the participants their behaviour towards disclosure

information is explained by economic theories and empirical studies in subchapter 3.3.1. These theories assume rational investors as market participants. To supplement the literature review in regard to a crisis situation additional psychological factors like

trust and fear that affect the behaviour of capital market participants at that time are

explained in subchapter 3.3.2. All explanations of market participants behaviour are

summarized and applied to the research methodology in subchapter 3.3.3.

62

3.3.1. Participants in the capital market Literature identifies different types of capital market participants. Investors are

regarded as the key element of the capital market in disclosure research (Verrecchia,

2001). This is supported by the perspective of international accounting standards

which focus the disclosure of information on investors' needs aiming to provide them

with information that should be useful for their investment and divestment decisions

(IASB, 2001 a). Investors can either be private or institutional. How these different types

of investors regard disclosed information by companies is not adequately resolved in

the literature.

Researchers often assume similar preferences and decision making of private and institutional investors. They do not distinguish between the two investor types (Nagar

et al., 2003). However following theoretical problem can enumerate the distinction that

would be helpful in this research.

Shleifer and Vishny (1986) state that in a company owned by many small owners it

does not pay for any individual owner to monitor the performance of the management because each single owner lacks the power to change wrongdoings by the

management. By contrast institutional investors like family trusts, pension plans, banks, insurance companies or investment funds (Shleifer and Vishny, 1986) are

empowered to intervene in management decision if they possess a significant part of the shares.

Institutional investors can therefore be defined as belonging to a certain organisational

structure and holding significant power to influence management decisions. This

makes institutional investors more sensitive to corporate disclosure and there is an incentive for them to consider disclosed information in their financial decision making

process. Private investors lack this incentive. It is more plausible for them to make their

investment decisions based on recommendations provided by analysts. Along with different motivations institutional and private investors differ in their decision making

process model as well.

All capital market participants have a common aim. They try to evaluate the fair value of companies they want to invest in or divest from. If the share market values of a

63

company is lower than the value estimated by the participants they buy shares of the

company. If the share market value of a company is higher than its estimated value

they sell their shares in it (Vater et al., 2008). Investors are supposed to take their

decisions based on this economic explanations. This decision making model is based

on assumptions of fully available information, no transaction costs and fully rational decision makers. However certain market phenomenons do not make sense on the

basis of traditional economic models according to Hirshleifer and Teoh (2003). They

can only be explained by psychological theories of which the private investor are

exposed to by a higher degree than institutional investors. How psychological factors

affect private investors is the subject of subchapter 3.3.2.

In contrast to private investors institutional investors tend to standardize their decision

making process in order to evaluate the value of companies. They use uniform measurement-models for this task. These models include quantitative data from

financial statements in addition to current ones historical data are processed to predict the future value of companies. Thus, both types of data are included in the calculation

of the current value of the company which summarizes the value of a company based

on the future development of its business activities. These models - like discounted

cash flow analysis - are used for accounting purposes as well (IASB, 2005).

How good the results of this process are depending as well on the quality of qualitative information that the market participants get about the company. Vater et al. (2008)

argue that qualitative information disclosed in financial statements are leading

indicators for the market value of the company. This argument is empirically supported by research from Pechtl (2000). He analyzed the forecast disclosure of German

companies and could prove that these are an integrated part of the decision making

process of institutional investors. The growing importance of forecast disclosure is

supported by research of Sinah and Watts (2001) who state the growing relevance of future oriented information to investors.

Yet an advance decision making model is necessary to integrate qualitative information about the future when calculating the value of companies based on a

probability weighted cash-flow basis on scenario basis. It is assumed for further

64

research that private investors have a reduced capability to process such type of information that comes along with advanced standardized methods.

Financial analysts are intermediators in the capital market. They are important players in that market and are frequently assumed to represent and influence investors belief

and their activities. However the effects of extensive disclosure on analysts are

ambiguous according to Healy and Palepu (2001). They argue that extended disclosure enables analysts to make superior forecasts and recommendations. On the

other hand better public disclosure preempts analyst ability to distribute private information of managers to the investors. Healy and Palepu (2001) state that

enhanced disclosure leads to lower demand of analysts' services. This is plausible for

institutional investors. But considering a crisis situation private investors might rely

more on expert opinion during that time for psychological reasons such as fear.

In the following the behaviour of each category of capital market participants towards

extended risk disclosure is predicted. This is done based on empirical research

studies. Whether participants sell or buy shares of companies that disclose risk information is influenced by their decision making process and their role in the capital

market.

Research on private investors found out that personality related and situational factors

influence their decision making. Research showed that personal experience investors

have made during their early childhood affect their investment behaviour as adults (Anonymous, 2009a). This influence can hardly be overcome by additional disclosure

of companies. Research analyses a variety of other personality related factors that

influence their financial decision making. An empirical study on situative factors by

Porcelli and Delgardo (2009) found out that under stress private investors tend to

chose riskier investment alternatives. When facing the possibility of a minor loss or a

major one the research participants chose the major loss alternative when being

exposed to stress factors. Applied to crisis situations when investors are exposed to

such stress they hold on to shares and do not realize the loss that can be determined

right now. Instead they hope that the situation might improve in the near future.

65

Apart from personal or situative aspects research found out that investors are more likely to hold on to shares of companies that raise their confidence by a long track

record of extended disclosure (Diamond and Verrecchia, 1991) (Kim and Verrecchia,

1994). This research supports the assumption that it takes a long period of time to

build up confidence in investors in regard to a certain company. Arpan and Roskos-

Ewoldson (2005) modify the assumption to crisis situations. They state that by

disclosing bad news early and fully in a crisis situation a company can build up trust.

Summarizing research results on the behaviour of private investors it can be stated that their financial decision making is hard to predict due to personal traits that

question the existence of uniform investors. Even if uniform investors are assumed to

exist situative factors that change the level of stress must be recognized in order to

predict financial decision making of private investors. This aspect is analyzed in

subchapter 3.3.2. with psychological factors like fear and trust which might dominate

the decision making process of private investors in stress situations.

Research on institutional investors does not find situative factors to influence their

decision making behaviour. Like private investors they tend to hold to shares of

companies with enduring good disclosure performance according to Diamond and Verrecchia (1991) as well as Kim and Verrecchia (1994). It can be assumed that this

factor influences institutional investors more than private ones due to the lack of disturbing situative factors and their increased decision making process capability to

store and process historic disclosure information and forecasts.

Research by Bushee and Noe (2000) states that institutional investors have an

excessive focus on short term performance of companies they invest in. This trait may lead institutional investors to disregard risk disclosure by companies and cut their

losses instead of holding on to their shares in times of crisis situation. No empirical

research supports this theory.

Literature does not provide indications on how the decision making process of analysts is affected by extended corporate disclosure but only the other way around. Research

by Healy, Hutton, Palepu (1999) as well as Lang and Lundholm (1996) show that more information disclosure leads to a larger analysts following of that company. According

to Barth and Hutton (2004) this has a positive effect on corporate disclosure as those

66

companies integrate more rapidly corporate information on accruals than less followed

firms. It is safe to assume that the quality of recommendations improves the more

analysts deliver their view on the corporation. However no research has been

performed on how analysts react to enhanced disclosure of companies during times of financial distress. Although common sense suggests that they reward disclosure because it enables them to make better and information based recommendations on these corporations.

Summarizing the analyzed research studies no coherent response of private, institutional investors as well as analysts to enhanced corporate disclosure can be

assumed drawn from existing research. Predominantly, research is based on the

assumptions of uniform investors and rationality of their decision making process. Additional psychological factors that influence foremost the decision making process of private investors in times of distress are presented in the next chapter.

3.3.2. Psychological factors that influence the capital market Private investors do not predominantly use standardized measurement-models. Their

decision making process is more opaque and seems to be influenced to a higher

degree by the interaction with other individuals (Hirshleifer and Teoh, 2003, p. 27). This

might lead to a substantial dependence on analysts recommendation.

The assessment by Hirshleifer and Teoh (2003) discussed in literature under the name

of "herd behaviour" is based on Keynes (1935) who stated that the markets are moved by "animal spirits" and not by reason. Hirshleifer and Teoh define herding as any behaviour similarity or dissimilarity brought about by the interaction of individuals. It

includes sub-rational mechanisms that induce an individual to be influenced by the

behaviour of another individual to behave in the same way (Hishleifer and Teoh 2003).

Several empirical studies try to measure herd behaviour in investment decisions focussing on private investors (Jain and Gupta, 1987), (D'Arcy and Oh, 1997) (Foresi

et al., 1998). All studies found only weak evidence of herding in different investment

decisions. The disappointing results of empirical research are partly to blame on the

impossibility of researchers to eliminate other causal factors that might affect the

67

financial decision making process of private investors aside from interaction that

causes herding behaviour as explained in theory by Hirschleifer and Tech (2003).

A theoretical explanation by Vater et al. (2008) of how companies communicate with

capital markets can be used to describe the relationship that leads to herding. It states

that collective psychological and emotional decisions are affected by the reputation of the company. They define reputation as the sum of all images that stakeholders have

about a company (Vater et at., 2008). The effect of a good reputation is trust that

investors have in the abilities of the management. Vater et al. (2008) assume that trust

in the company leads to a loyal behaviour of the market participants towards the

company. In times of crisis instinct based behaviour prevails and investors become

more sceptial (Keynes, 1935). So trust can balance this scepticism and make market

participants stay invested in the company (Vater et al., 2008) because they assume

that those effects forecasted by companies in their disclosure are to remain or become

true.

This idea is supported by the "stealing thunder-theory" (Aspan and Roskos-Ewoldson,

2005) that increases trust by early and extended disclosure of bad news at the

beginning of a crisis. This idea has been empirically supported by Lundholm and Myers (2002) and Zarowin and Gelb (2002). Other theories why companies disclose

bad news are discussed by Skinner (1994) who does not pursue an approach that

focuses on crisis situations.

Nevertheless, scepticism can also prevail and overturn trust into fear. Acting according

to that feeling market participants sell shares of all companies regardless of their

previous disclosure performance. What is easily explained in theoretical models cannot be supported by empirical research. One point objecting empirical research is that

herding is very hard to measure because of contradicting effects (Hirshleifner and Teoh 2003). So research studies performed on a company-level by Jain and Gupta

(1987) leads to mixed results seeing no herding among the bigger market participants

and herding in the behaviour of the smaller market participants following the bigger

ones.

68

It can be summarized that the images of companies in the mind of capital market

participants can influence their investment decisions. Trust in a company has a

stabilizing effect and herd behaviour a destabilizing effect on the investment decision

of market participants. Whereas trust is a feeling that influences private investors to

proceed in their own financial decision making process herding leads private investors

to follow recommendations of analysts or act without individual consideration in

accordance with the behaviour of other private investors.

According to empirical research by Choe et al. (1997) herding takes place before a

crisis period but not within the crisis. This is recognized in the timing of the event study

within the research methodology. Whereas the decision making process of private investors cannot be predicted in regard to psychological factors research by Sias and Starks (1997) found out that institutional investors reduce price effects of herding.

When private investors tend to emotionalize, institutional investors build a rational

counterpart that can - after gaining a critical mass in number or reputation - lead to

tipping private investors in favour of rational decision making back again. This is

considered by Hirschleifer and Tech (2003) as a possibility of future research.

Literature revealed that the relationship between disclosing companies and investors

are influenced by multiple factors among them psychological ones which make it hard

to determine a causal relationship. Emotions as part of the decision making process of

capital market participants harrass attempts to predict investors behaviour.

3.3.3. Implication for the research methodology

In respect with the results of analysing the reaction of the capital market following

aspects on the variables and the timing of the event study are recognized in the

research methodology.

Research shows that investors are not homogenous and must be studied according to

categories of private and institutional investors. This implies that the ownership

structure of all companies analyzed in this research methodology must be known and

categorized. The categories of private and institutional investors can be defined using definitions of the research by Shleifer and Vishny (1986).

69

Another implication that derives from the literature review is that investor decision

making can depend on long term experience with the disclosure performance of the

company. However it is understood that psychological factors influence the financial

decision making process of private investors too but that they can hardly be singled out

as a factor in this research study. They can be used to explain unusual effects that do

not comply with shared economic sense. Herding cannot be eliminated as a factor

influencing private investors behaviour because it is considered high before a financial

crisis peaks.

The reaction of the capital market towards disclosure has been empirically tested

during the South East Asian financial crisis in 1997. Research by Baek et al. (2002)

analyzed the effect of disclosure quality on the share value of Korean companies during this regional financial crisis. They found out that Korean firms that 'had higher

disclosure quality (... ) suffered less' in their share value (Baek et al., 2002 p. 269). This

study is supported by earlier research with a similar structure by Mitton (2002). Baek et

al. (2005) used the "event study method" which measures the development of the

share prices at days around a significant event. Share prices are called by the

researchers "return" and consequently the deviation of share prices form an individualized industry share index around a significant event is called "abnormal

return". The accumulation of all abnormal returns of each international insurance group in the days around the event is called "cumulative abnormal return". The significance

and direction of the abnormal return shows that investors held on to the share of a

company or sold it around that event, resulting in a lower share price. The application

of the event study method is regarded by Bagnoli and Watts (2007) as the standard in

the research field of market extremity. Considering psychological aspects like trust and fear such tool should be applied when the crisis situation peaked in order to test the full

extension of these psychological factors.

The influencing effect of reduced herding according to research by Sias and Starks

(1997) by institutional investors on private investors is monitored in the research

methodology of this study. It might be helpful to assume that a high level of institutional

investors in an European insurance company has a calming effect on private investors

that reduce their herding behaviour and increase the benefit of disclosed risk information by lower cumulated abnormal returns. However it cannot be ruled out that

70

situative circumstances can superimpose trust of investors that has been build up by

fear with significantly different consequences in regard to investment or divestment

decisions. Stress imposed on investors can lead to these psychological effects that

cannot be sufficiently forecasted and affects predominantly private investors. Because

trust of investors is hard to measure it is necessary to analyse the risk disclosure

performance of companies over a period of time. In this research study it is done over two reporting periods because of accounting restrictions not longer (see subchapter 3.1.3. ).

Literature indicates that institutional investors as one owner group influence the extend

of risk disclosure by companies (Chung et al., 2002). Similar behaviour has been

empirically testified for Chinese listed companies by Chen and Cheng (2007). A

calming effect of institutional owners on private investors during crisis situations has

been indicated by Kim and Verrecchia (1994). This influencing factor modifies the

intensity of the relationship between independent and dependent variables and can be

defined as a moderator variable (Mbengue and Vandangeon-Dorumez, 1999).

Additionally literature indicated that existence and timing of manager compensation

schemes are influencing the extent of risk disclosure. Therefore the type of manager

compensation is recognized as a control-variable in this research study. After catching the phenomenon of risk disclosure and the reaction of capital market participants in a

model which is enriched with additional variables the validity of this concept is verified. In a first step using a quantitative deductive method several hypotheses that arise from

the concept are stated (see chapter 5.1. ) and afterwards tested.

71

Chapter 4

4. Hypotheses This chapter develops the hypotheses of the relationship between risk disclosing

listed European insurance groups and the capital market is gathered. In order to

answer the research question whether those listed European insurance groups with

higher risk disclosure performance suffered less in their share value during the recent

crisis or not the relationship is categorized in hypotheses in chapter 4.1., the relevant

variables identified in chapter 4.2. and measured in chapter 4.3. Chapter 4.4.

describes the research methods necessary for data processing and chapter 4.5. the

data and sample selection.

4.1. Development of Hypotheses The aim of this research study is to detect how risk disclosure by listed European

insurance groups influences the investment/divestment decisions of capital market

participants during the financial crisis. Based on theories and empirical studies

analyzed in chapter 3a concept has been built that creates a model that consists of

variables and assumes premises among them. This process is called

conceptualization by Charreire and Furieux (2001) and it requires hypothesis that

illustrate relationships within this concept. The theoretical concept is assessed in a real

life crisis situation and the researcher has to determine whether a hypothesis is

acceptable or not aside from limits that enables the researcher to make a decision

about the acceptability. Further prerequisites for testing hypothesis are identified

variables quantified by using a risk disclosure index, an event study and a modified

equity ratio as well as quantitative and qualitative methods are needed.

The six hypotheses that derive from the literature review in chapter 3 cover the

following areas of the above mentioned relationship. Hypothesis H1 questions a

relationship between the capital market and risk disclosing insurance groups. The next hypotheses cover the activities of the insurance groups and the capital market. Hypotheses H2 an H3 ask how listed European insurance groups adapt their risk disclosure performance in times of crisis or to their ownership structure. Hypotheses H4

and H5 question how the capital market evaluates the fair value of insurance groups in

72

a crisis situation based on trust that has been built up by disclosed risk information. The hypothesis H6 is for technical reasons in this research analysis and questions the

risk disclosure performance difference from 2007 to 2008. The last hypothesis H7

summarizes the relationship between all analyzed variables.

As stated in chapter 3.1.4. theory does not provide a simple cause and effect model that would explain to what extend the quality of risk disclosure of insurance group

effects the financial decisions of investors. Although no single effect concept between

accounting and capital market can be created in a functioning market situation Lutz

and Diehm (2010) conclude that during crisis situations disclosed information can be of higher relevance for the decision making process of investors. With the intention of

eliminating other confounding factors the hypothesis of a relationship between

disclosure and capital market is tested. The hypothesis is based on an empirical study

of Baek et al. (2002) that higher disclosure quality of companies in a crisis situation

stabilizes their share prices.

There is no relationship between risk disclosure quality of listed European insurance

groups and their fair value in a crisis situation. This H1 hypothesis is analyzed by the

testing for a correlation between the risk disclosure performance of listed European

insurance groups in their 2007 and 2008 consolidated financial statements and the

cumulated abnormal return that represents the capital market reaction via the loss or

built up of trust towards disclosure. The correlation must be statistically significant with a correlation coefficient within the confidence interval.

The motivation of listed European insurance groups to disclose their risk is supported by the principal-agent theory. However organizational factors like the selfishness of their managers can restrict the disclosure of risk information. Specified to crisis

situation Wagenhofer et al. (2007) state that listed European insurance groups are

advised to disclose as much risk information as possible because investors otherwise

assume the worst. This would bring those insurance groups with the highest exposure to mortgage portfolio based financial instruments that are in the center of the financial

crisis according to Scharpf and Schaber (2009) to disclose most risk information in

their consolidated financial statements 2007 and 2008. Extended and timely risk disclosure about these instruments would be awarded with high marks in the self

administered risk disclosure performance measurement of this thesis. The reduction in

73

modified equity ratio from 2007 to 2008 is a proxy that is used to measure the extent of

risk each listed European insurance group is exposed to via mortgage portfolio based

financial instruments.

It is assumed for this research that no situations occurred between 2007 and 2008 that

would have increased the provisions of listed European insurance groups. The variable

change of modified equity ratio is based on the assumption that no significant insured

event took place in the period 2007 to 2008 that would have hit a single listed

European insurance groups more than others and would have caused it to record for a

provision that would have reduced equity.

There is no relationship between risk disclosure quality of listed European insurance

groups and the degree they are affected by the financial crisis situation measured

using the change in their modified equity ratio. This H2 hypothesis is analyzed by

testing for a correlation between the risk disclosure performance of listed European

insurance groups in their 2007 and 2008 consolidated financial statements and the

negative effect that the recent financial crisis produced on their equity ratio. The

correlation must be statistically significant with a correlation coefficient within the

confidence interval.

Based on chapter 3.2.1. corporate governance of listed European insurance groups

can influence the disclosure quality. The empirical study by Chen and Cheng (2007)

found out that a certain degree of institutional shareholders is beneficial for the risk

disclosure performance of insurance groups.

There is no relationship between risk disclosure quality of listed European insurance

groups and their degree of institutional ownership. This H3 hypothesis is analyse

testing for a correlation between the risk disclosure performance of listed European

insurance groups in their 2007 and 2008 consolidated financial statements and the

minimum 10% level of institutional investors as defined by Chung et al. (2008) on the

voting shares of listed European insurance groups. The correlation must be statistically

significant with a correlation coefficient within the confidence interval.

The relationship between corporate risk disclosure and the capital market response

cannot be analyzed within the same period of time. The format to disclose risk

information is standardized and required in the consolidated financial statements of

74

listed European insurance groups. The financial crisis situation peaked during

September 2008 at a point of time when the last published financial statements has

been half a year old and the next one expected in half a year. This time gap has been

closed with the assumption of trust that can be build up by extended risk disclosure

and helps investors to hold on to their shares during financial crisis situations. According to research in psychological factors that influence behaviour of market

participants, extended disclosure leads to a good reputation of a company (DIRK

2007). Trust by market participants that derives from a good reputation of a company

makes them hold on to the company's shares or to buy more. Bad reputation in

combination with a crisis situation might lead to short sighted behaviour by market

participants. This behaviour is recognized as "herding" in the literature (Hirshleifer,

Teoh, 2003) and makes market participants divest from shares of companies with bad

risk disclosure performance.

The share prices of listed European insurance groups do not decline during the

analyzed event of the financial crisis. This H4 hypothesis is tested by analysing

whether or not the share market valuation of the insurance industry dropped after AIG

announced its near insolvency. This is done by measuring the average standardized

cumulative abnormal return of all thirteen listed European insurance groups that sould be zero when no effect takes place and be negative when the share market valuation

dropped.

The next hypothesis is based aside from theoretical knowledge analyzed in chapter 3.

Based on the assumption that risk information are communicated between listed

European insurance groups and the capital market using structured ways like financial

statements the capital market could not detect differences between the insurance

group. The information about drops in the fair value of crisis relevant financial

instruments and impairments that would reduce the equity of insurance groups have

been disclosed later than the point of time when the event study that measured the

reaction of the capital market has taken place.

There is no relationship between the capital market reaction and how the modified

equity ratio of listed European insurance groups has been affected by the financial

crisis. This H5 hypothesis is analyzed by testing for a correlation between the

cumulative abnormal return that represents the capital market reaction and the proxy

75

for the negative effect that the recent financial crisis produced on the equity ratio

change from the consolidated financial statements as of 2007 to 2008. The correlation

must be statistically significant with a correlation coefficient within the confidence interval. This hypothesis test whether or not the capital market is efficient in detecting

those listed European insurance groups with the highest risk and the resulting drop in

their equity disregarding their risk disclosure performance.

The risk disclosure performance is not measured in an index that includes the

checklists of 2007 and 2008. Instead the analysis of each financial year is measured

separately. The reason for this is that the accounting requirements changed during the

crisis and that a search of key words has been performed in the checklist 2008 that

could not be performed in the checklist 2007. This separation allows additional

research on how listed European insurance groups changed their discretionary risk

disclosure. A prerequisite for these conclusions is that there is a relationship between

the results of the risk disclosure analysis for 2007 and 2008.

There is no relationship between the risk disclosure performance 2007 and 2008 of the

listed European insurance groups. This H6 hypothesis is analyzed by testing for a

correlation between the risk disclosure performance of listed European insurance

groups based on their 2007 consolidated financial statements and their risk disclosure

performance based on their 2008 consolidated financial statements. The correlation

must be statistically significant with a correlation coefficient within the confidence

interval.

The next hypothesis summarizes the relationship between capital market and listed

European insurance group that disclose risk information during the crisis situation. The

dependent variable that measures the response of the capital market towards the risk

disclosure and the two independent variables that measure the affect the crisis had on

the groups by their reduced equity ratio and their risk disclosure performance

measured in an index are combined. The goal of this hypothesis is to specify the

relationship between the capital market reaction towards risk disclosure tested in the

H1 hypothesis by adding the proxy for negatively effected by the financial crisis as

another independent variable.

There is no relationship between the capital market reaction and the risk disclosure

performance and how the modified equity of listed European insurance groups has

76

been affected by the financial crisis. This H7 hypothesis is analyzed by testing for a

correlation between the dependent variable of capital market reaction measured by

cumulative abnormal returns and the two independent variables risk disclosure quality 2007 and 2008 and the change in equity ratio as a proxy for the effect of the financial

crisis on listed European insurance groups. The multivariable regression analysis

necessary for hypothesis H7 should improve the predictability of the capital market

reaction by including two variables that influence it. The correlation must be statistically

significant with a correlation coefficient within the confidence interval and a higher R2

value than in the H1 hypothesis.

4.2. Selection of Variables The relationship between disclosure of listed European insurance groups and the

capital market is well researched (Verrecchia, 2001). However studies that measure

the capital market reaction on the risk disclosure performance of companies are rare, due to lacking data. To be able to predict capital market reaction as well as the effort listed European insurance groups make with their risk disclosure the researcher needs

a model that is based on measurable variables. In this chapter the implications for the

research methodology that has been analyzed in the literature review are applied to

the variables and documented how these variables are quantified in order to analyse their relationship during a crisis situation with quantitative research instruments.

The behaviour of the capital market is the dependent variable. It is influenced by the

risk disclosure performance and the negative effect of the financial crisis on the equity

ratio of listed European insurance groups as independent variables. The structure of

institutional ownership is assumed in the literature by Chung et al. (2008) as an independent variable but is classified as a control variable in this thesis due to a similar ownership structure among all analyzed insurance groups.

The four variables risk disclosure performance, the capital market reaction measured

with cumulative abnormal returns, the change in equity ratio as a proxy for the negative

effect of the financial crisis and institutional investors are described below.

Risk disclosure performance can be considered an independent variable in the

relationship between listed European insurance groups and the capital market during

77

crisis situations. Insurance groups can influence the dependent variable with the extent

of risk disclosure they pursue. The preference of investors for risk information remains

stable over time according to AICPA (1994) and EFRAG (2009) covering blocks such

as measuring risk and forecast information. As response to this crisis situation with its

dramatic drop in the fair values of investments the investment risk with its three sub- blocks market risk, credit risk and liquidity risk is added to these blocks and defined as the risk disclosure performance. The disclosure about the three sub-blocks of investment risk empowers insurance companies to explain how the unusual drop in fair

value of assets they have invested in affects their recent financial situation and can

predict future implications for the financial situation of their group. For the purpose of this research study disclosure about these risk factors in the consolidated financial

statements 2007 and 2008 of listed European insurance groups is measured using a

check list. The measurement process is described in subchapter 4.3.1. and documented in appendices A, B and C.

The dependent variable analyzed in this research study is the cumulative abnormal

return. It can be considered the reaction of the capital market in response to risk disclosure of listed European insurance groups aggregated in a short period of time.

The reaction of the capital market towards risk disclosure has been researched during

the Korean crisis situation by Baek et al. (2002). According to them Korean companies

with higher risk disclosure quality had a more stable share value over the crisis period then companies that disclosed less risk information (Baek et al., 2002). Although the

Korean financial crisis differed from the global financial crisis in 2007 and 2008 a

similar response of the capital market towards risk disclosure can be assumed for this

research study. But it should be considered that unusual phenomenon in the share

price as a response of the capital market towards certain events can exist according to

Frino et al. (2009). Two aspects that have come up in the literature review should be

applied to the accumulative abnormal return as a dependent variable of this research

study. First, the response towards disclosure of the capital market can differ in crisis

situation because of psychological aspects that influence the decision making process

of investors as analyzed in chapter 3.3.2. Another aspect that can influence the capital

market response is the access of private and institutional investors to risk information.

As analyzed in chapter 3.3.1. institutional investors do not solemnly depend on financial statements but have access to non structured ways of communication with

78

companies. Both psychological and information aspects can deter the response of the

capital market from the assumed response that is based on research by Baek et al. (2005). Whether an unusual phenomenon in the response of the capital market to a

crisis situation exists or not can be tested using the "event study method" which

measures the development of the share prices at days around a significant event. Share prices are called by the researchers "return" and consequently the deviation of

share prices form an individualized industry share index around a significant event is

called "abnormal return". The accumulation of all abnormal returns over the period

around the event of each listed European insurance group is called "cumulative

abnormal return" (CAR). Summarizing the cumulative abnormal returns of all

researched insurance groups is measured in the research result ACAR or Z. The

significance and direction of the abnormal return may indicate a phenomenon that

requires deeper analysis or it may confirm the assumed response that the capital

market holds on to shares of risk disclosing companies so that their share value

remains stable during a crisis situation. The measurement process of this dependent

variable is described in subchapter 4.3.2. and documented in appendix D.

Another independent variable is the change in equity ratio. This variable is necessary

as an indicator to classify listed European insurance groups in two categories. On the

one hand there are those listed European insurance groups that have not invested in

financial instruments that have dropped in fair value as a result of the financial crisis. On the other hand there are those insurance groups that invested heavily in those

mortgage portfolio based financial instruments to achieve higher returns. That

expected higher return as it turned out in the crisis has been connected to a higher risk

that materialized in a drop of the fair values in these financial instruments. Those listed

European insurance groups with a heavy exposure to these toxic financial instruments

and with significant losses due to their drop in fair value should have reported more

about the associated risk to investors via their annual consolidated financial

statements. Those insurance groups without this exposure did not need to disclose risk information to a similar degree. Therefore the change in equity ratio is used as a proxy to indicate the degree of investment by listed European insurance groups in toxic

assets that have dropped in value during the crisis.

Why is the change in equity ratio a useful proxy for the degree of exposure to these

toxic assets? In the recent financial crisis financial instruments quickly lost their market

79

value in three ways. First directly by a drop in the value of their underlying mortgage

portfolios, second by an indirect decrease of their values due to higher market interest

rates when they have been fixed interest rate bonds that led to lower fair values or fourth by default of bond issuers. Lower fair values lead to depreciation of these

financial instruments which in time leads to reduced equity and consequently to a

reduced equity ratio. The equity ratio reflects the loss of companies' fair value more

closely than the figure of depreciation which otherwise seems to be the most relevant figure to analyse the loss in fair value of international insurance groups. However, [AS

39 "Financial instruments: Recognition and Measurement" allows companies to

influence depreciation as a figure with effect on the profit and loss account depending

on the (AS 39 categories they place their financial instruments into and to interpret the

criterion of "lasting" lower fair values. But equity is affected by drops in these financial

instruments fair values regardless of the chosen categories either by impairments or by

unrealized losses. An additional reason why the figure of equity ratio is useful for

measuring the impact of the financial crisis is that equity is fundamental for companies

during a crisis because it increases the possibility of severely hit companies to survive

the economic struggle. The measurement of this independent variable is described in

subchapter 4.3.3. and documented in appendix E.

The last independent variable in the relationship between capital market and listed

European insurance groups are the institutional investors that own a significant part of the equity of those groups. It is analyzed because according to empirical research by

Eng and Mak (2003), Hotchkiss and Strickland (2003) as well as Chen and Cheng (2007) institutional investors as shareholders improve the disclosure quality of listed

companies. This theorem is based on the principal agent theory and assumes that long

term investors control the management by demanding additional information to be

disclosed. Due to a similar ownership structure the variable has the same number for

all analyzed listed European insurance group and can therefore be recognized as a

control variable. The measurement of this control variable is described in subchapter 4.3.4.

80

4.3. Measurement of Variables In the following subchapters methods are documented that have been used to

measure the one independent and two dependent variables and one control variable in

this thesis. Starting in subchapter 4.3.1. with the risk disclosure performance the steps lead to its measurement with a self administered risk disclosure index are presented. Then in subchapter 4.3.2. the behaviour of the share market in a crisis moment named

cumulative abnormal return as the dependent variable is measured using the event

study. After that in chapter 4.3.3. the change in equity ratio as a proxy variable for the

impact the financial crisis had on the each listed European insurance company as an independent variable is measured by the change in modified equity ratio in the

consolidated financial statements as of 2007 and 2008 of each insurance group. Last

the control variable of institutional investors is measured by defining the type of investors and the minimum level of significant influence. This measurement is

presented in subchapter 4.3.4. The measurement methods are stated to increase the

reliability of the research. All calculations and results are presented in appendices to

this research study.

4.3.1. Self-administered risk disclosure performance measurment To measure the risk disclosure performance of listed European insurance groups in a

self administered risk disclosure index four steps have been taken. As a first step the

analyzed content of risk disclosure has been defined. In a second step two checklists

for each analyzed financial year have been constructed that are based on weighted

categories and a scoring model that is based on investor information preference. The

third step has been reading and putting the relevant risk disclosure passages of the

consolidated financial statements in the two checklists. The forth and last step has

been the calculation of the scores in each checklist and the calculation of an index.

Each step is explained in the following. To increase the understanding of assumptions

made and decisions taken by the researcher a current research study by Ewelt et al. (2009) is cited and analyzed to provide a contrasting example.

In step one the content of risk analysis has been defined based on the results of the

literature review and the comparison of accounting requirements in chapter 3. Risk

disclosure is analyzed over the time period of January 2007 to March 2009 when the

accounting requirements remained fairly stable and two group financial statements

81

have been published by each listed European insurance group. In addition to

international accounting standards [AS 1, IFRS 4 and IFRS 7 national German

requirements such as GRS 5, GRS 5-20 and GRS 15 are used to provide the

categories that listed European insurance groups can fill with their discretionary risk disclosure published in their consolidated financial statements. This category based

measurement concept is based on empirical research by Core (2001). Based on the

result of chapter 3.1.3. the highest level of risk disclosure required in one of the above

mentioned accounting standards is taken as the necessary risk disclosure level in the

checklists. This content of risk analysis has been implemented as categories in the

checklist for the analysis of the consolidated financial statements 2007 and 2008.

Risk disclosure in the group financial statements 2007 (referred to as pre-crisis) has

been recognized according to the investment risk, the measuring risk and forecast

information. The content of risk analysis for the financial year 2008 has been modified because during that period of time the crisis worsen and reached its peak. This lead to

a change in accounting requirements and offered listed European insurance groups the possibility to disclose additional information to crisis related topics that had become

of public interest. The IASB as the standard setter for international accounting

standards released amendments to IFRS 7 at the beginning of 2009 (IASB, 2009a).

These amendments eased the pressure on companies to write down investments that

negatively affecting their profit and loss account by reclassifying assets into other [AS

39 categories. These easing requirements could be considered by listed European

insurance groups in their group financial statements 2008 (referred to as post-crisis). Risk disclosure in the group financial statements 2008 included in addition to the above

mentioned risk blocks the documentation whether or not listed European insurance

groups used the IFRS 7 amendments or not. Additionally the disclosure of the

companies about certain urgent topics that came up during the crisis is considered

relevant risk disclosure for the analysis of the 2008 consolidated group financial

statements. This corporate risk communication on urgent topics has been included as

a criteria in the past crisis checklist 2008. In order to find this qualitative information in

the consolidated financial statements 2008 on this new category key words have been

used. These key words have been selected from an report by Business Insights (2008)

"Impact of US subprime crisis on the insurance industry" that provided key words

82

relevant to the insurance industry in the crisis. These key words have been reconciled

with crisis relevant key words stated by Scharpf and Schaber (2009).

Due to this modification of the risk content. that has been analyzed in the 2008 post

crisis consolidated financial statements two checklists have been set up in a second

step. The approach to use checklists for the risk disclosure performance measurement is recommended by Chen (2006) and has been used in other studies by Rahman

(1998) as well as Cheng and Courtney (2006). These checklists build the data and

scoring basis for the self administered risk disclosure index. Such indices have been

recommended by Hussainey et at. (2003) for future research on disclosure

performance in the absence of the discontinued US-based disclosure index by AIMR

as mentioned in chapter 3.1.2. Critical parts of such self administered disclosure

indices are to weighten categories and to score the content of risk disclosure. Both

parts are subjectively performed by the researcher and reduce the reliability of the

research results. To minimize this subjective influence the information preference of investors published in previous research has been used to weighten categories and to

score content in the self administered risk disclosure index used in this research study. Based on research by AICPA (1994) and EFRAG (2009) the three block of investment

risk, measuring risk and forecast information have a similar importance for investors

that remained constant over the years. However in this crisis situation investors

requested more risk information about the prognostic value of companies and about the effects of drops in the fair value of investments on the companies. This crisis based

modification of preference is recognized by adjusting the weight of the forecast block

categories in the checklist to 40%, the weight of the investment block categories to

35% and the weight of the measuring block categories to 25%. This modification has

been applied uniformly in the pre crisis checklist on consolidated financial statements 2007 and on the post crisis checklist on these group reports 2008.

Each risk block consists of sub-criteria that are specified in international and national German accounting requirements. The weight of the block had to be distributed among these sub-criteria. This distribution could barely be based on previous research on investor preference because the accounting standards changed too often during that

period of time so that empirical studies that collected the preference of investors could

not keep up with the speed of change. Only the insurance industry specific adaption of

83

the German disclosure index (see chapter 3.1.2. ) that is published annually by Balzer

(2008) provides an orientation for the distribution of weights to the sub-criteria. The

weight of sub-criteria such as the amendments of IFRS (IASB, 2009a) or the key word

related sub-criteria could not be based on research and have been applied subjectively

by the researcher. The weight to each sub-criteria is documented in the checklist 2007

and 208 in appendices A and B.

Measuring the risk disclosure performance of listed European insurance groups

require a judgement on the quality of information that have been published in their

consolidated financial statements 2007 and 2008. All the accounting requirements

used to create sub-criteria are vague as to whether only qualitative or quantitative

information are required. International standards such as IAS 1, IFRS 4 and IFRS 7 as

well as national German standards such as GRS 5, GRS 5-20 and GRS 15 allow

alternatives or different level of detailed information depending on the situation of each

specific company as analyzed in chapter 3.1.3. In order to compare the risk disclosure

performance of listed European insurance groups each published risk information had

to be scored according to its usefulness to investors. Former research by Armeloh

(1998) discovered in a survey of users of financial statements that they regard

quantitative data as more informative than qualitative ones. Another discovering by the

same researcher has been that company specific information are preferred by

investors to general information. Both preferences have been implemented into this

research study by setting up following scoring model.

Quantitative information on sub-criteria is rewarded with five points. Entity-specific

qualitative information is rewarded with three points. General qualitative information

that is not related to specific situation of the listed European insurance group is

rewarded one point. When no information on the sub-criteria has been disclosed then

one point has been subtracted. This has been done because it is possible to

distinguish between information that a company must not or cannot disclose and information that companies just omit. This different approach of companies towards

information demands of investors is recognized in the scoring model based on

research by Armeloh (1998). This principal of punishing companies that are unwilling to

disclose by subtracting points had to be applied to the easying of accounting

requirements by IASB (2009). Each listed European insurance group had to state

whether or not it has facilitated the choice to restructure their investments into different

84

AS 39 categories. No information on this sub-criteria would result in the deduction of

one point. A negative statement that this accounting choice has not been utilized by the insurance group would result in three points. A listed European insurance group that has used this choice is required to disclose the quantitative effect of the

recatogrization on the profit and loss account and equity. The disclosure of this

quantitative risk information would result in five points granted.

An additional step has been included into the scoring process in order to increase the

transparency and to document the scoring decision. The additional documentation of the scoring process of each sub-criteria should increase the reliability of this research

study and enable other researchers to redo the scoring process with the same results. Each sub-criteria included in the pre-crisis checklist and the post-crisis checklist has

been thoroughly analyzed concerning its detailed requirements. For this purpose the

text of each relevant national or international accounting requirement has been stated. After that the scoring classes have been specified to each sub-criteria. As a result of this additional documentation every scoring process can be traced from the index

score back to the score on the subchapter and the company's disclosed risk information out of its consolidated financial statements 2007 and 2008 that in stated in

the checklists. The scoring documentation is added as appendix C to this research

study.

As a third step the consolidated financial statements as of 2007 and 2008 of all thirteen

listed European insurance groups have been analyzed. All these financial statements have been accessed via the internet page of the insurance groups. The notes as part

of the consolidated financial statements of the researched insurance groups have been

read completely by the researcher. Relevant text passages and numbers from the

notes have been copied and placed into the checklist. Risk information in the

consolidated financial statements 2007 have been included into the pre-crisis check list

2007 and those risk information from financial group reports 2008 have been

transferred into the post-crisis checklist 2008. This checklist has been filled additionally

with risk information concerning crisis relevant topics that have been disclosed by

listed European insurance groups. The crisis relevant risk information has been

searched by key-words using the search function of the word-processing programme

within the consolidated financial statements 2008 on the companies' internet pages.

85

All results of this third step are documented in appendix A and appendix B. The filled

out and rated checklists of all thirteen listed European insurance groups are attached

as these appendices.

The process of reading the financial statements, detecting relevant information and

copying it into the pre-crisis and the post-crisis checklist has been tested with a dummy

financial statements. The consolidated financial statements of ERGO Versicherungs-

gruppe AG, a company not included in the unit of analysis, has been used to test this.

No adjustments or corrections have been done as a result of this dummy testing.

As a fourth step the scores of the checklists have been calculated. A prerequisite for

summing up all points is that the risk disclosure of each listed European insurance

group has been graded. These grades have been multiplied with the weight associated

to each sub-criteria. The weighted grades have been added up to the level of the

blocks investment risk, measuring risk and forecast information. In recognition of the

critic in chapter 3.1.3. that listed German insurance groups have an advantage because they are required to disclose forecast information whereas other listed

European insurance groups do not oblige to similar national requirements two different

results of each checklist have been calculated. One result that includes all three block

investment risk, measuring risk and forecast information and another result without the

scoring of the forecast block have been calculated for each checklist. This

differentiation proved useful in the quantitative analysis in chapter 5.1. because the

forecast information have been deleted from the further research study in order to have

a level playing field of listed national German and European insurance groups. Instead

of an index that would level detailed information a variable for each checklist has been

created. Therefore a risk disclosure reporting performance 2007 and a risk disclosure

reporting performance 2008 have been measured. The degree of fulfilment of the risk

criteria have been measured in percentage. A result of each year form 0 to 1 is

possible for the risk disclosure performance variable for each analyzed financial year.

Following comparison between a recent empirical study on risk disclosure by Ewelt et

al. (2009) and this research study should demonstrate the difficulties in measuring risk disclosure performance. Ewelt, Knauer and Sieweke (2009, p. 706) analyse the risk disclosure performance of thirty German industry groups that are listed in the German

86

share index DAX. The researchers focus on risk disclosure because of the "clearly

defined accounting requirements that limit choices of companies" in preparing their

group financial statements. This would increase the objectivity of the research study

according to Ewelt et al. (2009). The researchers ignore the concept of discretionary

disclosure by Core (2001) and the unravelling principal by Leuz and Diehm (2010) to

get comparable risk information. By analysing the risk disclosure performance of listed

insurance groups form different European countries the concept of Core (2001) and Leuz/Diehm (2010) should be applied in this research study. To illustrate the necessity

the formal accounting requirements of IFRS 7 are equal but the risk information

disclosed is also based on additional or supplementary national accounting

requirements. This makes it difficult to compare the quality and quantity of risk

disclosure. Only by assuming that each listed European insurance group can exceed

these differences because an industry wide risk disclosure level is recognized makes a

comparison of them possible. The companies analyzed in the research study by Ewelt

et al. (2009) exclude listed banks and insurance groups because of their different

business model and varying accounting requirements. This limitation creates a homogenous group of companies that can be compared in their risk disclosure,

however significant the industries that have been directly affected by the financial crisis like banks or indirectly affected like insurance groups are not considered in the

research study by Ewelt et al. (2009).

Differing from research by Ewelt et al. (2009) this research study specifies risk disclosure demand of investors according to their long term preference to a specific

crisis situation. Rooted in empirical background (PwC, 2007 and Coval et al. 2009)

investment risk, measuring risk and forecast information are seen to be relevant to

investors in the analyzed financial crisis. These three information parts are the basis of

the risk-disclosure checklist that has been used in this research study to measure the

risk disclosure performance. These information parts have been operationalized in

similarity with the research study by Ewelt et al. (2009) by applying accounting

requirements and using their detailed requests codified in the accounting standards as

categories and sub-criteria.

The risk information provided by companies in their group financial statements about these sub-criteria is scored. Ewelt et al. (2009) use a0 to I scale instead of a0 to 5

87

scale as in this research study. In difference to this research study Ewelt et al. (2009)

measures in addition to the quality also the quantity of risk disclosure by companies. This approach is criticized by Schleicher et al. (2007) who states that quantity does not equal quality in the risk disclosure. Other studies used the amount of disclosure

measured in pages or words as a proxy for disclosure performance of companies (Botosan 1997, Lang and Lundholm 2000). The approach in this research study is to

analyse the quality of risk disclosure performance solemnly disregarding its quantity.

Another difference in the two research studies is the information preference of investors that is recognized in one and is not recognized in the others research

methodologies. Ewelt et al. (2009) assume that investors value every risk information

provided in the consolidated financial statements the same way. This general

approach is contrasted in this research study as it adapted weighten sub-criteria to

crisis situations. Doing so this study approach assumes that investors appreciate

certain risk information more than others, depending on what they regard as important

during the crisis. The weighten of criteria is based on investor preferences that has

been empirically researched, however certain adaptation within the risk blocks have

been set subjectively in this research study. Both studies measure the index value as a

percentage of the maximum points that could possibly be granted to each disclosure

performance.

4.3.2. Measurement of the capital market reaction variable (CAR) Measuring the cumulative abnormal return of European insurance group shares by

applying the event study method at the peak of the financial crisis should indicate any

share value phenomenon. The result of this measurement is a prerequisite to verify an

analogy in the observation of Baek et al. (2005) gained during the Korean financial

crisis that in a crisis situation companies with high risk disclosure performance remain their share value. The test of a relationship between risk disclosure and the capital

market reaction measured in the individual cumulative abnormal return of each listed

European insurance group will be performed in chapter 4.3.4. using quantitative research methods.

88

The measurement of the accumulative abnormal return of listed European insurance

groups that are part of the unit of analysis takes place in five steps. After an introduction to the method as a first step the theoretical background of the event study is presented in a second step. The timing of the event study is critical to the useability

of its result. Unlike in other event studies in which a cause and reaction take place at the same point of time this event study is based on a longer lasting cause effect that

influences the reaction at a crisis point of time. Not the immediate response of the

capital market toward announced trades (Frino et al, 2009) or towards changes in the

accounting policies of companies (Cornett et al, 1996) is measured in this research

study but how trust based on risk disclosure performance of companies affects the

financial decision making of investors during a crisis situation. Assuming a long lasting

cause that arises from the publication of annual financial statements the share price

effect at a point of time is tested. The reasons why a certain point of time has been

chosen to perform the event study are presented as a third step. In a forth step

measures to improve the information usefulness of the event study results are described. This is being done by eliminating confounding events that are stated in this

section. In the last and fifth step the formulas applied in the event study are listed that

build up to the cumulated abnormal return. The event study is based on data that

Bloomberg as an independent financial information system provides on a daily basis.

The data have been gathered for one hundred share market trading days before and five trading days after the event. For this research study GoldmanSachs provided the

share market data.

As a first step the event study is introduced. Event study methodology has been

started by Ball and Brown (1968) as well as Fama et at. (1969) and further developed

by Brown and Warner (1985). Event studies are designed to examine market reactions

and excess returns around specific information events. They measure the impact of a

specific event on the value of a firm according to MacKinlay (1997). These studies focus on the relative changes in share price following the effect of an event. Over the

years this research method has been applied to a large number of events. Announcements of dividends, earnings, mergers, capital expenditures and new issues

of share are a few examples of vast literature in this area (Ross et al., 1996).

What the event study does, it measures the abnormal return of shares during a time

period around the event (the "event window"). The definition of abnormal return is one

89

that exceeds the return justified by the riskiness of the investment according to Besley

and Brigham (2000). Applied to a crisis situation this definition must be enlarged to

lower return that justified by risk due to an overreaction of the capital market caused by

psychological factors as stated in chapter 3.3.2. For the period of the event window the

abnormal return is contrasted to the so-called "normal return". The normal return is

determined by applying the same deviation of the company's share price from. a

reference index which had been calculated for a time period before the event window (the "estimation window"). The cumulative abnormal return is the sum of the

differences between normal and abnormal return for over the event window. The

reference index used in this research study is Bloomberg BINSUR, an industry specific index that includes in addition to the thirteen IFRS insurance groups that are part of the

unit of analysis other companies of the same industry that apply national accounting

requirements such as US-GAAP. These later mentioned companies are not included in

this study.

The useability of the abnormal returns for this analysis has been verified by testing the

trade volume of each day in the event window. Trade days with low trading volume

tend to result in unrepresentative share returns which impair the informative relevance

of the event study results according to Andersen (1996). The analyzed trade days all had trade volumes of at least 30% of the mean trading volume of each listed European

insurance group during the longer estimation window. Based on this the results of the

event study can be considered fit for further research according to Brailsford (1996).

In a second step the theoretical underpinning of the event study method is presented. Event study methodology makes the assumptions of rationality in the marketplace so

that the effects of an event will be reflected immediately in security prices and that

asset returns are jointly multivariate normal and independently as well as identically

distributed through time according to MacKinlay (1997). In case these statistical

preconditions are not met nonparametric event study tests exist and can be applied

according to Corrado (1989) and Cowan (1992). However concerns about the

useablility of results from the event study have been eliminated by McWilliams and McWilliams (2000). According to them normal distribution can be assumed despite

the limited sample size of thirteen listed European insurance groups in this research

90

study. McWilliams and McWilliams recommend the event study method for even

smaller sample sizes.

Another central assumption is that the event has been unanticipated by the market

participants. In order to verify these assumptions researchers are urged to check for

confounding events, that are additional events that took place during the research

period and influenced market participants without formerly being addressed in the

research analysis. The effects of these confounding events should be eliminated in

event studies according to McWilliams and Siegel (1997).

The theory which is central to the practice of event studies is the operation of efficient

capital markets, and the efficient market hypothesis. As stated by Ross et al. (1996)

an efficient capital market is one in which share prices fully reflect available information. Besley and Brigham (2000, p. 315) define the efficient market hypothesis

as 'the hypothesis that securities are typically in equilibrium - that they are fairly

priced in the sense that the price reflects all available information on each security'. When a share continues to react to an unfavourable development, it often signifies

adjustment to the new pieces of information about the situation that has become

available, rather than over a long adjustment period. This makes event studies an important tool to analyse market reactions during crisis situations. This ability of the

market to adjust to new information comprises the foundation of the efficient market hypothesis (Besley and Brigham, 2000).

Event studies are an ideal tool for examining capital market reactions. The results of the event study have a higher information usefulness in those cases where the sample

companies have a common characteristic such as being members of the same industry (MacKinlay, 1997). This method can therefore best be used to evaluate the

capital market opinion about companies that are similar in size, business model and belong to the industry such as the listed European insurance groups that are being

studied in this research project. Although the assumptions regarding rationality and

efficiency of the capital market remain doubtful during crisis situations the event study

method is applied to this research study because by isolating a single effect within a

narrow time frame they control for important differences within firms that cannot be

observed by other means according to King and Lennox (2001).

91

The timing of the event study as the third step is critical for analysing the relationship between risk disclosure and capital market reaction that is performed in chapter 4.3.3.

Not the direct cause and reaction could be tested in this research study but whether or

not trust that has been build up with extended risk disclosure holds in a crisis situation. Therefore the timing is essential for the reaction of the capital market and should take

place at the peak of the crisis in order to minimize the influence of other factors on the

capital market that are not considered in this research study. The event study method

as used in this research methodology needs a specified date to be applied to. Previous

research by Gonedes and Dopuch (1974), Leftwich (1981), Barth (1994), Barth et al. (1992) and Beatty et al. (1996) used the release of new fair value accounting

requirements to test the relationship between companies' disclosure and capital

market. Because the release of new accounting requirements or better the relaxation

of these requirements (JASB, 2009a and 2009b) has been the reaction of the standard

setter towards the financial crisis dates connected to their release could not be used.

The event study with a research date is preferred over research methods that analyse data over a longer period of time using arithmetic averages that do not recognize the

character of a crisis and reduce the explanatory power of crisis data. The climax of the

financial crisis has been set as a research date for the event study. This point of time

can be specific to each industry and it enables researchers to test the relationship between accounting disclosure and the reaction of the capital market shortly before

and after the event.

The near insolvency of AIG - then the biggest insurance group of the world - on September 14th 2008 has been chosen as the event date because it can be singled out

as an unanticipated event of equal significance to all analyzed listed European

insurance groups (Goldfarb, 2008). The effect has been unanticipated because on September 16th 2008 AIG's share price fell more than 90 percent according to

Brunnermeier (2009). Such massive share price reaction epitomizes the loss of trust

that is important for the long term business of insurance.

The event study method has been applied and the cumulative abnormal return of each insurance group calculated in an event window of 10 days before and 5 days after the

event date of September 14'h 2008. The estimation window is the period starting 99

days before the event date until September 14th 2008. According to Ryngaert and

92

Netter (1990) the length of 99 days chosen in this research study as an event window is short enough to capture the significant effect of the event. Because the near bankruptcy event of AIG took place on a Sunday the information

following the publication, particularly of newswire articles, information was made

available to investors by news-services and newspapers almost immediately and could therefore influence their trading behaviour from Monday which can be considered the

day after the announcement even on the same day of the announcement. In this

specific case due to the event on a Sunday the information could not leak prior to its

announcement and be used for share trading before the information but has been

widely available on Monday when trading started. However, often information has been

found to leak prior to the official announcement date so it is possible that the event

studied could have influenced the share prices of the days preceding the

announcement (Ross et al. 1996). Therefore, taking the information leakage factor into

consideration and knowing that difficulties of AIG were already known the event

window includes 10 days prior to the announcement of AIG near bankruptcy at September 14th 2008 and 5 days after (-10, +5).

As a forth step two corrections in the event study structure have been made to

increase the informative value of its results. The share market data and the event study results included in the event window ten

days prior and five days after the event date have been tested for confounding events

applying the recommendations of McWilliams and Siegel (1997). It has been checked

whether or not active mergers or acquisition processes took place during that period of time that might have influenced the share prices of the researched listed European

insurance groups. Doing so the ZEW-ZEPHYR M&A index has been checked for that

period. It can be assumed that no such activity took place because the index has been

on its lowest level for four years in September 2008 when the event date has been set. The web-sites of the analyzed listed European insurance groups have been tested on dividend announcements during that time and no indication has been found. However

September 19th has been a triple witch day on share exchanges with unusual trading

activities. Based on recommendation by McWilliams and Siegel (1997) the data and the results of this day have been eliminated from the research study. The same has

happened to the data and results from September 8'. At that day floods in China and hurricane season in the US have been topics of the day. Around that date an industry

specific gathering of international reinsurers took place in Monte Carlo. Within this

93

gathering reinsurance contracts for future time are closed or previous ones modified. This might explain the unusual share price reaction of this day, because the insurance

industry being in the spotlight in Monte Carlo has been sensitive to information about

possible claims that arise from the hurricane and floods catastrophes. The elimination

of the data of both trade days reduced the data to thirteen from the original fifteen

trading days in the event window. By doing so the researcher tried to eliminate trade

day results with inducements that undue influence the results of cumulative abnormal

return. The theoretical aspects of the four steps are considered in the application of the

event study method. This is documented in appendix D.

As a fifth step the formulas needed for an event study and the prerequisite of regression for each listed European insurance group in the market model are presented as follows. The abnormal return of insurance group shares during the event window can be

calculated as the ex post return of the securities over the event window minus the

normal return of the firm over the event window according to MacKinley (1997). To

price data of securities the Bloomberg 500 Insurance Index called BEINSUR is used. For this event study the constant mean return model is used (MacKinlay, 1997 and McWilliams, Siegel 1997). Although the constant mean return model is the simplest

model, its results are often similar to those of more sophisticated models according to

Brown and Warner (1980).

Normal returns were assumed to be modelled for the share prices by the Ordinary

Least Squares (OLS) market model. The calculation of the regression parameters has

been performed using a scientific internet site

www. arndt-bruenner. de/mathe/scripts/reqr. htm which is freely accessible and verified by applying SPSS as a statistical software.

The rate of return on the share price of insurance group i on day t is expressed as:

R;,, =a+ßR,,,, t +e,,, (1)

where

R,,, = the rate of return on the share price of insurance company i on day t

94

R,,., = the rate of return on a market portfolio, the Bloomberg Insurance Index

BEINSUR on day t

a, ß are the regression parameters for firm i, is the return on the share of each

insurance group in the sample group for day t

e;, t is the residual term or prediction error.

The definition of an abnormal or excess return is one that exceeds the return justified

by the riskiness of the investment (Besley and Brigham 2000). Thus, the reduction of

excess or less abnormal returns associated with investors trust due to high disclosure

performance on crisis related information were estimated by using the OLS market

model, solving for the residual term as follows:

ARi,, =Ri,, -(ä+ßRmt) (2)

AR,,, = the abnormal return for company i on day t, which is the residual term in the

regression equation. The abnormal return, based on the market model discussed by McWilliams and Siegel (1997) for company i on day t using

slightly different denuminations: AR; t = Rit - (at + btRmt) where at and bt are the ordinary least squares (OLS) estimators, calculated over the estimation

period, for the market model for insurance group i

R,, = observed arithmetic return for insurance group i on day t

R, �, = the return on the specific insurance index m for day t

ä ,,

ß or a and b according to McWilliams and Siegel (1997) = the Ordinary Least

Squares estimates from the regression of return data on firm i, with index m, for both their respective values during the estimation period days (-100, -11).

Sit = the standard error of the OLS market model regression for firm i T= the number of days in the estimation period K= the number of days in the event window

95

Process to calculate b und a: Based on the website http: //www. arndt-bruenner. de/mathe/scripts/reqr. htm the

parameters and the standard error for each of the listed thirteen European insurance

groups have been calculated and verified by applying SPSS a statistical software.

Results OLS analysis for a, b and St:

9 Allianz AR; t = Rit - (-0.09 + 0,64 * Rmt)

St=1,0086665698508985

9 AXA AR; t = R; t - (-0,13 + 0,52 * Rmt)

St = 0,94959529954217

. Generali

ARit = Rit - (- 0,16 + 0,13 *Rmt)

St=1,64994748443866

9 Aviva

AR; t = R; t -( -0,18 + 0,03 * Rmt)

St=1,653157058765464

" ING ARit = Rit - (-0,1 + 0,56 * Rmt)

St=1,1117734391394798

" Aegon AR; t = R; t - (-0,05 + 0,51 *Rmt)

St=1,127421364502298

" CNP Assurance AR; t = R; t - (-0,24 + 0,65 *Rmt)

St=1,2868315026554742

96

" Legal & General AR; t = R; t - (- 0,18 + 0,04 *Rmt)

St= 1,6517209442920797

9 Zurich

AR; t = R; t - (- 0,17 + 0,17 *Rmt)

St=1,6294925332317694

" MunichRe

ARit = Rit - (- 0,06 + 0,73 *Rmt)

St=1,2355239115045748

" Old Mutual AR; t = R; t - (- 0,19 + 0,0 *Rmt)

St = 1,655074658371793

" Standard Life

AR; t = R; t - (- 0,19 + 0,03 *Rmt)

St=1,6531482953882306

" Hannover Re

AR; t = R; t - (- 0,12 + 0,38 *Rmt)

St=1,401592243298169

According to the efficient market hypothesis, firm i's share abnormal return at time t,

AR,; should reflect the release of information at the same time. The expected value of

AR;, is zero if investors view the near bankruptcy of AIG as not conveying significant

information to consider it for their market activities. Therefore, in order to derive at the

excess or abnormal returns for every company, first the ä and ß respectively a and b

parameters were estimated for each firm using a OLS calculation method. These

regression parameters resulted from the regression of return data on firm i with index

97

m for their respective values during the estimation period days, which is the interval of 90 days before the event window (-10, +5).

Many researchers calculate a standardized abnormal return SAR following Dodd and Warner (1983) where the abnormal return calculated is standardized by its standard deviation:

SARit = AR/i / SDit

With Wit is the standard deviation or AR; t:

(Rmt-Rm) sDit = St 1i+! +

2

T2 E, y(Rmt-Rm)

Where R. represents the mean return over the insurance group's estimation period of the insurance index.

k

CAR ik ± SAR ;t W 1=1

Where k is the number of days in the event window. This number has changed with the

confounding events that have been taken into account during the research. Thus the

number of days considered in the event windows has been reduced from 15 to 13.

This formula leads to the individual cumulated abnormal return of each insurance

group, which provides a result that can be analyzed individually for each listed

European insurance group and used in correlation analysis in chapter 4. It is assumed

that the individual values of CAR; are independent and independently distributed and

therefore t distribution is used as in the guiding literature by McWilliams and Siegel

(1997).

The following equation does provide information about the abnormal return of the

whole research group. Parameter n is the number of insurance groups included in the

research sample.

98

_4n ACAR ;=1+TE CAR it n2 t-1

The average standardized cumulative abnormal return has been -10.1 when the two

trading days as confounding events have been eliminated.

Alternatively the test statistic Z can be used to express the effect the AIG near

insolvency has had on all listed European insurance groups together.

Z=ACARx-

The cumulative abnormal return of 36,5 is significant. It can be stated that the near

insolvency of AIG had a significant impact on the share values of the thirteen

researched listed European insurance groups. Further analysis will be based on the

CAR results of each insurance group. The data, the formulas and the application are

documented in appendix D for all analyzed listed European insurance groups. The

relationship between the CAR results as the independent variable and risk disclosure

performance as well as change in modified equity ratio as dependent variables will be

analyzed in chapter 5.2.2.

4.3.3. Change in modified equity ratio 2007 to 2008

The drop in the fair value of derivative financial instruments had a "devastating impact

on companies during the Korean financial crisis" according to Rahman (1998, p. 17).

These drops all affect the equity of companies that prepare their books according to

IAS 39 "Financial instruments" (IASB, 1989). However some listed European insurance

group can cope with such a negative financial impact better than other ones. To make

the impact comparable ratios can be used that relate the affected equity to assets

under management or total assets. Based on research by Nissim and Penman (1999)

equity ratios are research tools that can be used to eliminate effects that derive from

other economical or accounting related factors and to compare the economic

performance of different companies. As listed European insurance groups all had the

same ability to add equity from the capital market without a restricting ownership

structure. Thus the starting equity ratio as of December 31st 2007 has been similar to

all researched insurance groups.

99

However change in equity from one financial year to the next does not solemnly

consists of the financial results of core business of a company. That is why changes in

equity are explained to investors in the change in equity statement that is according to

IAS 1 (IASB, 2003) a mandatory part of consolidated financial statements. This

statement lists all effects on the equity within one financial year and classifies them.

Because no industry-wide description of the classes exist the researcher tried to

distinguish and compare all equity effects in the analyzed change in equity statements

as of 2007 and 2008 in the consolidated financial statements of the listed European

insurance groups. Depending on their business structure and their international

business locations and activities these insurance groups disclosed in their change in

equity statements effects such as reserves for currency transactions, transactions

among subsiduaries and tax effects. These effects have been subtracted from the

starting equity ratio of all insurance groups. Also accounting effects and factors specific

to each listed European insurance group such as changes through shadow

accounting, unspecified reserves and changes in pension plans schemes have been

eliminated from the starting equity. Another subtraction from starting equity stated in

the 2007 consolidated financial statements has been all changes in equity that are not

based on core business activities and did not affect the insurance specific

performance. These changes relate to own shares transactions and buy back

programs, dividends paid and other distributions made to shareholders, the issue of

new share capital and changes in subordinated debt that can be considered as a

substitute for equity.

Based on the information disclosed by the listed European insurance groups in their

change in equity statements the above mentioned substractions have been made. To

arrive at ratios the stated equity as of consolidated financial statements 2007 has been

cleaned from all the above mentioned non core business factors. The resulting figure

for each insurance group has been divided by its total assets stated in their

consolidated group financial statements 2007. To arrive at the modified equity 2008

the cleaned 2007 equity is used as a starting point and the above mentioned non core

business changes in equity stated in the 2008 consolidated financial statements

substracted. The resulting equity figure has been divided by the total assets stated in

the group financial statements 2008.

100

The change from these two ratios has been used as an independent variable in this

research study. The change in percent from the modified equity ratio 2007 to the

modified equity ratio 2008 is the analyzed variable. The economic sense of this

variable is the impact by the drop of fair value in investments that had to be accounted for by reducing equity. All disturbing factors that are specific to each company's size, location, organization etc. have been eliminated in order to purify the unrealized losses

and impairments on financial instruments. Especially governmental support that have

been provided to struggling listed European insurance groups have been eliminated. They appeared especially in the consolidated financial statements 2008 as share issues or subordinated debt issues and have been eliminated in order to provide a

same level playing field approach for all listed European insurance groups that are

analyzed in this research study. To improve the reliability of this research the

measuring method to calculate modified equity ratios are presented in appendix E.

4.3.4. Percentage of institutional investors of stated equity Institutional investors are considered an independent variable in this research study, however due to their measurement they have become a control variable. Institutional

investors control the management with their demand on additional disclosure as

analyzed in chapter 3.3.1. In order to arrive at a figure on institutional investors in listed

European insurance groups two steps have been taken. First the groups of investors

that can be considered institutional must be defined. Second the level of this

institutional investors sufficient to assume their significant influence must be set.

For this research study institutional investors are defined as families, trusts, significant individual investors, banks or insurance companies. This definition is based on similar

empirical research by Chung et al. (2002).

More disputed in the literature is the level of institutional investors on the shares that is

necessary for them to have a significant influence on the business activities and the

connected accounting disclosure decisions of the management. The lowest level used in empirical studies is 5% of institutional investors owning of shares by Eng and Mak

(2003) as well as Chen and Cheng (2007). Both research studies analyzed single

substantial shareholders that can influence the business and accounting decisions of

101

the management. This approach is not followed in this research study because all listed European insurance groups analyzed have such a size and global expansion that a single investor owning shares on this level is unlikely. A higher level of 10% that

are directly or indirectly held by institutional investors is used in research by Attig et al. (2006). This level is the same as the one used in a research study by Chung et al. (2002) and will be followed in this study. To gather the information about institutional

investors this research relied on the internet websites of the listed European insurance

groups that are part of the unit of analysis. Twelve out of thirteen insurance groups

provided the necessary information on their websides. The researcher tried to gather

the missing data on one insurance group 'Standard Life' by sending emails to the

investor relation department that have not been answered. The missing data have

been assumed on the basis of the majority of the ownership structure of the other twelve listed European insurance groups. The date the share ownership structures have been analyzed and published on their websites differed among the insurance

groups. Ownership analysis dates strayed from year end 2005 to mid 2010 on the

websites of the insurance groups.

When institutional investors exceed the 10% level they are considered as a1 and

when they fail to reach this level a0 is measured in the statistical research in chapter 5.3. All analyzed listed European insurance groups have surpassed the 10%

institutional ownership level by far. All insurance groups are considered in the

statistical analysis with their value 1. This makes institutional investor a controlling instead of an independent variable for this research study.

4.4. Research Methods for Data Processing Data processing includes the collection of data, the processing of data and the

analysis of data. These steps are undertaken as a preparation before applying the

research tools and partially performed by using computer programmes.

The collection of data has been performed by the researcher to gather data on the

variables risk disclosure performance, change in modified equity ratio and institutional ownership of listed European insurance groups. The data on the capital market reaction has been provided by GoldmanSachs and verified in sample by the

researcher using Bloomberg information that is accessable via internet.

102

The processing of data has been done using Excel sheets that include formulas and

other calculations. The processing of the data about risk disclosure performance has

been detailed for each criteria and documented in appendix C. The processing of data about the capital market reaction has been done in excel sheets. The different

formulas that are applied to calculate the ACAR and Z results are included and step by step documented in the excel sheets which are attached to this document as

appendix D for each listed European insurance group. The formulas and connections

of data on the capital market reactions have been checked by an independent

researcher to avoid mistakes. The data for the modified equity ratio has been

collected from the consolidated financial statements 2007 and 2008 of each insurance group. The method to deduct positions from the equity is documented in

appendix E. The data about the ownership structure has been gathered from the

internet websites of twelve listed European insurance group. Standard Life as the

only insurance group did not provide sufficient data about their ownership. It has

been assumed that this insurance group surpasses the level of a 10% institutional

ownership share that all other listed European insurance group as well.

The processing of the data has been done using 'Statistical Package for Social

Science' (SPSS) in Version 17.0 that has been provided by the university. For the

correlation analysis a code book with all variables, their SPSS names and coding instructions has been prepared.

The analysis of the data has been performed based on graphics. This has been done

in order to check the distribution using SPSS and it has been used to cluster the

listed European insurance groups based on the change in equity and their risk disclosure performance as of 2007 and 2008. This has been done to detect those

insurance groups that have been extremely exposed to risk from mortgage based

portfolio financial instruments and did not sufficiently disclose this risk position in their

consolidated financial statements 2007 and 2008. These insurance groups can be

considered high risk shares for investors. The same has been performed for

companies with low equity ratio reduction and better risk disclosure performance. These insurance groups can be considered low risk shares for investors. The

clustering is documented in appendix F and G. It has been used as a preparation for

the expert interviews that have been applied as qualitative research tools in this

thesis.

103

4.5. Data and Sample Selection The selection process of companies that are analyzed in research studies should be

based on logical restriction according to Kraft and Nolte (2005). The number of

companies that are included in the group of analysis is limited due to two logical

restrictions that derive from the literature review in subchapter 3.2.4. The first logical

restriction is that similar accounting requirements are necessary to compare risk

disclosure performance on a detailed level. The second logical restriction is that all

analyzed companies should belong to the same industry to have a coherent capital

market response that is not diffused by effects that influence industries in different

ways.

The first logical restriction that leads the analysis of risk disclosure only over a period

of two financial years is that the crisis communication of listed European insurance

companies is comparable from January 2007 to March 2009 when all companies

applied nearly equal accounting requirements related to risk disclosure. In order to

compare risk disclosure and recognizing the absence of external disclosure indices a

situational risk disclosure index has been developed. This index is based on

categories that derive from international accounting requirements. Because

accounting requirements differ International Financial Reporting Standards (IFRS)

have been chosen for this research study as the ones that are most detailed on the

aspect of risk information with the acting IFRS 7 as analyzed in subchapter 3.1.3. To

base the research on this accounting standard reduced the unit of analysis to

insurance groups that applied this standard during the researched period 2007 and

2008. This lead to the exclusion of significant listed international insurance groups

that applied US-GAAP a different set of US national accounting requirements from

this research study. Beside the application of the same set of accounting standards

the structure of companies with similar business risks, similar ownership structure,

similar listing status and application of similar accounting requirements are a

prerequisit for a homogenous group of companies to be analyzed. These limitations

lead to a small sample of thirteen listed European insurance groups as predicted in

the literature by Core (2001) for disclosure research studies.

104

Referring to the second logical restriction is that including other less homogeneous

companies in the unit of analysis would diffuse the causal relationship between the risk disclosure performance and the change in equity ratio as independent variables and the capital market response during a crisis situation as the dependent variable. This

diffusion is indicated in Diamond and Verrecchia (1991) as well as Kim and Verrecchia

(1994). They recommend that a homogenous group of companies from the same industry should be chosen in order to limit inter-industry effects and other hard to

analyse relationships. Instead of banks that were affected differently from the crisis depending on being either an investment bank that sold crisis related financial

instruments or a saving banks that bought them, being in the US or in Europe, having

made acquisitions before or grown by themselves, insurance companies were chosen

as the single industry analyzed in this research study because they were not directly

related to the sub-prime crisis. Beside impairments on their financial instruments with

mortgage portfolio underlying - which also happened to other financial companies - their relationship with investors was affected by a single event. This allows the use of

the event study method to analyse the effect of this single event on a limited number of listed European insurance groups. Due to this indication financial groups other than

insurance business related are not included in this research study and insurance

companies with a similar business structure and product portfolio have been chosen.

The researched financial crisis can be characterized by the drop of fair value in

mortgage portfolio based financial instruments first. That drop has been followed by a

decline in the fair value of other investments that lasted longer than one financial

period. This market development has affected the insurance industry as a pool of

investments by its business model most severely and uniformly whereas the affect on banks depended on their degree of activities in the securitization of mortgage based

financial instrument, their exposure in certain regions of the world and their mix of business activities.

Following this argumentation only a unit of analysis consisting of thirteen listed

European insurance groups has been selected as a sample for this thesis. This size

might restrict the information usefulness of quantitative methods. A generalization from the behaviour of thirteen insurance groups can be misleading when applied to

companies from a different industry, size or groups that disclose accounting

105

information based on country specific accounting standards. This problem is

recognized and solved by using a mixed method approach in this research study. By

using qualitative methods statistical research results based on the existing unit of analysis are triangulazed and improved. In the structured interviews with experts in the field of listed European insurance groups statistical results from quantitative methods of this thesis are verified by experts that do not take part in the preparation of financial statements and can be considered as independent in their opinion.

By focussing on the insurance industry in this research on risk disclosure a lack is

closed that other studies on disclosure like Ewelt (2009) leave when they exclude insurance groups from their unit of analysis because of their unique business model

and the specified accounting regulations. So this research study that is based on a

small unit of analysis follows the recommendation by Fields et al. (2001) to

complement existing large sample studies with research study that analyse

accounting effects with small sample studies. This is done on insurance industry

specific risk disclosure during a crisis situation.

106

Chapter 5

5. Research Methodology The hypotheses will be tested using quantitative and qualitative research tools.

However, by choosing these tools and by setting the research methodology the

researcher brings his personal touch to this study. This influencing factor that is

recognized as the epistemological presuppositions of the researcher by Girod-Seville

and Perret (2001) is analyzed and stated in chapter 5.1. Other factors that influence

the methodology as well as limit the usefulness of the researched results are the

assumptions made for this research study that are stated in chapter 5.2. The research

methodology is detailed in chapter 5.3 and 5.4. The statistical relationship between the

variables is topic of chapter 5.3. In this chapter the search for correlations between the

variables is documented as well as the attempt to find causal relationships between

more than two variables applying regression analysis although the statistical

prerequisites for this statistical method are not met by the small unit of analysis in this

research study. Chapter 5.4. introduces structure interviews with experts as qualitative

methods to triangulize the statistical results as part of the mixed method approach. In

this chapter the experts commented on the theoretical assumptions, the research

structure, the statistical results and the relevance of the research results on future

accounting standards. Chapter 5.5. states confidentiality aspects and ethical

considerations of the research.

The goal of chapter 5 is to provide transparency on the methods to analyse

relationships among variables. The results of the methods described in this chapter

and whether or not the hypothesis as the implementation of the research question can be falsified or not are content of chapter 6 'Results and Discussion'.

107

5.1. Epistemology According to Girod-Sevill and Perret (2001) two epistemological questions should be

answered by the researcher to disclose his personified influence on the study. One is

his understanding of knowledge and the second are the ways the researcher takes to

produce this knowledge. The understanding of knowledge has been influenced by

the literature review in chapter 3. Based on this literature review the researcher found

out that the behaviour of listed companies and capital market participants is based on

economic theories that are constant. Regardless of the statistical insufficiency this

allows the researcher to use statistical tools in order to predict investor's behaviour

on extended risk disclosure according to theories and test it. This approach can be

considered a model that the researcher affiliates himself with according to Kuhn

(1962). By measuring variables and analysing their relationship using statistical tools,

the researcher belief is that objective knowledge will be gathered in this approach. The benefit of this positivist approach is a high degree of confirmation and logical

consistency according to Girod-Seville and Perret (2001).

However subchapter 3.3.2. about psychological factors that influence capital market

participants and is impossible to include as a variable in this research approach

proofed that emotional factors influence the decision making of investors. This

questions the positivist approach which is based on a model that recognizes all influencing variables. Considering this capital market participants are integrated in a

social systems and tend to act within one. This consideration is supported by the

"herding theorem" in subchapter 3.3.2. which describes that individual investors tend

to rely on the decision of other capital market participants and do not decide on a

pure rational basis. Considering this in the research study social factors of investors

need to be recognized considering a critical realism paradigm of the researcher.

The dual paradigm approach in researchers the understanding of knowledge

influences the way this research is pursued. The methodology of this research

recognizes economic theories and develops a model in order to predict the behaviour

of the capital market in a crisis situation. Considering the social context in which financial decisions are taken by individual investors both paradigm should be used to

answer the research question of this study. The positivist and critical realism

paradigm should be combined to a multi paradigm perspective that is applied to this

108

research in a mixed method approach. Based on the literature review the researcher developed the theoretical cause and effect model and created hypotheses. After this

abductive process the hypotheses are tested using an event study and correlation

statistics as a hypothetico-deductive process (Chalmers, 1976). To triangulate these

results interviews with expert users of financial reports are held in order to explain the

results implemented in the mixed method approach. The results of the quantitative

event study are analyzed with a qualitative approach. This mixed method approach in

the research of disclosure was suggested by Beattie et al. (2004). Interviews are held

with experts specialized in the field of European insurance companies from banks or

rating agencies. The combination of these two approaches is defined as

methodological triangulation (Yin, 2003) and allows the researcher to benefit from the

advantages of the quantitative and the qualitative approach (Baumard and Ibert,

2001). The quality of the quantitative event study is impaired by the small sample

size. However the sample size cannot be extended to other Insurance groups, because they differ in size and structure or follow other accounting requirements than

IFRS. Interviews with preparers of group annual financial statements are dismissed

because of their personnel involvement with the company and their inability to freely

disclose the motives of their accounting decisions. The information provided by

experts and analysts can be biased by their need for justification of the financial

judgements they have made about the companies. The combination of both

approaches using statistics and interviews enhances the quality of the research

study.

Another aspect of epistemology is the degree to which the research results can be

generalized and transferred. To achieve this, the validity and reliability of the

research should be considered in the research methodology. Validity represents the

relevance and the precision of research results that correspond with the extent to

which the research results can be generalized according to Drucker-Godard et al. (2001). Although the concept and the measurement of the tools used in this research

study are scientific standard and explained in detail in chapter 4.3. due to the

research subject the results can hardly be generalized to other research situations. This research focuses on a financial crisis situation that has been considered unique in regard to its intransparent risk situation by Baetge et at. (2008). To catch the risk information published by insurance companies in their groups financial statements as

of 2007 and 2008 accounting standard based criteria have been selected and

109

extended by key words special to this financial crisis in analogy of research by Chen

(2006). Although the criteria have been weighten based on preferences used before

the crisis (Armeloh, 1998) the measuring of the disclosure performance based on an

index cannot be generalized to other non-risk related parts of group financial

statements. As a result the specification to the financial crisis 2007/2008 reduces the

ability to generalize the results of this research study.

Another aspect of epistemology is the ability to transfer the results. This is possible

when the study could be repeated by another researcher or at another point in time

with the same results according to Drucker-Godard (2001). To increase this reliability

all procedures to measure and analyse the variables are described in detail in the

subchapter 4.3.1. and 4.3.2. Additionally the checklists filled for all listed European

insurance groups are documented and attached as appendices A and B to this

research. As a further step to increase reliability all assumptions made in the research

methodology are documented in chapter 5.2. All these means and documentation are

provided to the reader for transparency purposes.

Contrasting these reliability improving procedures certain limits exist. A recent study on

risk disclosure by Lenz and Diehm (2010) however illustrates that in this research field

limits must be accepted until technical improvements occur. Lenz and Diehm in their

empirical study state that "selecting the types of risk to be analyzed and choosing the

category the risk is refered to is based on a high degree of subjectivity" (Lenz and

Diehm, 2010, p. 391). However they stress the fact that the setting of a research

method in risk disclosure research is based on subjective decision they point out that

this critical point applies to the majority of empirical risk disclosure studies namely

Kajüter (2001), Bungartz (2003), Vielmeyer (2004) and Filipiuk (2008).

The researcher faced the problem that corporate governance policies prohibit most

rating agencies to discuss their policies and judgment with non-clients. As a result only

one rating agency has been willing to discuss questions with the author of this

dissertation under anonymity. Due to this lack of qualified experts the researcher

extended his search for experts by looking for other knowledgeable users of financial

statements of listed European insurance groups. Those users have been selected based on whether or not they represented a certain group of users of financial

statements of listed European insurance groups. Contacting the experts has been

110

based on personal recommendations by friends and collegues of the author of the

dissertation.

Another weakness in the risk disclosure research is that due to the high manual effort

to read, extract, compare and rate risk information in group financial statements this

analysing process is not performed parallel by another researcher. Lenz and Diehm

(2010) state that the analysis of group financial statements has been performed by

Diehm only. His research results have not been double checked by performing another

analysis of the same group financial statements by another researcher. They admit that this research procedure restrains the objectivity of the research and sets limits to

its reliability. Lenz and Diehm (2010) suggest that computer based methods to analyse

qualitative risk disclosure in group financial statements should be developed and

applied in further research.

Certain steps to increase the reliability of this research study have been done by

adding the results of nearly every research step as appendices to this work. This

enables the reader to trace any risk relevant statement of an insurance group disclosed in their group financial reports to the checklist, to the scoring process and to

the risk disclosure performance. This transparency in the research should increase the

transparency of this research methodology. However restrictions concerning the

subjectivity of the analysis remain. Because they are common in the leading empirical

studies on risk disclosure they are accepted as a given in this research study. The

validity of the research remains low because due to the emphasis of this research

study on a specific financial crisis situation its results can hardly be generalized.

However they can indicate how listed European insurance groups act on other

sensitive accounting information that may be required by future fair value measuring

and disclosing international accounting standards.

5.2. Assumptions To research the relationship between risk disclosing listed European insurance groups

and the capital market assumptions have been set. Several assumptions are set in

regard to variables researched and other assumptions are implied in the research

methods applied in this research study.

111

Following assumptions are made concerning the risk disclosure performance of listed

European insurance groups that is included in this study as a variable.

Although the preference of investors about the type of risk disclosure has been

analyzed by AICPA (1994) and EFRAG (2009) the preference of investors to have or

not to have risk information has to be assumed. Lenz and Diehm (2010p. 387) call it

the "unravelling principal" that investors assume the worst risk situation when they are

not provided with entity-specific risk information. Based on this assumption there is an incentive for the management to extent their risk disclosure in times of crisis. This

leads to implications in the measuring of the risk disclosure performance when it is

assumed that listed European insurance groups are willing to exceed the requirements

set by international accounting standards to meet the expectation of investors taking

an active approach towards risk disclosure. Additionally it implies in the measuring

process that little disclosure of risk information or the statement that such a risk is not

present is regarded higher than no risk information at all.

According to empirical research by Kajüter and Winkler (2003) companies prefer to

disclose unspecific general risk information instead of individualized risk information

specific to the company. On the contrary investors need company specific information

to ensure rational financial decisions. Because quantitative risk disclosure must be

company specific whereas general risk can be described qualitative investors prefer

quantitative risk disclosure. In a crisis situation the assumption of constant information

requirements by investors set in empirical studies like Sengupta (1998) should be

modified. Companies that hold greater risk shall provide more information than

companies with smaller risk. These assumptions are implemented in the measurement

of risk disclosure performance in subchapter 4.3.1.

Following assumptions are made concerning the capital market participants, whose

reaction is analyzed in this study as a variable.

It is theorized that timely, extended and specified disclosure of insurance companies about financial investments such as those that originated the crisis enhances the

company's credibility and causes their investors to respond favourable towards the

company during a crisis (Vater et al., 2008, p. 2606). This company - shareholder

112

relationship has been researched simulating crisis situations and empirically testing an increase in the purchase intent of stakeholders to buy a product of the disclosing

company (Arpan and Roskos-Ewoldsen, 2005, p. 430). Along the same lines previous

research by Diamond and Verrecchia (1991) and Kim and Verrecchia (1994) indicates

that in a crisis situation investors are more likely to hold on to shares of companies that

raise their confidence by a long track record of discretionary disclosure is applied to a

crisis situation. Consequently each company's disclosure performance is measured by

analysing the group financial statements of companies in 2007 and 2008 are analyzed

and to find statistical evidence that those listed European insurance groups with high

performance are held and not sold by their shareholders during the peak of the

financial crisis. Pre crisis experience of investors with the insurance company is a key

predictor of the investor's evaluation of the company during the crisis according to

Arpan and Roskos-Ewoldsen (2005). The assumption that investors that are informed

about risk do not sell the shares they own of that company during a crisis situation is

implied in the quantitative methods that analyse the relationship between risk disclosing and share prices of listed European insurance groups in subchapter 5.3.1.

Following assumptions are made concerning the measuring of the capital market

reaction, the quantitative and qualitative methods used in this research study

Measuring the capital market reaction certain assumptions are made when using the

event study method. The assumption of market efficiency as the basis of the event

study method is provided with short event windows according to McWilliams and

Siegel (1997). However market efficiency is impaired during crisis situations. Therefore

the trade volume of each share per day is considered in order to determine that the

financial market is not inactive or tight so that it still can provide accurate share prices.

With a minimum trade volume defined by Brailsford (1996) each insurance group has

been tested on a sufficient trade volume during the event window. The normal distribution of the researched data in the event study can be assumed, because

McWilliams and McWilliams (2000) assume it for smaller sample sizes than in this

research. Another assumption for the use of the event study is that the event has been

unanticipated. Although a crisis develops the near insolvency of the then world largest

insurance company AIG can be assumed as unanticipated due to the magnitude of this

event. Outliers have been identified using the indirect and direct approach by

113

McWilliams and Siegel (1997). As a result the data of three days within the event

window have been excluded from the research as detailed in subchapter 4.3.2.

The amount of statistical data gathered based on the unit of accounts restricts

quantitative research methods. This restriction is accepted for this research study because the risk disclosure quality can only be compared between groups that apply the same accounting standards, that have a similar business structure and that

account based on a constant set of rules over a period of time. It is assumed that the

resulting population of thirteen listed European insurance group is sufficient to be

statistically analyzed, although certain preconditions on normal distribution or sufficient

sample size might not be met. Due to the application of a mixed method approach the

verification of these insufficient statistical results by expert interviews is assumed to be

enough.

Applying quantitative methods certain assumptions about the type of experts that have

been asked to participate in this research study have been made. Experts in the field of European insurance industry have been asked on the research methodology and

results. Those experts cover the full range of users of groups financial statements

representing the buyer-side, the seller-side, transaction bankers and rating agencies. Preparers of financial statements have not been included in this group, because it is

assumed that they are not sufficiently objective. Being responsible for preparing

consolidated financial statements of listed insurance groups it is assumed that these

managers are not in the position to switch sides and commend from an investor side

on the information that they provided themselves.

5.3. Quantitative Methods The hypotheses are detected and tested by finding the relationship between the

dependent and the independent variables using two different quantitative methods. According to Pallant (2007) these methods explore in a first step the association between pairs of variables by testing their correlation and in a second step predict

capital market reaction as a dependent variable towards the two independent variables

risk disclosure performance and equity ratio by testing multiple regressions. Institutional investors are considered a controlling variable, because all listed European

114

insurance groups have the same high level of institutional ownership in common that they all share the same data (see subchapter 4.3.4. ) that is considered in quantitative

methods.

In the following subchapters statistical test for normality are performed before methods to detect and test the relationship are applied. This is as a preparation for the results that are documented in chapter 6 'Results and Discussions'.

5.3.1. Testing for Normality Testing for normality of the distribution determines the statistical research techniques

that either parametric or non-parametric tests can be applied. The testing for normality has been performed for the dependent variable capital market reaction and the two

independent variables risk disclosure performance and change in modified equity ratio. Three tests are applied to test the distributions of the variables. The 5% Trimmed

Mean removes the top and bottom five per cent of the cases and recalculates a new

mean value. Comparing the original mean with the trimmed mean shows the effect

extreme scores have on the mean. The second test is the analysis of the distribution

concerning skweness and kurtosis. Negative results on skewness present a distribution that leans to the right. Positive skewness results present a distribution that

leans to the left. If a kurtosis result is negative, the distribution will be flat. If the kurtosis

is positive, the distribution will be peaked. This test provides information about the form

of the distribution that might falsify the assumption of normal distribution. The third test

is the Shapiro-Wilk test that is applied to small samples and that is more rigid than the

Kolmogorov-Smirnov test that is considered as additional information (Field, 2009). A

result of more than 0.05 indicates a normal distribution. A result of 0.00 indicates a

violation of the assumption of normality.

5.3.2. Setting of the confidence interval The confidence interval is necessary to decide whether or not hypotheses are true or

not. The Pearson correlation coefficients of all relationships analyzed in this research

study are different from zero. Therefore a correlation between the variable other than

the controlling variable exists. It should be determined if the results of the correlation

analysis are significant enough to falsify or verify the hypotheses.

115

In order to do this all data on correlation results can be converted into data that has a

mean of zero and a standard deviation of one that are called z-scores according to

Field (2009). The z-scores set individual confidence intervals for each Pearson

correlation coefficient. They have been calculated based on an SPSS syntax that is

freely accessable via a companion website published in the statistic book of Andy

Field at www. saaeDub. co. uk/field3e.

The z-scores are used to construct a 95% confidence interval for each of the six

analyzed correlations among the four variables. The variables are risk disclosure

performance 2007, risk disclosure performance 2008, change in modified equity ratio

and cumulative abnormal return that are tested for the following six correlations.

The two variables risk disclosure performance 2007 and change in equity have a lower

boundary of confidence interval of -0.751 and an upper boundary of 0.259.

The two variables risk disclosure performance 2007 and 2008 have a lower boundary

of confidence interval of 0,082 and an upper boundary of 0.867. The two variables

change in modified equity ratio and risk disclosure performance 2008 have a lower

boundary of confidence interval of -0.757 and an upper boundary of 0.245. The two

variables cumulative abnormal return and risk disclosure performance 2007 have a lower boundary of confidence interval of -0.267 and an upper boundary of 0.747. The

two variables change in modified equity ratio and cumulative abnormal return have a lower boundary of confidence interval of -0.141 and an upper boundary of 0.799. The

two variables risk disclosure performance 2008 and cumulative abnormal return have a lower boundary of confidence interval of -0.398 and an upper boundary of 0.674.

When the Pearson correlation coefficients for these variables that are normally distributed are within the boundaries of the confidence interval, then the correlations

can be assumed significant and the hypotheses can be falsified. A confidence interval

for multiple regressions cannot be set, because according to Pallant (2007) the sample

size is too small so that the researcher cannot rely on the results. The result of the

analysis on multiple regression is therefore taken as an indication that is not pursued further statistically but is included in the structured interviews with insurance experts that are utilized to triangulize this result.

Table 1: Confidence intervals

116

Correlation between Hypothesis Lower bound Upper bound Significance of of a 95% of a 95% r confidence confidence interval for r interval for r

Risk disclosure Cumulative H1 -0.267 0.747 0.274 performance abnormal 2007 return Risk disclosure Cumulative H1 -0.398 0.674 0.530 performance in abnormal 2008 return Risk disclosure Change in H2 -0.751 0.259 0.262 performance in modified equity 2007 ratio Risk disclosure Change in H2 -0.757 0.245 0.242 performance modified equity 2008 ratio Change in Cumulative H5 -0.141 0.799 0.131 modified equity abnormal ratio return Risk disclosure Risk disclosure H6 0.082 0.867 0.026 performance in performance in 2008 2007

A confidence interval for hypothesis H3 has not been calculated because all listed European insurance groups have institutional investors. A confidence interval for hypothesis H4 has not been calculated because no correlation is measured. A confidence interval for hypothesis H7 has not been calculated because it is a multivariate analysis that is not used for statistical purposes but only as an indication for the qualitative analysis.

5.3.3. Correlation analysis between two variables Correlation coefficients as the product of correlation analysis provide a numerical indication of the direction and strength of the linear relationship between two variables

according to Pallant (2007). A negative or a positive correlation coefficient indicates the

development of one variable when the other variable increases. The size of the

correlation coefficient indicates the statistical strength of the relationship. In a first step it is tested in this research study whether or not the relationship between two variables

is linear or not. The kind of relationship being linear or not-linear determines the type of

correlation coefficients used in this part of the research study. That is Pearson

correlation coefficient (r) for linear relationships as a parametric test and Spearman's

Rank Order Correlation (rho ) for curvilinear relationship as a non-parametric test

according to Pallant (2007). The results of the correlation coefficient used need an interpretation in order to be facilitated in this research study. The strength of the

relationship between two analyzed variables can be interpreted based on Cohen

(1988, p. 79) that a correlation coefficient of r=0.30 to 0.49 is considered medium

strong and of r=0.50 to 1.0 is considered very strong. However individual confidence

intervals that recognize the distribution of each variable are recommended by Field

117

(2009). Because the statistical sample that is based on the unit of analysis with only thirteen listed European insurance groups is quite small z-scores are calculated in

order to set improved correlation specific confidence interval that are necessary to

verify or falsify the hypotheses from chapter 4.1. Additionally the coefficient of determination R2 is used to explain the amount of variability in one variable that is

shared by the other variable according to Field (2009).

5.3.4. Regression between multiple variables Multiple regression techniques are used to explore the relationship between one dependent and a number of independent variables according to Pallant (2007). Applied

to this research study the relationship between the capital market response as a dependent variable towards the risk disclosure performance and the proxy of change in modified equity ratio is tested using a standard multiple regression. According to

Pallant (2007, p. 147) this approach explains 'how much unique variance in the

dependent variable each of the independent variables explains'.

The variable of institutional investors is indirectly recognized as a control variable because its value is 1 with twelve out of thirteen listed European insurance groups that

are part of the unit of analysis. Lacking data on the thirteenth insurance group is

assumed to be similar to the other groups with a value of 1, too. Multiple Regressions

are performed because they allow 'a more sophisticated exploration of the

interrelationship among (... ) variables' than the correlation analysis according to

Pallant (2007, p. 146). The knowledge gain of this method is that it is able to predict a

particular outcome of a certain combination of variables.. This would support the idea of Baiman and Verrecchia (1996) that an optimal level of risk disclosure exist, even when the referring dependent variable is not cost that has been critizised by Vielmeyer

(2004) but share price stability measured using the cumulative abnormal return as described in subchapter 4.3.2. The variable that is a proxy for the effect the crisis had

on the equity of each listed European insurance group can be controlled for to set this

optimal level. Using a standard multiple regression to achieve this requires certain

conditions that are not completely met in this research study. The small sample size that has been used in this research study in order to maintain a high quality information

level of the risk disclosure variable hinders the application of multiple regression tools.

According to Tabachnick and Fidell (2007) the required sample size for a regression

118

analysis with two independent variables in analogy to this research design would be 66

listed European insurance groups. This number cannot be achieved without comparing financial companies with different business models, different global reach and different

accounting standards they apply. To base this risk disclosing research on sufficiently large group of analysis would lead to unspecific risk disclosure data. The response of

the capital market towards non industry specific risk information diffuse the cause and

effect relationship because the financial industry has been affected by the crisis

situation differently depending on their individual business model and exposure to US-

mortgage based financial instruments.

Although the statistical assumptions for multiple regressions are not met it is continued. The results may be invalid from a statistical point of view. They can be used however

as indications that can be verified or refused in the structured interviews that are

performed with experts in the insurance industry as described in the following chapter.

5.4. Qualitative Methods By applying mixed method research using an explanatory design the shortcomings of

the research methodology especially in the small unit of analysis that restrains the

explanatory use of its statistical results are overcome. Mixed methods according to

Baumard and Ibert (2001) are used to expand the results of the statistical analysis. To

gain an in-depth understanding of the relationship between risk disclosure quality and

the capital market reaction experts interviews are applied. Statistical analysis from the

positivist perspective and expert interviews from the constructivist perspective are

applied in this mixed method research approach. Both methods are sequenced and

combined to maximize the understanding of the relationship in a crisis situation. The

statistical results can be explained by experts (Morse, 1991) and those international

insurance groups that are most relevant in the crisis can be selected for a follow up

analysis (Tashakkori and Teddlie, 1998).

Selecting those international insurance groups that have been analyzed in the interviews is done by clustering them as documented in appendix F and appendix G.

The thirteen insurance groups are clustered according to how they were affected by

the crisis (change in equity ratio) and their risk disclosure performance. Those three international insurance groups that were negatively affected most and disclosed least

119

can be regarded as the most dangerous cluster for the capital market. Those three

insurance groups that were affected least and disclosed most can be regarded as the

most beneficial ones for the capital market. These six international insurance groups

are contrasted and analyzed in structured interview with experts. As experts Managers

of rating agencies as well as bankers and institutional investors, who are not involved

in the preparation of financial statements and with expertise in the insurance sector are

asked for structured interviews. The interviews shall focus on specific quantitative finding that need additional explanation (Creswell and Clark, 2007).

The following research method of structured interviews has been used in this research

study. Five one-on-one telephone interviews have been undertaken during the

research process in the period of time from December 2010 to January 2011. The

participants in these interviews were qualified in the field of listed European insurance

groups by their daily work.

The participants have been chosen according to their business position and their

expertise in the field of listed European insurance groups. They are working at

international banks that recommend shares of insurance companies their clients to

buy, at investment banks that provide transaction services such as acquisitions of

insurance companies and therefore follow them, at funds that buy shares of insurance

companies to put them into their portfolio that they hold for their customers and at

rating agencies that provide clients with their opinion about the financial strength of

listed European insurance groups. The work-places of these participants represent

experts that are using the group financial statements of European insurance groups for

their business activities. They are not included in the preparation of these statements

themselves. The participants selected for interviews represent the buyer-side by funds

managers, the seller-side by bank managers and the rating-side by managers of a

rating agency. Managers working in listed European insurance groups that take part in

the preparation have not been included in the research study. It is assumed that they

are biased in answering questions about financial statements that they could be held

legally responsible for. The participants have not been selected randomly. Due to a low

response rate from emails asking banks and rating agencies for a interview

opportunities a different approach has been chosen. Based on the recommendation by

friends and collegues personalized emails have been sent asking for an interview. This

120

increased the response rate and allowed the researchers to cover buyer-side, seller- side and rating-side as users of consolidated financial statements of European insurance groups with expertise on the researched group of analysis.

The interviews were undertaken in a semi-structured way. This allowed the researcher to broach a series of subject areas defined in advance (Ibert et al., 2001). Recognizing

the limited time participants were able to allocate to these interviews topic and sub- topics as well as questions were sent to them via email at least two days in advance. This interview guide has been used to ensure that the same basic line of inquiry is

pursued with each expert who is interviewed (Patton, 2002). These 'guidelines for a

structured interview' as documented in appendix H have been pre-tested by a banking

expert experienced in the field of European insurance groups who has not been taking

part in the later interview process in order to increase the reliability of the qualitative data (Ibert et al., 2001). The interviews lasted between thirty minutes and one hour. As

the interviews have been scheduled before or between meetings of participants the

structured questions allowed an efficient use of the limited period of time available (Patton, 2002). All participants asked for not being recorded so their answered were

written down by the researcher and later completed. The documentation of these

interviews has been sent to each participant and has been confirmed by each expert. Ethical issues have been recognized in the guidelines that were sent to all participants

ahead of the interview and mentioned verbally at the beginning of each interview

session. As a response the majority of participants opted for anonymity. The

participants told the researcher that corporate governance and internal requirements

prohibit them to provide information or an opinion about the researched listed

European insurance groups.

The topics and questions discussed in the semi-structured interview reflect the theory,

structure and results of the quantitative research part. Using interviews experts were

asked to comment on each part. The questions asked related to four sub-topics. The

three research variables were three sub-topics and the other sub-topic the wider implications of the researched risk disclosure on the accounting process and the future

of accounting standard setting. The sub-topics are the effect on equity ratio, the risk disclosure behaviour, the capital market functionality as the three research variables

121

and the sub-topic of overall questions. Four questions on each sub-topic added up to

sixteen questions overall.

Aside from the sequence of the questions the purpose of the interviews has been to

verify the implied theory, the research methodology, results of the quantitative analysis

and to ask experts about future implications.

The implied theory has been discussed with the experts in three questions that

focused on the theoretical background of the research. The purpose of this approach

was, to gather the opinion of experts on theoretical assumptions between the

relationship of corporate risk disclosure and the capital market that have been used in

research studies but that have been questioned and specified in the face of financial

crisis situations. The first question has been about the ability of the capital market to

integrate all available information in its decision making process as assumed by

Blackwell and Girshik (1954). The second question asked the experts on the predictive

value of disclosed risk information. This question specifies the assumption of Blackwell

and Girshik (1954) to financial crisis situations. As researched by Lenz and Diehm

(2010) it would enable experts to predict the effect a crisis would have on the listed

insurance groups based on its disclosed information ahead of the crisis. The other

theoretical question asked the experts about the relationship between risk disclosure

performance and how the capital market recognized this activity. Although this

relationship has been thoroughly researched by Welker (1995), Sengupta (1998),

Healy et al. (1999), Botosan and- Plumlee (2002) as well as Lambert et al. (2007) it has

been questioned recently whether or not it can be applied to crisis situations by Menz

and Nelles (2009) who could not empirically support this relationship during a crisis

period.

The research methodology has been discussed with experts in six questions. This

research has focused on the insurance industry, on specific insurance groups,

assumed preferences of investors regarding information disclosure, the proxy of equity

ratio as a way to measure the negative effect of the crisis on listed insurance

companies and has used the event study method at a certain point of time assuming it

to be the peak of the financial crisis. The purpose of these questions was to discuss

the research methodology with experts in the business field and not in the academic

122

field where they are either uniformely accepted like the event study method (McWilliams and McWilliams, 2000) or rarely discussed like proxies or investor

preferences. This approach did not work in regard with the event study method that

was unknown to the experts, however it has worked out with the preferences that

supported studies by Al CPA (1994) and EFRAG (2009).

The results of the quantitative research that analyzed the statistical correlation of the

three research variables have been discussed with experts in six questions. The focus

of these questions has been to explain the weak statistical relationship between better

disclosure quality and a better opinion of the capital market of the disclosing company. This has been specified by clustering the researched listed European insurance

groups. The cluster in which companies have been negatively affected by the financial

crisis and did not disclose much information about their equity destroying structured financial instruments like Asset Backed Securities, Commercial Mortgage Backed

Securities, Residential Mortgage Backed Securities and Collateralized Debt

Obligations (Scharpf and Schaber, 2009) have been subject of questioning to experts. In addition to that questions have been asked about the opposite cluster of listed

insurance companies that were less affected by the crisis and provided a lot of information. These questions however could not be answered by two of the three

experts due to corporate governance regulations of their companies. The last question has been about a paradox in the relationship between disclosing companies and the

financial market with that is statistically relevant. The capital market detected those

companies that have been affected worst by the crisis way before they disclosed

information about their risk. This paradox. has been explained by the experts.

The future implication of the research has been the last question to the experts. All

experts related this question to efforts of the accounting standard bodies to introduce

market related fair value measurement to both sides of the balance sheets. The

disclosure of fair values can been considered as sensitive as the disclosure of risk information about structured financial instruments during the financial crisis. With this

analogy the researcher tried to determine a constant preference of listed European

insurance groups not to disclose information they consider sensitive. The purpose of this question is to broaden the research results to future accounting matters and to

other aspects of the communication between listed companies and the capital market.

123

Although the answers of the experts are not specific they provide an idea about the

topics that can used in future research studies.

5.5. Confidentiality and Ethical Considerations To analyse confidentiality and ethical aspects or this research study the data gathering

and the results must be distinguished. As far as the gathering of data infringes

individual and organizational rights, the results do not place a threat towards these

rights.

The research study is based on publicly accessable and private date. Listed European

insurance groups are required by national law or listing requirements of share

exchanges to disclose annual consolidated financial statements. When these

statements are prepared according to International Financial Reporting Standards

(IFRS) they must include risk information according to IFRS 7 in the researched period.

These risk information is publicly accessable to everybody. The analyzed listed

European insurance groups all present this information voluntarily on their corporate

websites. The structuring of this data in categories and scoring them has no effect on

the organization or any individual. The same can be said about the quantitative

analysis using statistical methods to research the relationship between risk disclosure,

change in modified equity ratio and capital market reaction.

However the qualitative methods used to verify, test and question the statistical results

depend on interviews with experts. They are entitled to personnel rights and rights of

the organization they represent. According to Patton (2002) questioning experts

requires an interactive style of communication. This restricts the possibility of experts

to maintain their anonymity by answering prepared questions. The interview of experts

requires a personnel contact between them and the researcher either vis-ä-vis or by

phone. Such an exposure endangers their confidentiality rights and those of the

organization they represent. Ibert et al. (2001) specify that the attitude or behaviour of

experts or their organization may be revealed and that can have consequences for the

organization. The confidentiality rights are endangered of investment banks, funds and

rating agencies that all experts that have been taken part in the interviews belonged.

Investment banks are providing transaction advice to different parties. With statements

in the interviews they might lose their neutrality that is required by their customers. The

124

same applies to rating agencies that sell their advice about the financial situation and

outlook of organizations. The loss of neutrality would threaten their business model. Internal regulations of funds limit each fund manager in his or her external

communication. Third parties should not be enabled to conclude about investment or divestment decisions taken by the funds by analysing external communication.

To consider confidentiality aspects in this research study, the researcher has taken

following steps in preparation of structured interviews with experts. Based on the

recommendation by Patton (2001) the researcher explained the purpose of the

research study to each expert in a letter that has been sent to them at least two days

ahead of the interview date. Another part of this letter which is attached as appendix H

to this research study includes a confidentiality clause. This clause enabled each

expert to opt for anonymity. As part of the introduction at the beginning of each

structured interview it has been mentioned by the researcher that questions can be

skipped by the experts. This right to refuse answers on certain questions has been

used by several experts in the process of the interviews.

Based on the results of the literature review in chapter 3a model of the relationship between risk disclosing insurance groups and the capital market has been developed.

Within this chapter aspects of this model such as the analyzed variables, the methods

used to measure these variables, the qualitative and quantitative research instruments

as well as the unit of analysis have been set and explained. In performing this research

no ethical problems have risen and the confidentiality of experts that have voluntarily

taken part in this research study has been preserved.

In the next chapter the model-relationship is tested. According to Royer and Zarlowski

(2001, p. 113) the testing requires that 'hypotheses have to be formulated in such a

way as to be (... ) falsifiable'. In a second step the hypotheses are tested with

quantitative statistical methods. Those results have been questioned or verified by

experts in the field of the European insurance industry. The result of these structured interviews is the last part of the following chapter.

125

Chapter 6

6. Results and Discussion The relationship between the variables capital market reaction as the dependent

variable, risk disclosure performance and change in equity ratio as the independent

variables and ownership structure as the control variable have been statistically

analyzed. The results of applying the quantitative research tools are presented and discussed in chapter 6.1. In order to verify statistical results, the theory implied and the

research methodology of this thesis, professional users of financial statements have

been interviewed and the implications of the research results discussed with them. The

results of the interviews are presented in chapter 6.1. The main findings are summarized in chapter 7.1.

6.1. Statistical Results and Discussion 6.1.1. Results of the test of normality The results on the tests of normality for the cumulative abnormal return of the

dependent variable, the risk disclosure performance of the two independent variables

and the change in modified equity ratio are presented in the following. The distribution

of the control variable's institutional ownership is not measured because it has the

same value for nearly all listed European insurance groups. The variable risk disclosure performance has been measured based on two checklists that refer to risk information that have been published in 2007 and 2008. Thus the results of two

disclosure performances are presented.

The distribution of the cumulative abnormal return of the dependent variable that

represents the capital market reaction can be considered normal. The trimmed mean

of -10.08 is close to the mean -10.02 and the scores are clustered to the right with a

negative Skewness of -0.061 an flat with a Kurtosis of -1.349. The Kolmogorov-

Smirnov test with 0.134 and the Shapiro-Wilk test of 0.924 are higher than the

significance level of 0.05 according to Field (2009). These results lead to the

assumption of a normal distribution. This also supports the assumption of McWilliams

and McWilliams (2000) that the data of event studies even on small samples can be

assumed to be normally distributed.

126

The distribution of the risk disclosure performance in 2007 of the independent variable can be considered normal. The trimmed mean of 53.518 is close to the mean 53.671

and the scores are clustered to the left with a positive Skewness of 0.202 an slightly peaked with a Kurtosis of 0.075. The Kolmogorov-Smirnov test with 0.147 and the Shapiro-Wilk test of 0.953 are higher than the significance level of 0.05 according to

Field (2009). These results lead to the assumption of a normal distribution.

The distribution of the risk disclosure performance in 2008 of the independent variable

can be considered normal. The trimmed mean of 53.956 is close to the mean 53.677

and the scores are clustered to the right with a negative Skewness of -0.530 and

relatively flat with a Kurtosis of -0.765. The Kolmogorov-Smirnov test with 0.177 and the Shapiro-Wilk test of 0.945 are higher than the significance level of 0.05 according to Field (2009). These results lead to the assumption of a normal distribution.

The distribution of the independent variables' change in modified equity ratio can be

considered normal. The Kolmogorov-Smirnov test with 0.166 and the Shapiro-Wilk test

of 0.967 are higher than the significance level of 0.05 according to Field (2009). These

results lead to the assumption of a normal distribution.

For all dependent and independent variables it can be assumed that their distribution is

normal. Therefore parametric tests have been applied to analyse the relationship between the variables with statistical methods.

Table 2: Test of Normality Mean Trimmed Skewness Kurtosis Kolmogorov- Shapiro

mean Smirnov Wilk Dependent variable: Cumulative -10.02 -10.08 -0.061 -0.075 0.147 0.953 abnormal return Independent variables: Risk disclosure 53.671 53.518 0.202 0.075 0.147 0.953 performance 2007 Risk disclosure 53.677 53.956 -0.530 -0.765 0.177 0.954 performance 2008 Change in 0.324 0.388 -0.244 0.200 0.166 0.967 modified equity ratio

There is no test of normality for the control variable 'institutional ownership' because all listed European insurance groups have significant institutional ownership

127

6.1.2. Correlations between the variables The form chosen to present the correlations between variables is summarized in

accordance with Pallant (2007, p. 133).

The relationship between risk disclosure performance in 2007 and the capital market

reaction measured by the cumulative abnormal returns in a crisis situation has been

investigated using a Pearson product-moment correlation coefficient. Preliminary

analyses were performed to ensure no violation of the assumption of the normality and linearity. There was a medium positive correlation of r=0.333 between the risk disclosure in 2007 and the cumulative abnormal return and a weaker positive

correlation of r=0.196 between risk disclosure in 2008 and the cumulative abnormal

return with n=13. Both correlation coefficients are between the lower and upper boundary of a confidence interval that have been calculated using z-scores for each

statistical distribution. The hypothesis H, that there is no relationship between risk disclosure quality of listed European insurance groups and their fair value in a crisis

situation can be rejected. High risk disclosure quality of insurance groups in their 2007

and 2008 financial statements leads to lower negative cumulated abnormal returns. The capital market appreciates the disclosure of risk information and does not sell their

shares in such companies as fast in a crisis situation. The statistical relationship is

stronger with the information disclosed in the 2007 financial statements before the

crisis peaked than with the 2008 data because this was the information considered by

the capital market during the event period. The result supports the empirical study by

Baek et al. (2002) so that their results can be transferred to the financial crisis situation

of 2007 and 2008. With a coefficient of determination R2 of 0.111 for 2007 and R2 of 0.038 for 2008 the shared variances between the variables are limited.

The relationship between risk disclosure performance in 2007 and 2008 of listed

European insurance groups and the degree they are affected by the financial crisis

situation using the change in their modified equity ratio as a proxy has been

investigated using a Pearson product-moment correlation coefficient. Preliminary

analyses were performed to ensure no violation of the assumption of the normality and linearity. There was a medium negative correlation of r=-0.340 between the risk disclosure 2007 and the change in modified equity ratio and a stronger negative

correlation of r=0.354 between risk disclosure 2008 and the cumulative abnormal

128

return with n=13. Both correlation coefficients are between the lower and upper boundary of the confidence interval that have been calculated using z-scores for each

statistical distribution. The hypothesis H2 that there is no relationship between risk disclosure quality of listed European insurance groups and the change in modified

equity ratio can be falsified. High risk of derivative financial instruments that are based

on mortgage portfolios are associated with lower equity ratio, because the instruments

lost a significant portion of their value during the crisis. This loss is recognized by

reducing the equity of insurance groups. The smaller the equity and thus the higher the

materialised risk from these derivative financial instruments the better the risk disclosure performance. Therefore insurance groups under financial stress try to

improve their disclosure about their financial risks more strongly in their 2008 financial

statements after they experienced the shock and awe of the event as a peak of the

effect from the financial crisis on the insurance business. With a coefficient of determination of a R2 of 0.116 for 2007 and a R2 of 0.125 for 2008 the shared

variances between the variables are limited.

The relationship between risk disclosure performance of listed European insurance

groups and their degree of institutional ownership is not correlated because the

ownership structure of all analyzed insurance groups is measured with the number one

as all groups are owned by more than 10% institutional owners. It was not possible to

find out the ownership structure of Standard Life as one of the listed European

insurance group. Assumed that this insurance group is not owned by institutional

investors and provided with the value of zero in this statistical analysis a correlation

coefficient r =-0.062 is close to zero. Therefore the H3 hypothesis cannot be falsified.

Institutional investors as a significant proportion of shareholders have no statistical influence on the risk disclosure quality of companies. The reason for this is that 12 of the 13 analyzed international insurance groups have significant institutional investors.

The results on the H4 hypothesis are not based on a correlation analysis. The results sums up the cumulative abnormal returns of all thirteen listed European insurance

groups analyzed. The H4 hypothesis that the share prices of listed European insurance

groups do not decline during the analyzed event of the financial crisis cannot be

supported. The average standardized cumulative abnormal return that summarizes the

share price development of all listed European insurance groups in the unit of analysis 129

is -10.1. This means that the share value of each insurance group deviated negatively from the industry index during the period of the event window by 10,1%. It can be

stated that the near insolvency of AIG indeed troubled the shareholders of listed

European insurance groups, let more owners sell their shares than shares bought and therefore leading to lower share prices and a decreased fair value of the insurance

groups.

The relationship between the capital market reaction and the change in modified equity

ratio of listed European insurance groups at a crisis situation has been investigated

using a Pearson product-moment correlation coefficient. Preliminary analyses were

performed to ensure no violation of the assumption of the normality and linearity. There

was a medium high positive correlation of r=0.444 with n=13 between the capital

market reaction represented by the cumulative abnormal returns of each listed

European insurance group and their change in modified equity ratio. The correlation

coefficient is between the lower and upper boundaries of the confidence interval that

has been calculated using z-scores for each statistical distribution. The H5 hypothesis

can be falsified. High risks of derivative financial instruments that are based on

mortgage portfolios are associated with lower equity ratio, because the instruments lost

a significant proportion of their value during the crisis. This loss is recognized by

reducing the equity of insurance groups. The smaller the equity and thus the higher the

materialised risk from these derivative financial instruments, the higher the negative deviation of the share price of the listed European insurance groups from the industry

benchmark. Therefore the lower the risk and the higher the equity ratio the lower the

negative deviation of the share price of the insurance company from the industry

benchmark. The capital market has worked efficiently by reducing the share price of international insurance groups in September 2008 that had high financial risk. This risk

arose from financial instruments based on mortgage portfolios. The risk from these

financial instruments materialized four months later when the financial statements of 2008 of the insurance companies disclosed the decrease in equity caused by the drop

in fair value of these instruments. With a coefficient of determination R2 of 0.197 the

shared variance between the variables is limited.

The relationship between the risk disclosure performance in 2007 and 2008 of listed

European insurance groups has been investigated using a Pearson product-moment

130

correlation coefficient. Preliminary analyses were performed to ensure no violation of the assumption of the normality and linearity. There was a high positive correlation of r=0.606 with n=13 between the risk disclosure performance of both years. The

correlation coefficient is between the lower and upper boundary of the confidence interval that have been calculated using z-scores for each statistical distribution. The

H6 hypothesis can be falsified. Those listed European insurance groups with a high

disclosure performance in their consolidated financial statements 2007 also have a high risk disclosure performance a year later in 2008. Those insurance groups with a low disclosure performance in 2007 showed a low disclosure performance based on their 2008 consolidated financial statements. The impression that the risk disclosure

performance of listed European insurance groups is relatively stable shows that their

willingness to disclose risk information is not situational and is not significantly influenced by the financial crisis situation.

Table 3: Univariate analysis measured by Pearson correlation

Cumulative Change in Risk disclosure Risk disclosure abnormal return modified equity performance 2007 performance 2008

ratio Risk disclosure r=0.333 r--0.340 r=0.606 performance 2007 Risk disclosure r=0.196 r--0.354 r=0.606 ----- erformance 2008

Cumulative ---- r-0.444 r=0.333 r-0.196 abnormal return

6.1.3. Multivariate regression The relationship between the capital market reaction as dependent variable and the

risk disclosure performances 2007 and 2008 as well as the change in modified equity

of listed European insurance groups as independent variables of this research study have been investigated using a Pearson product-moment correlation coefficient

although the sample of listed European insurance groups is statistically too small to

apply such regression analysis according to Pallant (2007). Therefore no confidence interval has been calculated. The statistical results are used as an indication for the

qualitative method applied in this research study. The results of the H7 hypothesis are used to cluster listed European insurance groups according to their relevance for the

capital market. Those that disclose little and bear significant risk positions are frightening to the capital market. Those that disclose a lot and bear little risk are the

stars of the capital market. These results are documented in appendices F and G. The 131

response of the capital market is influenced by a medium correlation of r=0.282 with the change in modified equity ratio and to a high correlation of r=0.584 with the risk disclosure performance of 2007. The response of the capital market is influenced to a

slightly lower degree with a medium correlation of r=0.262 with the change in equity

ratio and to a lower degree with a medium correlation of r=0.292 with the risk disclosure performance of 2008. The H7 hypothesis can be falsified although this result is not based on statistically valid confidence intervals and the necessary sample size. The better risk disclosure performance and a lower risk from derivate financial

instruments that are based on mortgage portfolios resulting in a higher equity ratio, the

less insurance groups suffer from a negative effect on their share price in a crisis

situation. The capital market does not sell the shares of those companies with small financial risk and a good risk disclosure performance in a crisis situation. The capital

market relies more on the risk disclosure performance before the crisis event with a

correlation coefficient of r=0.584. The statistical significance of this multivariable

correlation is higher than the Ho hypothesis because the R2 is high with 0.461 and even the R2coRected of 0.354 is higher than the R2 of the Ho with 0.111 according to Backhaus

et al. (2008). So it is worth using the multiariante regression analysis with the change in modified equity ratio and the risk disclosure performance of 2007 as independent

variables in this research study.

After the quantitative results are presented, qualitative methods are used in the

combined research approach to improve the precision of the statistical measurement

according to Baumard and Ibert (2001).

6.2. Qualitative results and discussion Interviews with users of consolidated financial statements that are experts in the

European insurance industry were performed and documented in appendix I. By using

clusters those insurance groups were selected that should be analyzed by experts from rating agencies. The sample of thirteen international insurance groups was

subdivided into four clusters using their disclosure performance of 2007 and the

change in equity ratio. Two of the clusters each containing three insurance groups with bad disclosure performance and low equity ratios as well as a cluster of three insurance groups with high disclosure performance and high equity ratio in 2007 and 2008 were further analyzed as documented in the appendices F and G.

132

The three insurance groups Allianz, ING and Standard Life are the ones with the

lowest disclosure performance and the highest risk from financial instruments in the

crisis situation measured by the reduction of equity used as a proxy. Can they be

considered the most dangerous investments in the industry?

The three insurance groups MunichRe, AXA and Zurich are the ones with the highest

disclosure performance and the lowest risk from financial instruments in the crisis

situation measured by their reduced equity ratio used as a proxy. Can they be

considered the safest investments in the industry?

The interviews have been performed with five experts in the fields of banking,

investment banking, fund management and rating agencies that cover the full

spectrum of professional users of risk information in group financial statements. The

daily work of all experts is focused on the European insurance industry. The purpose of the interviews was to receive comments about the implied theory, the research

methodology, results of the quantitative analysis and to ask experts about future

implications of the research performed.

6.2.1. Experts on the implied theory Answers on these questions differed among experts. The first question about the

ability of the capital market to include all information available at a given point of time in

the share price was answered by four experts. The question is based on the

assumption mentioned by Blackwell and Girshik (1954) that can be considered the

cornerstone of empirical research on the capital market. Although the experts agreed

on the assumption they voiced criticism that this is an academic assumption. They

pointed out that the business model of listed European insurance groups is far too

complex so that not all information is available to the market and can be processed by

the capital market. Anonymous (2010) asserts that insurance shares suffered unduly

during the crisis. Seel (2011) provides the information that the special regulations in

each European market and entity-specific assumptions on disclosed information make

it almost impossible - especially in the life insurance business - to gather the

necessary information to compare new business ratios among listed European

insurance companies. The information was based on the attempt to create an internal

133

report on European life companies. This questions the capacity of the capital market to

receive understandable and sufficient information in order to make financial decisions.

The second question about the assumptions of a relationship between corporate disclosure and the capital market can be adapted to the financial crisis. As mentioned by Lenz and Diehm (2010) it would be possible to predict the effect of the financial

crisis on the listed insurance groups in September 2008 based on risk information that

was disclosed in the group financial statements 2007. The question was answered by

four experts. Anonymous (2011 a), Anonymous (2011 b) and Anonymous (2010) are

supportive that the effect of the crisis on European listed insurance groups can be

predicted. However Anonymous (2011 a) specifies that information needed to make

such predictions exceeds that which is disclosed in the group financial statements or

disclosed in other ways. Seel (2011) opposes the opinion of the other experts by

stating that the insurance industry is far too complex in order to be able to make

precise predictions about the financial results of European insurance groups.

The third question about the relationship between risk disclosure performance and how

the capital market recognizes this activity has been thoroughly researched by Welker

(1995), Sengupta (1998), Healy et al. (1999), Botosan and Plumlee (2002) as well as Lambert et al. (2007). They empirically supported better financial possibilities on the

capital market for listed companies that exceed requirements and disclose additional

information. The psychological aspect of this relationship gains importance during

crisis situations. Williams and Dolnik (2001) found out that such disclosure behaviour

of companies increases the credibility of these companies in the eye of the investor.

These investors are endangered of panicking during crisis situation in a phenomenon

called "herding" by Hirshleifer and Teoh (2003). Only the trust of investors in the

credibility of companies can reduce such herding effects (Vater et al., 2008). All

experts agreed that timely and extensive disclosure of companies build up trust by

investors. Seel (2011) points out that it is the quality and not the amount of information

that is significant for investors. He defines this quality as being able to compare disclosed information with other benchmarking European insurance groups.

Summarizing the experts verified existing theories about the relationship between

disclosing European insurance groups and the capital market, they support the

134

theoretical ability of the capital market to consider every information available at any

given point of time in its decision making process. However, some experts constrict this opinion when they stress the complexity of the insurance business model, the

country-specific regulations and the entity-specific assumptions. These constraining factors limit the efficiency of the capital market in regard to the share evaluation of listed European insurance groups. The experts question the interdependency between disclosure performance of companies and their share evaluation by stating that the information disclosed must be comparable in order to be useful for investors.

6.2.2. Experts on the research methodology The research methodology has been determined by the kind of industry and the

companies that were analyzed, by the event study method and its timing, by the

proxies to measure the effect of the crisis and the risk disclosure performance by the

companies.

The focus of this research is on the insurance industry. Only insurance companies

were analyzed because they did not initiate the financial crisis but predominantly

experienced side effects of the crisis like illiquid markets for financial instruments and lower market interest rates on corporate and governmental bonds. These side effects

mattered to them because insurance companies are major pools of financial assets. In

order to research this industry it is essential that listed European insurance groups

were affected in a similar way in order to compare their risk disclosure behaviour and the reaction of the capital market on their disclosure. All five experts answered this

question. Their shared opinion was that the insurance industry was hit hard by the

financial crisis. The industry was affected indirectly by the crisis according to Noack

(2010) unlike banks which were hit harder and directly by selling credit default swaps that originated the crisis (Anonymous, 2010). The experts explain this difference

between banks and insurance companies by pointing out that insurance companies have a longer liquidity position that did not force them to sell as many assets as banks

during a time when the market was down (Anonymous, 2011 b). The experts did not

address the problem in this question that several of the researched listed European

insurance groups owned banking subsidiaries. This leads to organizations with a

mixed structure of banks and insurance segments. It disturbs the idea of a homogenous group of analysis that was analyzed in this research study.

135

In addition to the focus on the insurance industry only listed European insurance

groups that record their books according to International Financial Reporting

Standards (IFRS) were included in this research study. The purpose of these

restrictions was to build a comparable and homogeneous group of companies to be

analyzed. The request about the homogeneity of the insurance companies was part of the question: Which European insurance group was affected most negatively by the

crisis? Four experts answered the question. They agreed upon two main differences in

the structure of European insurance groups. First, the exposure to the US-life

insurance market forced these groups to meet financial guaranties of policies which

they sold that could not be achieved with the actual return of their investments which

lead to significant losses during the researched period of 2007 to 2008 according to

Noack (2010) and Anonymous (2011a). Second, by being involved in banking activities

European insurance groups were affected directly by the financial crisis and not only

indirectly like those groups with pure insurance business (Anonymous, 2010).

Anonymous (2010) and Noack (2010) specify that AIG and SwissRe as two

international insurance groups that were involved in initiating the financial crisis by

selling credit default swaps. These two groups were excluded from the sample of this

thesis because they record their books according to US-Generally Accepted

Accounting Standards (US-GAAP) that differ in their requirements to disclose risk

information from the other internationally applied accounting standards IFRS. This

leads to the conclusion that the thirteen European insurance groups that were

analyzed in this research are only partially homogenous. Groups with banking activities

like Allianz and ING could be considered separately or given special attention in the

analysis of their statistical results. However, the thirteen groups are homogenous in the

accounting standards they apply and in the way that they are predominantly indirectly

affected by the financial crisis and did not initiate it by selling credit default swaps.

The event study method was used to analyse the reaction of the capital market

towards risk disclosure at a certain point of time. The method is uniformly accepted in

the academic world to analyse the relationship between events of listed companies

and the capital market reaction (McWilliams and McWilliams, 2000). Only one of the

five experts answered this question. Anonymous (2010) did not respond to the

application of the event study method but pointed out and provided examples, that the

136

financial crisis produced some extraordinary share price reactions. The event study

method contrasts according to Kothari (2001) average market reaction in the past with

extraordinary share price reactions around the chosen event date. Therefore this

scientific method is applicable to the analysis of this financial crisis. What Anonymous

(2010) might suggest is that the analysis should not stop by stating that there are

extraordinary share price reactions but why these reactions occurred. At least this is

the way he explains the share price of MetLife and AEGON in the examples he

provides. Although none of the experts explicitly recommends the event study method,

one of them hints in the direction that research tools should be used to analyse

extraordinary share price reactions. As the event study method is accepted in the

scientific community to be such a tool (McWilliams and McWilliams, 2000) it is

assumed that this part of the research methodology is supported by the experts.

The timing of the event study was set in such a way as to minimize other confounding factors and to purify the relationship between risk disclosure and capital market

reaction. Based on the assumption of Vater et al. (2008) that during the peak of a financial crisis investors fear most and need trust to hold on to their investments the

event study was set on September 14th -, 2008 when AIG faced near insolvency and had to be rescued by governmental aid. This event was chosen because it complies

with the requirement set by Kothari (2001) that it is not clustered in calendar time and

with the requirement by MacKinlay (1997) of being preponderantly unanticipated. All

experts answered the question regarding the timing of the event study method and

alternative timing. Four experts agree that the near insolvency of AIG has been an

unanticipated one-time disaster. This incident left investors shocked and fearful

according to Anonymous (2010) and Anonymous (2011a). Noack (2010) and Anonymous (2010) suggest other dates like the insolvency of Lehman, an investment

bank, Hypo-Real-Estate, a German Pfandbrief-bank or the recapitalization of ING.

However these suggested events lack the same magnitude or are locally restricted to

certain countries. As a result of the expert interviews the chosen event date has been

confirmed. Alternative dates can be considered in future research studies.

Two proxies were used in this research project. One proxy is the change in modified

equity ratio of listed European insurance groups. This quantitative number should

stand for the effect the financial crisis had on the insurance groups. The underlying

137

assumption is that the financial crisis lowered the fair value of interest bearing assets

and equity based assets that insurance companies as pools of investments are holding. The difficulty is that insurance groups can categorize their assets into different

groups according to [AS 39. Depending on the chosen category lower market values

are accounted for in different ways either against loss or against equity. As a result the

effect on the annual results varies but on the equity of companies remains the same. Therefore experts were asked about modified equity ratio being a proxy to represent the effect each European insurance group suffered from the financial crisis. All five

experts answered this question. The experts explicitly agreed that the change in the

modified equity ratios can be used as a proxy for the effects of the financial crisis on insurance groups. Anonymous (2010) specified that banking losses from insurance

groups with banking segments lowered the equity ratio as well. As a result, direct and indirect effects of the financial crisis on European insurance groups are enfolded in the

proxy that is used in this research study. Seel (2011) and Anonymous (2011a)

suggested more specific proxies to use. Seel (2011) suggested to use several results

of European insurance groups in one year such as results from regulatory accounting

required by insurance companies by local regulators, results from rating agencies

concerning these groups and results from the embedded value that life insurance

companies voluntarily disclose. The problem with such an expanded proxy is that not

all results are available and comparable. The accounting requirements of local

regulators may differ and the business model of the analyzed European insurance

groups vary so that the embedded value that measures only the market value of life

insurance segments may not be available for all insurance groups or overrate the

significance for one group. Another suggestion by Anonymous (2011 a) was to extend the period of time in which the proxy is measured. He suggests that the write-downs because of lower market values of investments and the write ups when the market

value of these investments rises again should be measured. The advantage of this

method would be that the impairment policy of each insurance group could be

analyzed. Does the insurance group write investments down during the crisis because

of lower market values or does it wait until the prices come back to the original level?

The disadvantage of this method would be the longer period of time to measure. It is

not clear at what point of time the market values of all investments get to the same level as before the crisis. Another objection of the experts is that depreciation is not by

138

choice and that all insurance groups have to comply with the impairment rules of IAS

32 and IAS 39 so that a level playing field approach exists.

The other proxy used in the research methodology is that the disclosure performance is appreciated by investors. The disclosure requirements of IFRS 7 on risk information

set a standard that all listed European insurance groups are obliged to keep to.

However certain information can either be provided in a quantitative or a qualitative

manner. Based on empirical research by AICPA (1994) and EFRAG (2009) investors

prefer quantitative to qualitative information and entity-specific rather than general information. These preferences were recognized in the scoring method used to

measure the disclosed risk information of the listed European insurance groups in their

annual group financial statements of 2007 and 2008. This was not questioned by the

researcher. However Anonymous (2010) verified the investors' preference for

quantitative and entity-specific information in their decision-making process so that the

proxy concurs with the infrequent research studies performed on this topic.

To summarize the statements of the analysts regarding the research methodology, they agreed that the analyzed industry and the selected companies are a

predominantly homogenous group to be analyzed. Although this group includes

insurance groups with banking segments it was stated by the experts that all have

been indirectly affected by the financial crisis. A direct affect could be ruled out because none of the international insurance groups that sold credit default swaps is

included in the group of analysis. The experts supported the application of the event

study method without explicitly mentioning it and confirmed the set event date as the

peak of the financial crisis from the view of investors in all listed European insurance

groups. The two proxies in this research study used to measure the effect of the crisis

on companies and the effect of disclosed information on investors were confirmed by

all analysts. The chosen quantitative research methodology that is based on a

scholarly review of the existing theories and empirical studies were confirmed by users

of group financial statements and experts in the field of the European insurance

industry.

139

6.2.3. Experts on the results of the statistical results In the quantitative part of this research study the statistical correlation of the three

research variables was analyzed. This revealed that better risk disclosure is generally

appreciated by the capital market. In addition the analysis showed that the disclosure

quality improved during the crisis. In two questions analysts were asked about insurance companies that suffered from the financial crisis and did not reveal their risk

position or were not affected but revealed their risk. In order to generalize the

statements about this phenomenon the listed European insurance groups were

clustered as documented in the clustering in appendices F and G. In the last question

an unexplainable statistical result was presented to the experts that the capital market detected those insurance companies that suffered most from the financial crisis way before they disclosed their risk position in their annual financial statements.

In contrast to empirical research by Lehn et al. (2007) that good corporate governance increases the share value of companies this could not be statistically supported in this

research study assuming that risk disclosure is regarded as a means of corporate

governance. The capitalization of all analyzed listed European insurance groups dropped by 10% during the research period. In order to analyse a correlation between

positive or negative feelings of experts and the risk disclosure performance of insurance groups these experts were asked whether or not they were able to get a

clear picture of insurance groups in the crisis. Three experts answered the question. Anonymous (2011 a) stated that information in the group financial statements have

been insufficient and non-specific. But Noack (2010) generalizes that the timeliness

and extent of risk information is directly connected to his judgement about the value of

an insurance company. By doing so he supports the correlation between the risk disclosure performance of the two researched variables and capital market reaction. He specifies that the necessary information about the insurance companies' risk

position portfolio-based financial instruments was not available in 2007 and 2008 but

has improved since. The impression of improvement in the disclosure about these

crisis affected financial investments was shared by all experts.

The analysts were provided with information about the clustering of researched listed European insurance groups as documented in appendices F and G. They knew that

Allianz, ING and Standard Life lost equity as a result of the crisis and revealed less risk

140

information than their peers. Analysts knew as well that MunichRe lost less equity and

scored higher in its risk disclosure performance than other insurance groups. The

experts were asked to comment on these statistical findings. Four experts commented

on this issue. Seel (2010) and Anonymous (2010) confirmed that Allianz and ING

suffered equity losses and that MunichRe remained stable in its equity basis.

Anonymous (2011b) explains this by stressing that ING had a strong position in life

insurance that suffered during the crisis whereas MunichRe had a stronger position in

the property and casualty insurance that did better. The risk disclosure performance of

clustered companies was evaluated differently by the experts. This difference

epitomized on Allianz. Noack (2010) puts MunichRe and Allianz on the same level by

stating that both pursued an active course in their risk disclosure. However

Anonymous (2010) specifies that Allianz made an attempt to be transparent about their

risk exposure but could have done more, whereas MunichRe was fully transparent

during the financial crisis. He adds that ING disclosed excessive amounts of information but does not confirm that this information was impaired in their use for

investors because they could not be compared with other peers. The experts

predominantly support the statistical results of the clustered insurance groups about

the negative effect of the financial crisis on these groups and their disclosure

performance. Doing this they hinted that the risk disclosure measurement that was

performed in this research is reasonable and complies with the judgement of the

experts that were interviewed.

The experts were asked about the unexplainable phenomenon of the capital market that detects insurance groups with high risk although this risk was not disclosed in their

annual financial statements. Four experts answered this question. The experts agreed that the risk information in the annual financial statements of 2008 came too late to

provide investors with the necessary information during this crisis (Anonymous,

2011 b). Noack (2010) explains how the information was available before the disclosure

of the financial reports. He states that alarmed by ad-hoc announcements analysts

asked the management of all European insurance groups during conference calls and investor days about problems that one group had to announce and asked each

management about their position on the relevant problem. This created transparency

in the financial market and share prices reacted correspondingly according to Noack

(2010). Anonymous (2011b), Seel (2011) and Anonymous (2010) confirm this

141

procedure. Anonymous (2011 b) adds that investors rely more on these sources during

crisis situations. The common opinion of the experts is that during this financial crisis the annual financial statements were not the prime source of information for the capital

market. To cover the time lack until new information is released investors asked the

management about current issues and problems of other peers and demanded a reply. Aside from this question Anonymous (2011 a) states that when analysts do not get a

response from the management on specific questions the analysts expect the worst

and act accordingly. This puts listed European insurance groups into the situation that

according to Anonymous (2011 b) they have to disclose the right information in the right

way so that professional investors understand it. Therefore the phenomenon could be

explained by specific risk information that was disclosed in time and ahead of the

annual financial statements to investors during conference calls and investor days.

Summarizing the response of the experts on the results of the statistical research it can be stated that the statistical relationship between risk disclosure quality and the opinion

of the capital market is supported by the experts. They believe that good disclosure

improves their view and opinion about listed insurance groups. Although the

researched European insurance groups did not disclose much information about their

equity destroying structured financial instruments of this financial crisis like Asset

Backed Securities, Commercial Mortgage Backed Securities, Residential Mortgage

Backed Securities and Collateralized Debt Obligations that are relevant according to

Scharpf and Schaber (2009), their disclosure performance improved since the crisis

according to the experts. The experts confirmed the disclosure quality of the clustered insurance groups measured in this research study and explained that during a crisis

critical information is provided to institutional investors in informal ways such as on investor days or conference calls. This puts private investors at a disadvantage. Not

necessarily concerning the access to this information, because all information at such

events is placed on the internet according to Noack (2010) but because of the

complexity of the insurance business model that needs experience that only

professional investors can rely on (Seel, 2010).

6.2.4. Experts on the future implication of research results The future implication of the research was the last question to the experts. They

confirmed that the disclosure of fair values of insurance contracts is considered in the

142

industry to be as sensitive as the disclosure of risk information about structured

financial instruments during the financial crisis. Therefore, the disclosure of risk

information according to IFRS 7 that was the subject of this research study can be

seen as a process of enhanced disclosure required by international accounting

standards. This process continues with the new IFRS 4 standard on the measurement

of insurance contracts and the exposure draft on IFRS 9 future standards about the

measurement of financial instruments. Their measurement is either based on real

market derived values, on market orientated values or on entity-specific calculated

values. These values will be required not just for the debit but for the credit side of the

balance sheet as well. This step brings disclosure of chances, risks and their

quantification in fair values to the next level. The experts were asked about this new

level of transparency for listed European insurance groups. All experts answered this

question. Two experts welcome the higher transparency that a market oriented

measurement of the total balance sheet would bring (Anonymous, 2011a). It would be

a consistent approach that would be beneficial for investors (Anonymous, 2011 b). Four

experts stressed technical difficulties of such an approach that surfaced during the

financial crisis. Anonymous (2011 a) mentioned that market values did not exist in the

last crisis situation, although pressure mounted on accountants to mark them to market

value. Noack (2010) mentioned that the IASB allowed a new categorization of

investments away from the fair value category at the peak of the crisis for companies

which wanted to conceal drops in the fair values of such investments. Seel (2010) and

Anonymous (2010) mentioned the adaption of estimations that must be used in order

to calculate fair values of investments that are not traded on a market like insurance

liabilities. But aside from the technical problems during the crisis two experts question

the willingness of listed European insurance groups to increase the transparency of

their business to investors. Noack (2010) stresses that insurance groups would only be

willing to disclose bad news when these are of no relevance to the assessment of their

overall performance. This opinion is supported by Anonymous (2011a) who believes

that bad news are only disclosed when the insurance group could afford it. This

assessment of experts brings back the basic assumption of this research study

whether or not managers of listed European insurance groups are willing to disclose

the negative risk positions of their companies in a timely and extensive way under the

pressure of a financial crisis.

143

144

Chapter 7

7. Conclusions and Implications This last chapter summarizes the main findings of the thesis (chapter 7.1. ) and demonstrates their implications (chapter 7.2. ). After presenting the limitations of the

used research methods and the sample size in chapter 7.3. the researcher lists

suggestions to investors, accounting standard setters and regulators of the insurance

industry that derive from the findings of this research work (chapter 7.4. ).

7.1. Main findings The research was performed on the European insurance industry that was affected by

the financial crisis especially by the sharp drop in the fair values of investments it held.

The sample selected to analyse its risk disclosure performance and the response

towards it by the capital market is homogenous regarding its common implementation

of international financial reporting standards by all insurance groups but differs about their degree of banking activities and their geographic exposure of the thirteen listed

European insurance groups analyzed.

The main findings are categorized in how the insurance groups have disclosed their

risk positions during the crisis situation and how the capital market reacted towards this

risk disclosure.

Experts confirmed that it is possible to predict the future value of listed European

insurance groups based on the risk information that they disclosed in their consolidated

annual financial statements. They supported the theory that insurance groups can build up trust that investors have in them during crisis situations by extensive risk disclosure. By doing so the insurance groups hope that investors do not sell the shares during crisis situations despite the uncertainty that prevails during that period of time,

That they do sell is proven by the 10% drop in the market capitalization of the

European insurance industry sample that was analyzed in this recent crisis situation

using the event study method. The degree to which insurance groups disclose their

risk positions was analyzed using quantitative methods. By falsifying the H6 hypothesis

the researcher found out that the risk disclosure performance of the analyzed insurance groups remained relatively stable over the researched period of 2007 to

145

2008. However, individual insurance groups that were threatened to be exposed

improved their risk disclosure performance in their 2008 consolidated financial

statements. This disclosure trait of managements can be statistically supported with

the falsification of the H2 hypothesis. In contrast to former empirical research no

correlation between risk disclosure performance and the ownership structure of listed

European insurance groups could be found. The H3 hypothesis could not be falsified

because nearly all analyzed insurance groups have the same ownership structure with

a significant share of institutional investors.

The extraordinary share price reaction during the recent financial crisis was a result of

the capital market towards the companies' risk disclosure performance because

experts stated that in the interviews. Statistical evidence exists with the falsification of

the H6 hypothesis that the capital market appreciates better risk disclosure

performance of insurance groups. Investors did not sell off their shares in listed

European insurance groups with high risk disclosure performance easily during the

recent crisis situation. However, according to experts the capital market considers all

information available in order to take financial decisions. This market behaviour can be

tracked down statistically by falsifying hypothesis H5 and by expert opinion. The share

prices of those insurance groups that were exposed to fair value drops of mortgage-

based portfolio financial instruments and banking activities suffered most at the peak of

the financial crisis although those insurance groups had not disclosed their risk

positions in their consolidated financial statements. The efficiency of the capital market

in valuating the insurance groups was made possible by direct communication

between professional investors and managers when relevant, company-specific and

timely risk information was disclosed during investor days, conference calls and similar

occasions.

Experts agreed that the disclosure of a risk position of an insurance group is equally

sensitive to the disclosure of their assets' and liabilities' fair values. Therefore the

research results about the behaviour of the insurance groups as well as the reaction of

the capital market can be applied to the disclosure of other sensitive information as

well as to other situations that are not as extreme as the recent financial crisis

situation.

146

7.2. Implications of the findings It is unclear whether or not the research results analyzed for the insurance industry can be applied to other industries. Although the insurance industry can be regarded as a

secretive one that hesitates to disclose sensitive information it would be wrong to

conclude that other industries are more willing to disclose sensitive information.

Experts point out that the business model of the insurance industry is unique and difficult to understand for investors.

Aside from the attitude of managements towards risk disclosure in normal times, the

behaviour of management under pressure changes for those with risk positions in their

books and remains constant for those with a better position in their books. Statistical

methods indicate that the insurance groups exposed to high financial risk and decreased equity ratios as a consequence improve their risk disclosure performance in

their next financial statements. On the other hand the insurance groups without these

kinds of risk and better equity ratios maintained their high level of risk disclosure. As a

proof the cluster of the best risk disclosing insurance groups with limited exposure to

financial risks remained the same in both financial years, whereas the companies

within the cluster of least risk disclosing insurance groups which did however face risk,

changed. The risk disclosing behaviour of managers can be distinguished in eager

managers that hold on to their high level of risk disclosure regardless of a tough

environment and those managers that limit their risk disclosure and might reveal more

risk information only when pressed by professional investors or the public. This

behaviour remains relatively stable over time.

The participants in the capital market demand risk disclosure and want risk information

that suits their needs during crisis situations as soon as possible and as closely tailored to the individual risk situation of the insurance group as possible. Statistical

methods indicate that they get this information during a crisis situation nevertheless because the capital market detected those insurance groups that were affected most

severely even before they disclosed risk information in their financial statements. Experts pointed out that the capital market demands risk information from the

management in a crisis and receives it by unregulated ways of communication. This

additional source of information puts private investors in a position of disadvantage,

147

especially, as experts stress in an industry that is as complex and hard to understand . as the insurance industry. industry.

The results of this dissertation can be used in further academic research. According to

Verrecchia (1986) research in the field of disclosure should try to link corporate disclosure to its economic consequences on the disclosing company and the capital

market. This dissertation has highlighted some behavioural patterns of disclosing

companies and the capital market in connection with the handling of 'sensitive

information'. With fair value accounting and detailed notes requirements companies

are forced to disclose 'sensitive information' that can result in high external cost from

their unintended use by competitors. How companies comply with or try to circumvent

these disclosure requirements about 'sensitive information' can be subject of futures

empirical research studies.

7.3. Limitation of the research The results of this research study can be generalized based on the restrictions that

only listed groups of the insurance industry have been analyzed. In regard to the

willingness of companies to disclose 'sensitive' information the insurance industry can be regarded as a good proxy for companies from other industries because according to Meyer et al. (2004) insurance groups fear that the disclosure of risk can be judged

critically by investors and customers as a sign of weakness in their business model. This enables the researcher to identify traits of managements to retain 'sensitive'

information that should be considered by regulators and standard setters in their

attempt to provide all investors with relevant information.

This thesis considered the capital market reaction in only one short period of time.

Experts confirmed that this period had a significant impact on the insurance industry, however the long-term risk disclosure performance of companies and the reaction of the capital market over a longer period of time are not part of this thesis.

Statistical results that are based on an analysis of data from a limited sample size can be misleading and should only be generalized when confirmed by another research

method. This problem was addressed in this research study by verifying the results via structured interviews with experts. By using two methods to analyse the research

148

question the consistency of findings generated by the one method is validated by the

other (Patton, 2002). However, future research on risk disclosure with an extended sample size is recommended.

7.4. Policy orientated suggestions The starting assumption that additional risk disclosure of listed European insurance

groups is appreciated by the capital market during this current financial crisis situation

has been statistically supported. This result has implications for the management of

listed European insurance groups, for the capital market as well as for regulators and

standard setters.

The risk disclosure efforts of listed European insurance groups stay constant over time

and vary within a crisis situation. Expert interviews indicated that the same insurance

groups are holding up a high level of risk disclosure quality. This has been confirmed

by statistical methods that find a significant correlation between the risk disclosure

performance in 2007 and 2008 of each analyzed insurance group. Therefore

management that stresses the importance of investor relations and extensive risk

disclosure in financial statements have been pursuing this policy for a longer period of

time. Such management behaviour supports theories that only long-term and constant

disclosure of risk-relevant information builds up investors' trust in the company. On the

other hand, managements that limit risk disclosure or at least the quality of disclosed

risk information as indicated in expert interviews do this as a pattern and over a longer

period of time as well.

Regulators might want to empower private investors so that they receive explanations

from the management about their risk position and not merely the data that have been

communicated to professional or institutional investors and posted on the internet

which are understood by professional investors without further explanation but not by

private investors. The need for timely risk information that is specific to the crisis

situation has proven that risk disclosure in annual group financial statements is

insufficient. The lack of qualitative risk disclosure as a separate part of the notes in

quarterly financial statements does not close this risk information gap.

149

Standard setters reacted at the peak of the financial crisis by modifying their

accounting standards. This reaction has been the first of such interferences and can

only be explained by the magnitude of this global financial crisis and the mounting

political pressure on standard setters. However it is unlikely that standard setters will interfere in smaller crisis situations. It is also unlikely that the standard setters will

demand stricter disclosure and more risk information because they modified

accounting standards in this crisis situation by relaxing them. Therefore risk disclosure

requirements will most probably not be adapted to the increased risk information needs

of investors during a crisis. The standard setters rely on the willingness of companies

to disclose more risk information than set by their accounting standards and to disclose

the individualized effects on insurance groups that can hardly be prescribed in their

form by accounting requirements.

How might these results affect the ongoing discussion of full fair value accounting? Fair

values are considered one line risk indicators because all assets and obligations are

rated at market value and sum up the value of the company in a single market-based

number. At least in theory this sensitive information is based on the objectivity of the

financial market. Due to the lack of a functioning market for insurance obligations the

fair values of them must be calculated by the companies themselves using present

value methods. However, these methods require assumptions that must be set by the

management. This leaves the full fair value concept as a risk information that can be

easily influenced by management - an analogy to the risk information about mortgage

portfolio-based financial instruments analyzed in this research study.

As stated in the expert interviews investors would prefer full fair value as a one line risk indicator and a coherent system in contrast to the current mixed method concept of

cost and fair value accounting although they doubt that all insurance groups would

disclose the same degree of risk information or would use the same assumptions that

influence the calculations. There is a high likelihood that preparers of financial

statements would end up as the same two groups that have been analyzed in this

research study. One group would be highly motivated to disclose all assumptions and

methods of their presented fair values. This would be an approach which would build

trust in investors by the transparency it provides. Another group might use present full

fair values but withhold necessary information about the assumptions that have been

150

used to calculate fair values. This group might improve their disclosure when pressed by regulators, accounting standard setters or crisis situations. However, it is unlikely

that such improvements would last because the attitude towards risk disclosure of the

management remains stable over time according to quantitative results in this research

study.

These different management approaches towards risk disclosure cannot be levelled

out by very specific accounting requirements or diligent auditors because experts

confirm that the insurance industry's business model is much too complex to dictate

every calculation assumption that lead to fair market values. The willingness of

managers to exceed risk disclosure requirements is a necessity in order to provide

investors with the information they need to make their investment decisions. 'Rules

alone do not work; what matters is how they are applied' (Anonymous, 2009b, p. 12).

So it is up to investors to back the managers of those companies with good risk

disclosure performances and penalize those with bad ones. Indicated by statistical

results and supported by experts interviewed in this research study, this is what

investors did during the crisis. Therefore the implementation of full fair values might

succeed despite the complexity of its concept when investors' demand pushes

managers into more specific, up-to-date and understandable risk disclosure. Such

pressure seems to be necessary because, as Bagnoli and Watts (2007) pointed out

even before the financial crisis, the likelihood and content of voluntary disclosure

depends on whether good or bad news is communicated. Therefore a constant

pressure of the capital market should be maintained to ensure that all relevant

information - whether good or bad from the standpoint of the companies - is disclosed.

151

REFLECTIVE DIARY

The work on the DBA thesis has been a challenging experience. To find a topic, plan the process, coordinate with supervisors and information providers and to adapt the

plan to diverting situations has changed the way I work.

The DBA programme should lead individuals to create new knowledge, to gain a

systematic understanding of a body of knowledge and to conceptualise and design a

research project that can be tested with applicable research techniques. Because the

DBA programme is focused on professionals and set up as a part time programme that

allows individuals to pursue their career in correspondence with the academic

programme the thesis topic should be embedded in the business surrounding of the

individual. These goals are to be pursued by an individual who is placed in a social

environment with duties as well as demands by his or her spouses, children and friends.

The first problem is to 'think outside the box'. After years of professional work as a

preparer of consolidated financial statements of a European insurance groups it is a

challenge to find a topic that is embedded in your professional work. As the daily

workload is achieved by routine and standard procedures it is different to be ingenious

and find new topics. The search for a topic starts with the search for a question. This

question should not be superficial or too easy to solve. It should be something that has

the potential to last during years of research and something that is not too far fetched

or impossible to solve due to constraints.

During 2007 I invested private money in AIG shares. It looked to me like a good

opportunity to step into the capital market. This investment turned out to be a disaster.

I lost nearly all the money invested and had to ask myself whether or not I would have

been able to prevent that loss if I had read the consolidated financial statements of AIG

more thoroughly in order to detect the risk position that derived from the credit default

swaps of which AIG had so many.

The topic of disclosure has been with me for several years during my career. The

private loss in AIG shares led me to choose it as a topic of my DBA thesis. Consolidated financial statements offer a basis of information that is easy to access

152

and available from every listed company. The financial crisis that started in 2007 at a time when I was searching for a topic offered a point in time that made the disclosure

unique. It enabled me to apply established knowledge about disclosure to a new

situational setting. The result has been a contribution to the professional practice of

accounting from the empirical financial market point of view and from a non-

quantitative disclosure point of view. The uniqueness of the topic has been the

combination of these two points of view that was only possible because of the

extraordinary circumstances through which the financial market and the European

insurance groups were going in the recent financial crisis.

Flexibility in searching for a topic is essential. After having made it halfway through the

literature review and side stepping other variations of the topic, the idea was finalized.

However, a critical check for every idea is the complexity and the data availability. Although the relationship between investors and the communication of European

insurance groups is quite complex assumptions had to be made in order to make

research steps possible. Such assumptions are restrictions to the first idea but they

need to be done and they should focus on the availability of data. Quarterly financial

statements were not available from a lot of listed insurance groups, annual financial

statements however were. They are posted on the internet sites of the companies. Conference calls or other investor interviews were not comparable between companies

and ad-hoc messages about the disclosure of share price- sensitive information were

not uniformly regulated at the locations or countries of listing. Getting deeper and deeper into the topic the research approach needed further refining. After checking the

regulation requirements of annual financial statements a further restriction on international financial reporting standards limited the group of analysis even more. Well

known insurance groups that report according to national accounting requirements that

are not as explicit about risk disclosure had to be excluded from the research study. This included AIG because it reports according to US-GAAP accounting standards. However, at the end of this process a coherent group of listed insurance groups with a

similar business structure had been refined. Unfortunately the number of European

insurance groups that could be analyzed and compared in this research study was down to thirteen. This required a change in my perspective from choosing a

quantitative approach towards a mixed method approach using interviews to verify the

results that have been based on too small a statistical basis. The advantage of the

rigorous restriction on comparable, international financial reporting standards using

153

listed European insurance groups was that the group financial statements of the years

within the financial crisis were available to the researcher and possible to access via

computer. The advantage of this information availability stood in contrast to the

availability of experts that could be interviewed on the statistical results. The

researcher contacted a lot of friends and former co-workers in order to identify experts in the field of European insurance groups. However, it was difficult to contact them in

advance without results of the research and ask whether or not they would sit interviews with the researcher. So the researcher trusted promises of friends that these

experts would be willing to cooperate and be available for the interviews whenever the

researcher wanted to hold them. When the time came and the researcher was willing to discuss his statistical results the experts were either busy or corporate policy did not

allow them to speak directly and so on. Even contacts via friends and co-workers did

not help in contacting them. This restriction on experts willing to take part in interviews

endangered the process of completing the thesis. Again the availability of information

is essential in the process of the thesis and can overthrow the timetable made. So

researchers are encouraged to include some time slots for information gathering and

some reserve time.

In the end it helped to persistently contact the experts, and a few agreed to take part in

the semi-structured interview and to answer my questions. The quality of the content

varied in the interviews. Some experts answered my questions in detail and were

eager to provide additional information that helped my understanding. Other experts did not answer questions but generalized about the topic at best or talked about

neighbouring topics that were not connected to the research question. During the

interviews it was not possible to bring those experts back to my topic and approaches to question them were blocked.

When summarizing the results of the interviews however, it proved to be helpful that

data existed aside from the questions asked. It was possible to connect information

and to bridge data that I was not aware of before. It was necessary to plan research

steps ahead and to discipline oneself in the work process by planning work steps and defining sub-goals. However, every work plan in this thesis needed adaptation and no-

one was fully implemented. A significant degree of flexibility proved essential for the

progress of this work.

154

Flexibility is a key ability not only in this research process but in daily life as well. So

why did it feel so hard to rely on this flexibility? I felt that the solitude of research has

been a blessing and a curse. To work alone on a subject you have picked yourself is a

very rewarding activity. It is a pleasure to see that the work you have done can be

traced back to you. However, there is a certain price you have to pay that comes to

mind when you need flexibility. You cannot discuss the problems you face with

anybody. You do have friends, a spouse and other members of your DBA programme. You can use them to get motivated or to share disappointments. When you hit a limit

and you have to give up your old plan and find a new opportunity that is when you are

really alone. To give up the old plan for a new plan that no one has taken so far - that

is when flexibility hurts.

The support of the university was readily available. Response by the supervisors was fast, helpful in the subject-matter and very supportive. Technical support is sometimes difficult to come by for long-distance students - access to a helpline would be

appreciated. Technical support is not just there when you face technical problems with

your SPSS software and it takes valuable time until a solution is reached.

Another aspect that comes to my mind looking back is the need for persistency. This

character trait seems to be more important than brilliance in order to pursue and finish

the thesis. It helps to feel humble from time to time and not to allow oneself be

discouraged by the inevitable setbacks but to carry on regardless without losing the

flexibility mentioned above. Those who can be persistent should be grateful because

their living and working conditions remain stable and supportive. When you face

difficulties try to be persistent but do not break something worthy in order to continue.

155

REFERENCES

AICPA (1994) Improving Business Reporting -A customer Focus: Meeting the Information Needs of Investors and Creditors, Comprehensive Report of the Special Committee on Financial Reporting (The Jenkins Report), New York, NY: American Institute of Certified Public Accountants

Andersen, T. G. (1996) Return Volatility and Trading Volume: An Information Flow Interpretation of Stochastic Volatility, The Journal of Finance, Vol. 51, Issue 1, p. 169 - 204.

Anonymous (2008a) The invasion of the sovereign-wealth funds, The Economist, Jan. 17th 2008

Anonymous (2009a) The bonds of time, The Economist, Jan. 10th 2009

Anonymous (2009b) Sound and fury, The Economist, Jan. 14th 2009 Anonymous (2009c) The revolution within, The Economist, May. 14th 2009 Anonymous (2010) Semi-structured interview German investment fund on December

27,2010

Anonymous (2011 a) Semi-structured interview major investment bank on January 28, 2011

Anonymous (2011 b) Semi-structured interview major rating agency on January 25, 2011

Armeloh, K-H. (1998) Die Berichterstattung im Anhang, IDW-Verlag, Düsseldorf Arpan L. M, Roskos-Ewoldson, D. R. (2005) Stealing thunder: Analysis of the effects of

proactive disclosure of crisis information, Public Relations Review, Vol. 31, No. 4, p. 425 - 433

Attig, N., Fong, W. M., Gadhoum, Y. (2006) Effects of large shareholding on information asymmetry and share liquidity, Journal of Banking & Finance, Vol. 30, No. 10, p. 2875 - 2892

Backhaus, K., Erichson, B., Plinke, W., Weiber, R. (2008) Multivariate Analysemethoden, ed. 12th, Springer Lehrbuch, Leibzig

Baek, J. S., Kang, J. K., Park, K. S. (2002) Corporate governance and firm value: evidence form the Korean financial crisis, Journal of Financial Economics, Vol. 71, No. 2, p. 265 - 313

Baetge, J., Brembt, T., Brüggemann, P. (2008) Die Mark-to-Model Bewertung des IAS 39 in der Subprime-Krise, Die Wirtschaftsprüfung, Vol. 60, No. 21, p. 1001 - 1010

Bagnoli, M., Watts, S. G. (2007) Financial Reporting and Supplemental Voluntary Disclosure, Journal of Accounting Research, Vol. 45, No. 5., p. 885-913

Bailey, W., Karolyi, G. A., Salva, C. (2006) The Economic Consequences of Increased Disclosure: Evidence from International Cross-listings, Journal of Financial Economics, Vol. 81, No. 1, p. 175 - 213

156

Bailey, W., Karolyi, G. A. (2006) The economic consequences of increased disclosure: Evidence from international cross-listings, Journal of Financial Economics, Vol. 81, No. 1, p. 175 - 213

Baiman, S., Verrecchia, R. E. (1996) The Relation among Capital Market, Financial Disclosure, Production Efficiency and Insider Trading, Journal of Accounting Research, Vol. 34, No. 1., p. 1- 22

Ball, R., Brown, P. (1968) An empirical evaluation of accounting income numbers, Journal of Accounting Research, Vol. 6, No. 2, p. 150 - 173

Balzer, A. (2008) Der beste Geschäftsbericht, Manager Magazin, Vol. 36, No. 10, p. 102-119

Barleu, B., Haddad, J. R. (2003) Fair Value Accounting and the Management of the Firm, Critical Perspectives on Accounting, Vol. 14, No. 4, p. 383 - 415

Barth, M. E. (1994) Fair Value accounting: Evidence from investment securities and the market valuation of banks, The Accounting Review, Vol. 69, No. 1, p. 1- 25

Barth, M. E., Hutton, A. P. (2004) Analysts Earnings Forecast Revisions and the Pricing of Accruals, Review of Accounting Studies, Springer, Den Haag

Barth, M. E., Beaver, W. H., Landsman, W. R. (1992) The market valuation implications of net periodic pension cost components, Journal of Accounting and Economics, Vol. 15, No. 1, p. 27 - 62

Baumard, P., Ibert, J. (2001) What Approach With Which Data?, in: Doing Management Research, Thietart, R-A. (ed. ), Sage, London

Beattie, V., McInnes, B., Fearnley, S. (2004) Corporate financial communication and voluntary disclosure, Accounting Forum, Vol. 28, No. 3, p. 205 - 236

Beatty, A., Chamberlain, S., Magliolo, J. (1996) An empirical analysis of the economic implications of fair value accounting for investment securities, Journal of Accounting and Economics, Vol. 22, No. 1, p. 43 - 77

Beierle, T. C. (2004) The Benefits and Costs of Disclosing Information about Risks: What Do We Know about Right-to-Know?, Risk Analysis, Vol. 24, No. 2, p. 335 - 346

Berger, T. and Gleißner, W. (2007), Risikosituation und Stand des Risikomanagements aus Sicht der Geschäftsberichterstattung, Zeitschrift für Corporate Governance, Heft 2, pp. 62-68.

Besley, S., Brigham, E. (2000) Essentials of Managerial Finance, 12th ed., The Dryden Press, Orlando

Blackwell, D., Girshik, M. A. (1954) Theory of games and statistical decisions, Wiley, New York

Bloomfield, R., Wills, T. J. (2000) Disclosure effects in the laboratory: Liquidity, Depth and the Cost of Capital, The Accounting Review, Vol. 75, No. 1, p. 13 - 41

Botosan, C. A. (1997) Disclosure level and the cost of equity capital, The Accounting Review, Vol. 72, No. 3, p. 323 - 350

Botosan, C. A.; Harris, M. S. (2000) Motivation for a change in disclosure frequency and its consequences: An examination of voluntary quarterly segment disclosures, Journal of Accounting Research, Vol. 38, No. 2, p. 329 - 353

157

Botosan, C. A.; Plumlee, M. A. (2002) A Re-examination of Disclosure Label and the Expected Cost of Equity Capital, Journal of Accounting Research, Vol. 40, No. 1. p. 21-40

Brailsford, T. J. (1996) The Empirical Relationship Between Trading Volume, Returns and Volatility, Accounting and Finance, Vol. 36, No. 1, p. 89 - 111.

Brown, S., Warner, J. (1985) Using Daily Share Returns: The Case of Event Studies, Journal of Financial Economics, Vol. 14, No. 1, p. 3- 31

Brunnermeier, M. K. (2009) Deciphering the Liquidity and Credit Crunch 2007-2008, Journal of Economic Perspectives, Vol. 23, No. 1, p. 77 -100.

Bungartz, 0. (2003) Risk Reporting: Anspruch, Wirklichkeit und Systematik einer umfassenden Risikoberichterstattung deutscher Unternehmen, Verlag Wissenschaft und Praxis, Sternfels

Bushee, B. J. (2001) Do Institutional Investors prefer Near-Term Earnings over Long- Run Value?, Contemporary Accounting Research, Vol. 18, No. 2, p. 207 - 246

Bushee, B. J., Noe, C. (2000) Corporate disclosure practices, institutional investors, and share return volatility, Journal of Accounting Research, Supplement (2000), p. 171-202

Bushman, R. M., Piotroski, J. D. (2006) Financial reporting incentives for conservative accounting: The influence of legal and political institutions, Journal of Accounting and Economics, Vol. 42, No. 1-2, p. 107 - 148

Bushman, R. M., Piotroski, J. D., Smith, A. J. (2004) What Determines Corporate Transparency, Journal of Accounting Research, Vol. 42, No. 2, p. 207-252

Bushman, R. M., Smith, A. J. (2003) Transparency, Financial Accounting Information, and Corporate Governance, Economic Policy Review, April 2003, p. 65-87

Business Insights (2008) The Top Ten Global Insurance Companies, www. reuters. com, accessed on June 2nd 2009

Chalmers, A. F. (1976) What is this Thing Called Science? An Assessment of the Nature and Status of Science and its Methods, University of Queensland Press, St. Lucia

Chamberlain, A. B. S., Magliolo, J. (1996) An empirical analysis of the economic implications of fair value accounting for investment securities, Journal of Accounting and Economics, Vol. 22, No. 1-3, p. 42 - 77

Charreire, S., Furieux, F. (2001) Exploring and Testing, in: Doing Management Research, Thietart, R. A., Sage Publications, Los Angeles

Chen, J. (2007) Corporate governance and the harmonization of Chinese accounting practices with IFRS Practise, Corporate Governance: An international Review, Vol. 15, No. 2, p. 284 - 293

Chen, J. J., Cheng, P. (2007) Corporate Governance and the Harmonization of Chinese Accounting Practices with IFRS Practices, Journal of Corporate Governance, Vol. 15, No. 2, p. 281 - 293

Chen, K. C. W. (2006) Board composition, regulatory regime, and voluntary disclosure, The International Journal of Accounting, Vol. 41, No. 2, p. 290 - 292

158

Cheng, E. C. M., Courtenay, S. M. (2006) Board composition, regulatory regime and voluntary disclosure, The International Journal of Accounting, Vo. 41, No. 2, p. 262 - 289

Choe, H., Kho, B. C., Stulz, R. M. (1997) Do foreign investors destabilize share markets? The Korean experience in 1997, Journal of Financial Economics, Vol. 54, No. 2, p. 227 - 264

Chung, R., Firth, M., Kim, J. B. (2002) Institutional Monitoring and Opportunistic Earnings Management, Journal of Corporate Finance, Vol. 8, No. 1, p. 29 - 48

Cohen, J. W. (1988) Statistical power analysis for the behavioural sciences, 2nd ed., Hillsdale, NJ, Lawrence Erlbaum Associates

Core, J. E. (2001), A review of the empirical disclosure literature: discussion, Journal of Accounting and Economics, Vol. 31, No. 4, p. 441 - 456

Cornett, M. M., Rezaee, Z., Tehranian, H. (1996) An investigation of capital market reactions to pronouncements on fair value accounting, Journal of Accounting and Economics, Vol. 22, No. 2, p. 119 - 154

Corrado, C. J. (1989) A nonparametric test for abnormal security-price performance in event studies, Journal of Financial Economics, Vol. 23, No. 2, p. 385 - 395

Coval, J. D., Jurek, J., Stafford, J. (2009) The Economics of Structured Finance, Journal of Economic Perspectives, Vol. 23, No. 1, p. 3- 25

Cowan, A. R. (1992) Nonparametric Event Study Tests, Review of Quantitative Finance and Accounting, Vol. 2, No. 4, p. 343 - 358

Creswell, J. W., Plano Clark, V. L. (2007) Designing and Conducting Mixed Methods Research, Thousand Oaks, CA, Sage

Cropper, L. C. (1949) Book-Keeping and Accounts, MacDonald and Evans, London

D'Arcy, S. P., Oh, P. (1997) The cascade effect in insurance pricing, Journal of Risk and Insurance, Vol. 64, No. 3, p. 465 - 480

Daske, H.; Gebhardt, G. (2006) International Financial Reporting Standards and Experts'Perceptions of Disclosure Quality, Abacus, Vol. 42, No. 3, p. 461 - 498

Denis, D. K., McConnell, J. J. (2003) International Corporate Governance, Journal of Financial and Quantitative Analysis, Vol. 38, No. 1, p. 1- 36

Depoers, F. (2000) A cost-benefit study of voluntary disclosure: some empirical evidence from French listed companies, The European Accounting Review, Vol. 9, No. 2, p. 245 - 263

Diamond, D. W., Verrecchia, R. E. (1991) Disclosure, liquidity and the cost of capital, Journal of Finance, Vol. 46, No. 4, p. 1325 - 1359

Dillenschneider, R., Hyde, R. (1985) Crisis communication: Planning for the unplanned, Business Horizons, Vol. 28, No. 1, p. 35 - 38

DIRK e. V. (2007) Corporate Perception on Capital Markets, Qualitative Erfolgsfaktoren in der Kapitalkommunikation, DIRK-Deutscher Investor Relations Verband e. V., Institut für Medien- und Kommunikationsmanagement, Universität St. Gallen, 1. Auflage, Oktober 2007

Dobler, M. (2005) Rezension zu Uwe Vielmeyer, Uwe, Risikoorientierte Unternehmenspublizität, Theorie und Empirie, Verlag Peter Lang, Frankfurt am

159

Main, 2004, in: Zeitschrift für betriebswirtschaftliche Forschung, Vol. 57, p. 277 - 279

Dobler, M. (2008) Incentives for risk-reporting -A discretionary disclosure and cheap talk approach, The international Journal of Accounting, Vol. 43, No. 2,184 - 206

Drucker-Godard, C., Ehlinger, S., Grenier, C. (2001) Validity and Reliability, in: Doing Management Research, by: Thietart, R-A, Sage, Los Angeles, p. 196 - 219

Dye, R. A., Sridhar, S. S. (1995) Industry Wide Disclosure Dynamics, Journal of Accounting Research, Vol. 33, p. 157 - 179

EFRAG (2009) The needs of users of financial information, A user survey, www. standardsetter. de/drsc/news, accessed June 2"d 2009

Elliott, B. (2007) Financial Accounting and Reporting, 12th ed., Pearson, New York

Eng, L. L., Mak, Y. T. (2003) Corporate governance and voluntary disclosure, Journal of Accounting and Public Policy, Vol. 22, No. 4, p. 325 - 345

Ericson, R., Doyle, A. (2004) Uncertain Business: Risk, Insurance and the Limits of Knowledge, University of Toronto Press, Toronto

Ernst & Young (2009) IFRS 7 in the banking industry, www. ey. com, accessed on April 7`h 2009.

European Parliament (2002) Regulation No. 1606/2002 of the European Parliament and the Council of 19. July 2002 on the application of international accounting standards, Official Journal of the European Communities, p. L 2431 -L 2434

Ewelt, C., Knauer, T., Sieweke, M. (2009), Mehr = besser? Zur Entwicklung des Berichtsumfangs in der Unternehmenspublizität am Beispiel der risikoorientierten Berichterstattung deutscher Aktiengesellschaften, Zeitschrift für internationale und kapitalmarktorientierte Rechnungslegung, Heft 12, pp. 706-715.

Fama, E., Fischer, L., Jensen, M., Roll, R. (1969) The Adjustment of Share Prices to New Information, International Economic Review, Vol. 10, No. 1, p. 1- 21

Farny, D. (2000) Versicherungsbetriebslehre, ed. 3`d , Karlsruhe,

Verlag VersicherungsWirtschaft

FASB (2001) Improving Business Reporting: Insights into Enhancing Voluntary Disclosures, Steering Committee Report, Business Reporting Research Project, www. fasb. org, accessed on January 14th 2009

Field, A. (2009) Discovering Statistics Using SPSS, ed. 3`d, Sage Publications, London

Fields, T. D., Lys, T. Z., Vincent, L. (2001) Empirical research on accounting choices, Journal of Accounting and Economics, Vol. 31, No. 1-3, p. 255 - 307

Filipiuk, B. (2008), Transparenz der Risikoberichterstattung, Gabler, Wiesbaden. Foresi, S., Hamo, Y., Mei, J. (1998) Interaction in investment among rival Japanese

firms', Japan and the World Economy, Vol. 10, No. 2, p. 393 - 407 Freiling, A., Wehrle, B. (2008) IFRS: Noch mehr Einheitlichkeit bei den

Informationsangaben nötig, Versicherungswirtschaft, Vol. 63, No. 23, p. 1980 - 1982

160

Frino, A., Jarnecic, E., Lepone, A. (2009) An event time study of the price reaction to large retail trades, Quarterly Review of Economics and Finance, Vol. 49, No. 2, p. 617 - 632

German Accounting Standards Board (2001) German Accounting Standard No. 5-20, Risk Reporting by Insurance Enterprises, Stuttgart, Schäffer-Poeschel

German Accounting Standards Board (2008) German Accounting Standards (GAS), Stuttgart, Schäffer-Poeschel

German Accounting Standards Board (2009) Hinweise zum Prognosebericht gemäß DRS 15 - Lageberichterstattung, www. standardsetter. de/drsc/news, accessed June 2"d 2009

Girod-Seville, M., Perret, V. (2001) Epistemological Foundations, in: Doint Management Research, Thietart, R-A. (ed. ), Sage, London

Gleißner, W., Berger, T., Rinne, M., Schmidt, M. (2005), Risikoberichterstattung und Risikoprofile von HDAX-Unternehmen 2000 bis 2003, Der Finanzbetrieb, Heft 5, S. 343 - 352.

Goldfarb, Z. A. (2008) AIG Shares Tumble On Capital Concerns, washingtonpost. com, accessed Saturday, September 13th, 2008

Gonedes, N. J., Dopuch, N. (1974) Capital Market Equilibrium, Information Production and Selecting Accounting Techniques: Theoretical Framework and Review of Empirical Work, Journal of Accounting Research, Vol. 12, No. 3, p. 48 - 129

Gräfer, H., Scheld, G. A. (2005) Grundzüge der Konzernrechnungslegung, Erich Schmidt Verlag, 9th ed., Berlin

Großer, C. (2001) Erfolgsrechnung von Versicherungsunternehmen nach IAS. Gestaltungsempfehlungen unter besonderer Berücksichtigung der Informationsfunktion der Rechnungslegung, Thesis, Sternenfels, Verlag Wissenschaft und Praxis

Harrison, G. L., McKinnon, J. L. (1986) Culture and accounting change: a new perspective on corporate reporting regulation and accounting policy formulation, Accounting Organizations and Society, Vol. 11, No. 3, p. 233 - 252

Hartmann, C. (2006) Internationalisierung der Lageberichtserstattung - Darstellung und Vergleich des deutschen Lageberichts und der US-amerikanischen MD&A, Tectum Verlag, Marburg

Hayn, S., Prasse, S. (2009) Aktuelle Entwicklungen in der internationalen Rechnungslegung, Zeitschrift für kapitalmarktorientierte Rechnungslegung, Vol. 9, No. 2, p. 134 - 138

Healy, P. M., Palepu, K. G. (2001) Information asymmetry, corporate disclosure, and the capital markets: A review of the empirical disclosure literature, Journal of Accounting and Economics, Vol. 31, No. 1-3, p. 405 - 440

Healy, P., Hutton, A., Palepu, K. (1999) Share performance and intermediation changes surrounding sustained increases in disclosure, Contemporary Accounting Research, Vol. 16, p. 485 - 520

Hirshleifer, D., Teoh, S. H. (2003) Herd Behaviour and Cascading in Capital Markets: a Review and Synthesis, European Financial Management, Vol. 9, No. 1, p. 25 - 66

161

Hope, O. -K. (2003) Disclosure Practices, Enforcement of Accounting Standards, and Analysts'Forecast Accuracy: An International Study, Journal of Accounting Research, Vol. 41, No. 2, p. 235 - 272

Hotchkiss, E. S., Strickland, D. (2003) Does Shareholder Composition Matter? Evidence from the Market Reaction to Corporate Earnings, Journal of Finance, Vol. 58, No. 4, p. 1469 - 1498

Hussainey, K., Schleicher, T., Walker, M. (2003) Undertaking large-scale disclosure studies when AIMR-FAF ratings are not available: The case of prices leading earnings, Accounting and Business Review, Vol. 33, No. 4, p. 275 - 294

IASB (1989) Financial Instruments (IAS 39), London

IASB (2001 a) International Financial Reporting Standards, London IASB (2001 b) Framework for the Preparation and Presentation of Financial

statements, London

IASB (2003) Presentation of Financial Standards (IAS 1), London

IASB (2005) Business Combination (IFRS 3), London

IASB (2005) Financial Instruments: Disclosure (IFRS 7), London

IASB (2009a) Improving Disclosure about Financial Instruments (Amendments to IFRS 7), London

IASB (2009b) Exposure Draft of draft guidance on fair value measurement, London

IASB (2009c) Exposure Draft on Management Commentary, London

Ibert, J., Baumard, P., Donada, C., Xuereb, J-M. (2001) Data collection and managing the data source, in: Thietart, R. A. et al. (2001) Doing management research, Sage, London

IDW (2008) Besondere Prüfungsfragen im Kontext der aktuellen Wirtschafts- und Finanzmarktkrise, Versicherungsfachausschuss, IDW Verlag, Düsseldorf

Jain, A. K., Gupta, S. (1987) Some evidence on "Herding" behavior by U. S. Banks', Journal of Money, Credit and Banking, Vol. 19, No. 1, p. 78 - 89

Jensen, M. C., Meckling, W. (1976) Theory of the firm: Managerial behavior, costs and ownership structure, Journal of Financial Economics, Vol. 3, No. 4, p. 305 - 360

Kaiser, K. (2005) Auswirkungen des Bilanzrechtsreformgesetzes auf die zukunftsorientierte Lageberichterstattung, Die Wirtschaftsprüfung, Vol. 57, No. 8, p. 405 - 418

Kajüter, P. (2001) Risikoberichterstattung: Empirische Befunde und der Entwurf des DRS 5, Der Betrieb, Heft 03, pp. 105 -111

Kajüter, P. and Winkler, C. (2003), Die Risikoberichterstattung der DAX 100 Unternehmen im Zeitvergleich, Zeitschrift für internationale und kapitalmarktorientierte Rechnungslegung, Heft 5, pp. 217-228.

Kajüter, P., Winkler, C. (2004) Praxis der Risikoberichterstattung deutscher Konzerne, Die Wirtschaftsprüfung, Vol. 57, No. 6, p. 249 - 261

Keynes, J. M. (1935) The General Theory of Employment, Interest and Money, Brace, New York, Harcout

162

Khanna, T., Palepu, K. G., Srinivasan, S. (2004) Disclosure Practices of Foreign Companies Interacting with U. S. Markets, Journal of Accounting Research, Vol. 42, No. 2, p. 475 - 508

Kieso, D. E., Weygandt, J. J., Warfield, T. D. (2001) Intermediate Accounting, Wiley, New York

Kiethe, K. (2003) Das Recht des Aktionärs auf Auskunft über riskante Geschäfte (Risikovorsorge), Neue Zeitschrift für Gesellschaftsrecht, Vol. 6, No. 9, p. 401 - 416

Kim, 0., Verrecchia, R. (1994) Market liquidity and volume around earnings announcements, Journal of Accounting and Economics, Vol. 17, No. 1, p. 41 - 68

King. A., Lennox, M. (2001) Lean and green: Exploring the spillovers from lean production to environmental performance, Production and Operation Management, Vol. 10, No. 1, p. 1- 13

Kirchhoff, K., Balzer, A., Gazdar, K., Grewe, W., Gunther, E., Habisch, A., Hecht, J. (2007) Das Good Company Ranking 2007, Hamburg

Kirschenheiter, M., Jorgensen, B. N. (2003) Discretionary Risk Disclosure, The Accounting Review, Vol. 78, No. 2, p. 449 - 469

Kothari, S. P. (2001) Capital markets research in accounting, Journal of Accounting and Economics, Vol. 31, No. 1-3, p. 105 - 231

Kötzle, A., Grüning, M. (2009) Unternehmenspublizität aus Sicht der Praxis, Zeitschrift für kapitalmarktorientierte Rechnungslegung, Vol. 9, No. 1, p. 33 - 44

Kraft, M., Nolte, A. (2005) Risikoberichterstattung von Versicherungsunternehmen - Analyse der Risikoberichte im Zeitablauf, Zeitschrift für die gesamte Versicherungswissenschaft, Vol. 94, No. 3, p. 423 - 455

Krüger H., Schneider-Piotrowsky, K. (2008) Ergänzende Lageberichterstattung zum IFRS Konzernabschluss, Empirische Indizien zur Berichterstattung im HDAX, Zeitschrift für internationale Rechnungslegung, Vol. 3, No. 10, p. 471 - 479

Kuhn, T. (1962) The Structure of Scientific Revolutions, Chicago: University of Chicago Press, 1962

Lambert, R. (2001) Contracting theory and accounting, Journal of Accounting and Economics, Vol. 32, No. 1-3, p. 3- 87

Lambert, R., Leuz, C., Verrecchia, R. E. (2007) Accounting information, Disclosure, and Cost of Capital, Journal of Accounting Research, Vol. 45, No. 2, p. 385 - 420

Lang, M. Lundholm, R. J. (2000) Voluntary Disclosure and Equity Offerings: Reducing Information Asymmetries or Hyping the Share, Contemporary Accounting Research, Vol. 17, No. 4, p. 623 - 662

Lang, M., Lins, K., Miller, D. (2003) ADR's, Analysts, and Accuracy: Does Cross Listing in the United States Improve a Firm's Information Environment and Increase Market Value?, Journal of Accounting Research, Vol. 41, No. 2, p. 317 - 345

Lang, M., Lundholm, R. (1993) Cross sectional determinants of analysts ratings of corporate disclosures, Journal of Accounting Research, Vol. 31, No. 2, p. 246 - 271

Lang, M., Lundholm, R. (1996) Corporate disclosure policy and analyst behavior, The Accounting Review, Vol. 71, No. 4, p. 467 - 493

Lang, M., Lundholm, R. J. (1996) Corporate disclosure policy and analyst behaviour, The Accounting Review, Vol. 71, No. 4, p. 467 - 492

163

Lange, K. W. (2001) Risikoberichterstattung nach KonTraG und KapCoRiLiG, DstR, Vol. 39, No. 6, p. 227 - 232

Leftwich, R. (1981) Evidence of the impact of mandatory changes in accounting principals on corporate loan agreements, Journal of Accounting and Economics, Vol. 3, No. 1, p. 3- 36

Lehn, K., Patro, S., Zhao, M. (2007) Governance indexes and valuation: Which causes which?, Journal of Corporate Finance, Vol. 13, No. 5, p. 907 - 928

Lenz, H. and Diehm, J. (2010) Einfluss der Finanz- und Wirtschaftskrise auf die Risikoberichterstattung im SDAX, Zeitschrift für internationale und kapitalmarktorientierte Rechnungslegung, Heft 7-8, pp. 385-396.

Lundholm, R., Myers, L. A. (2002) Bringing the Future Forward: The Effect of Disclosure on the Return-Earnings Relation, Journal of Accounting Research, Vol. 40, No. 3, p. 809 - 839

Luo, S., Courtenay, S. M., Hossain, M. (2006) The effect of voluntary disclosure, ownership structure and proprietary cost on the return-future earnings relation, Pacific Basin Finance Journal, Vol. 14, No. 5, p. 501 - 521

MacKinlay (1997) Event Studies in Economics and Finance, Journal of Economic Literature, Vol. 35, No. 1, pp. 13 - 39

Mayr, G. (1999) Internationalisierung der Konzernrechnungslegung deutscher Versicherungsunternehmen, Wiesbaden

Mbengue, A., Vandangeon-Derumez, J. (1999) Causal Analysis and Modelling, in: Doing Management Research, by: Thietart, R. A., Sage, London

McWilliams, A., Siegel, D. (1997) Event Studies in management research: Theoretical and empirical issues, Academy of Management Journal, Vol. 40, Issue 3, pp. 626 - 655

McWilliams, T. P., McWilliams, V. B. (2000) Another look at theoretical and empirical issues in event study methodology, The Journal of Applied Business Research, Vol. 16, Issue 3, pp. 1 -11.

Meyer, C., Meisenbacher, M. (2004) Bilanzpolitik auf der Basis von IAS/IFRS, insbesondere in Zeiten der Krise, Deutsches Steuerrecht, Vol. 40, No. 13, p. 567 - 572

Miller, G. S. (2002) Earnings Performance and Discretionary Disclosure, Journal of Accounting Research, Vol. 40, No. 1, p. 173 - 204

Mitton, T. (2002) A cross-firm analysis of the impact of corporate governance on the East Asian financial crisis, Journal of Financial Economics, Vol. 64, No. 2, p. 215 - 241

Morse, J. M. (1991) Approaches to qualitative-quantitative methodological trangulation, Nursing Research, Vol. 40, No. 3, p. 120 - 123

Nagar, V., Nanda, D., Wysocki, P. (2003) Discretionary Disclosure and Share-Based Incentives, Journal of Accounting and Economics, Vol. 34, No. 34, p. 283 - 309

Nissim, D., Penman, S. (1999) Ratio Analysis and Equity Valuation, Review of Accounting Studies, Vol. 6, No. 1, p. 109 -154

Noack, T. (2010) Semi-structured interview Bank on December 22,2010

164

Pae, S. (2002) Optional Disclosure Policy in Oligopoly Markets, Journal of Accounting Research, Vol. 40, No. 3, p. 378 - 394

Pallant, J. (2007) SPSS Survival Manual, A Step by Step Guide to Data Analysis using SPSS for Windows, 3rd ed., Open University Press, McGraw-Hill, Berkshire, England

Pape, U., Schlecker, M. (2009) Reaktion von Credit Spreads auf Finanzmarktkrisen am Beispiel der Subprime-Krise und der LTCM-Krise, FinanzBetrieb, Vol. YY, No. 1, p. 38 - 45

Patton, M. Q. (2002) Qualitative evaluation and research methods, 3`d ed., Sage, Thousand Oaks, California

Pellens, B. (1997) Internationale Rechnungslegung, Schäffer-Poeschel, Stuttgart

Penman, S. H. (2007) Financial reporting quality: is fair value a plus or a minus?, Accounting and Business Research, Special Issue: International Accounting Policy Forum, p. 33 - 44

Pincus, M., Wasley, C. (1994) The Incidence of Accounting Changes and Characteristics of Firm Making Accounting Changes, Accounting Horizons, Vol. 8, No. 2, p. 1- 24

Porcelli, A. J., Delgado, R. (2009) Acute Stress Modulates Risk Taking in Financial Decision Making, Psychological Science, Vol. 20, No. 3, p. 278 - 283

PricewaterhouseCoopers (2006) Reporting under the new regime: A survey of 2005 IFRS insurance annual reports, www. pwc. com accessed on July 21st 2008

PricewaterhouseCoopers (2007) Corporate reporting: Is it what investment professionals expect? Corporate reporting survey 2007, www. pwc. com accessed on July 21th 2008

PricewaterhouseCoopers (2007) Seeing risk from the inside, The impact of IFRS 7 and revised IFRS 4 on insurers'financial reporting, www. pwc. com/insurance accessed on July 21st 2008

PricewaterhouseCoopers (2009) Lageberichterstattun%vor dem Hintergrund der Finanzkrise, www. pwc. com accessed on January 14 , 2009

Rahmann, M. Z. (1998) The role of accounting in the East Asian financial crisis: Lessons learned?, Transnational Corporations, Vol. 7, No. 3, p. 1- 52

Reinhart, C. M., Rogoff, K. S. (2009) This time is Different: Eight Centuries of Financial Folly, Princeton University Press, New Jersey

Rejda, G. E. (1998) Principals of risk management and insurance, 6th. ed., Addison- Wesley, Reading, MA

Rippel, M. (2006) Die Risikoberichterstattung von Versicherungsunternehmen nach HGB, IFRS und Solvency ll, Verlag Versicherungswirtschaft, Karlsruhe

Rockel, W., Helten, E., Loy, H., Ott, P. (2005) Versicherungsbilanzen, Rechnungslegung nach HGB, US-GAAP und IFRS, Schäffer-Poeschel Verlag, Stuttgart

Ross, S., Westerfield, R., Jaffe, J. F. (2005) Corporate Finance, 7th ed., Irwin, Toronto Royer, I., Zarlowski, P. (2001) Research Design, in: Thietart, R. -A. (2001) Doing

Management Research, Sage Publication, London

165

Ryngaert, M., Netter, J. (1990) Shareholder Wealth Effects of the 1986 Ohio Antitakeover Law Revisited: Its Real Effects, Journal of Law, Economics and Organizations, Vol. 6, No. 1, p. 253 - 262

Scharpf, P., Schaber, M. (2009) Handbuch Bankbilanz, 3rd. Ed., IDW, Düsseldorf

Scheytt, T., Soin, K., Sahlin-Anersson, K., Power, M. (2006), Special Research Symposium: Organizations and the Management of Risk, Journal of Management Studies, Vol. 43, No. 6, p. 1331-1337

Schleicher, T, Hussainey, K., Walker, M. (2007) Loss Firm's Annual Report Narratives and Share Price Anticipation of Earnings, The British Accounting Review, Vol. 39, No. 2, p. 153 - 171

Schmalenbach, E. (1953) Dynamische Bilanz, 11th ed., Westdeutscher Verlag, Köln Schwab, M. (2010) Prognoseberichterstattung in der Finanzkrise - Unmöglichkeit

gesetzestreuer Rechnungslegung, Zeitschrift für kapitalmarktorientierte Rechnungslegung, Vol. 7, No. 12, p. 652 - 657

Seel, A. (2011) Semi-structured interview Bank on January 24,2010

Sellnow, T. L., Ulmer, R. (1995) Ambiguous argument as advocacy in organizational crisis communication, Argumentation and Advocacy, Vol. 31, No. 2, p. 138 -150

Sengupta, P., (1998), Corporate Disclosure Quality and the Cost of Debt, The Accounting Review, Vol. 73, No. 4, p. 459-474

Shaw, K. W. (2003) Corporate disclosure quality, earnings smoothing, and earnings' timeliness, Journal of Business Research, Vol. 56, No. 12, p. 1043 - 1050

Shleifer, A., Vishny, R. W. (1986) Large Shareholders and Corporate Control, The Journal of Political Economy, Vol. 94, No. 3, p. 461 - 488

Sias, R. W., Starks, L. T. (1997) Return autocorrelation and institutional investors, Journal of Financial Economics, Vol. 46, No. 1, p. 103 - 131

Sinah, N., Watts, J. (2001) Economic Consequences of the declining relevance of financial reports, Journal of Accounting Research, Vol. 39, No. 3, p. 663 - 681

Skinner, D. (1994) Why firms voluntarily disclose bad news, Journal of Accounting Research, Vol. 32, No. 1, p. 38 - 61

Smith, D., Tombs, S. (1995), Corporate Social Responsibility and Crisis Management: the democratic organisation and crisis prevention, Journal of Contingency and Crisis Management, Vol. 3, No. 3, p. 135 - 149

Standard & Poor's (2002) Transparency and Disclosure: Overview of Methodology and Study Results - United States, Standard & Poor's

Surrey, I. (2006) Die Bilanzierung von Versicherungsgeschäften nach IFRS, IDW Verlag, Düsseldorf

Tabachnick, B. G., Fiddel, L. S. (2007) Using multivariate statistics, 5th ed. Boston, Pearson Education

Tashakkori, A., Teddlie, C. (1998) Mixed methodology: Combining qualitative and quantitative approaches, Thousand Oaks, CA, Sage

Vater, H., Meckel, M., Hoffmann, C., Fieseler, C. (2008) Zur Bedeutung qualitativer Erfolgsfaktoren der Kapitalmarktkommunikation für die Unternehmensbewertung und deren Auswirkungen, Der Betrieb, Vol. 61, No. 48, p. 2605 - 2611

166

Verrecchia, R. E. (1983) Discretionary disclosure, Journal of Accounting and Economics, Vol. 5, No. 4, p. 179 - 194

Verrecchia, R. E. (1998) Disclosure and the cost of capital: A discussion, Journal of Accounting and Economics, Vol. 26, No. 1-3, p. 271 - 283

Verrecchia, R. E. (2001) Essays on disclosure, Journal of Accounting and Economics, Vol. 32, No. 1-3, p. 97 - 180

Vielmeyer, U. (2003) Risikoorientierte Unternehmenspublizität, Theorie und Empirie, Peter Lang GmbH, Frankfurt am Main

Wagenhofer, A., Drücker, H. (2007) Die Messung von �Earnings" Qualität, Journal für Betriebswirtschaft, Vol. 57, No. 3-4, p. 263 - 297

Watts, R. L., Zimmerman, J. L. (1978) Towards a Positive Theory of the Determination of Accounting Standards, The Accounting Review, Vol. 53, No. 2, p. 273 - 305

Watts, R. L., Zimmerman, J. L. (1986) Positive Accounting Theory, Englewood Cliffs

Weick, K. E. (1995), Sensemaking in Organizations, Sage, Thousand Oaks, CA

Welker, M. (1995), Disclosure Policy, Information Asymmetry and Liquidity in Equity Markets, Contemporary Accounting Research, Vol. 11, No. 2, p. 801 - 827

Williams, K. D., Dolnik, L. (2001) Revealing the worst first: Stealing thunder as a social influence strategy. In: Forgas, J., Williams, K. D. (Eds. ) Social influence processes: Direct and indirect influences, New York, The Psychology Press, p. 213 - 231

Williamson, O. E. (1975) Markets and hierarchies, analysis and antitrust implications: A study in the economics of internal organization, Free Press, New York

Yin, R. K. (2003) Case Study Research, 3rd ed., Sage, London

Zarowin, P., Gelb, D. S. (2002) Corporate Disclosure and the Informativeness on Share Prices, Review of Accounting Studies, Vol. 7, No. 1, p. 33 - 52

167

Appendices

168

Uu o\

F w

u G1. u

0 U

x N

co .. Z N

cz C)

w n 0 0 N L N

U U U . .

Ü

< . o

0.

9

3 s e

F

2

F t

Ji Yc

U 22o g

(00ý, T: M. I

:

m 8ýo ä

vI'.; 0: °: E

a, -s Qi. i e

ä

h f

-B n ö ö ä x ED m m m

z_=

o

c`¬ J öF

ösQ 4ýis, E ., ¬ ä

nö_c:

:CF

Of

s` - PoýEö

n n f

n

N

n d Pm JÖ

PFC p°ý

m

p 4

°

6

mo.

6r m 4

m 41ý

m

ýs ¬` : - - ý a . °ý ý i; E ý" ä ' M

iý vE' iE . . °

P-äö£FE_ Se a

¬'0¬= _oýE

"{ "

;ý ý

lYS ¬ ¬

°lfs `cEcä! tE Sätý4. '!

c$e

öiä; ä

c". g

" E ýo °L=Oe i°ýE£ii

o : = e °ý5"S FEoQý"ýge)"=°

öE c° is a Eo"E'or 8"r

iEöýi ,°°Yi

g°"t Y`S "F'"fie8 ¬

ýýEBý, a ýý P ýE ö"�=5ý3 -. m 5 e%c; mÜöýEi gip; gE `F :F

;s 'pgE Y'E ; °

"g_: Aröýa; E' äT. ß. 5 `"c

"G eY Ecý. 3 F8PLP26F$° " 'g.!

" xä ", c=So`. ° EöluE $, _£¬"£°ö E t. tlý" Y. :t .

ýý ¬$ýS "° a V. 1 [. E"o:

U. E ö E °S Ya" E"ýaS':

°: 7 i F -'i

ýf E°i°ii Fei o°;

oE. e¬ "Q Y: `w

c -.

M ý¬ýZE>`c""SP"Eý ¬" £"

P "ö

SE E§S. t °.. : g

f62; äE' ;. .ö iP: ä

. Ep örö iö_ö ° SýY"" °° PyF°, "O

öfFE¬o 2F

ß ¬: OQ ewE"o�

.! ý¬i. ýýsleSYwS4

2 : ý`-£Y -¬ E $ <Ee. SE!: Eg : FY! f ö: eg8<E nE £ä: F ä`. YE: ýE:: ¬c Eý£E

- ?1 ° ö

ö

¬; =an em

gran e

gran ö

ý g ° cm

4ssg . 1R ; rE w 3

i ' 9ö'° R°öö öÖÖ ° ö' äPöe ý-öe o

/ N3ý

n äQw

S°

om

Ö 'xÖ iýä m` : ýöm

ä 'eýä

mm

: m n öm i; äm :; öm

n n n - m m m

X X m

n rv g m

ý ý

X m

°

ö£ ro X n E ö ä E9ý

iPý=o 0 N Nm NN N r-fig Z[_":

¢- t7°n

a� OA rc ý

°c SSS9 ¢.. LLS

ag wo s

.8 g e

rf" ! c; ` °ö ý

uz c

E0`{Xý p öö"3 !ý

ö° i5S °EC

i tS ° ' ` EEe l E. 2 gE

(. Z. -

5E3 ýöi¬

gý`"Eý

QEc`8ý y

¬gE lii

t E s 4

f ö ° $ c g Y

E : I I E y Q [ Ö

-

8 ¢

e

7 2 E x2 p

Ü

1 1

l

1

i

1

1

3

d

C 4

E i

aaaa `= a- aäa

C) n r-i

u cs E

w

V

a)

0 Uv

. fq

10

. I. bo

C) E

10- N 0 0 N

u a) U

a)

: He

S t£ 3" ° nc i¬, SF"

iE cö

iý F ' `

i

t-* inSSf.. e3 ITZ

¬ -, Fü 5: s"E fý

8L'F''v Sc nýöF; a3

" ý3ýi '

c! v5..

ö jai: °Cýý , °ý 24

" it öý. a s ö3

°w. ý_ "5 " 'Eias E ¬ T

E ýöý'°"e

-

4 E°'vft" ° S fi'[Sif

aý f

. i°itci 3£. °S 3ýý °: bgftS

! F 'S[g i:

5 !p p! b"ý" ¬ ¬E I ß

. ý Q ""ä. ""S S= E " '°' O (sc° y2 nY E ` 'F"`

. ¬a,: äysF °. >ý F" ý'io": {

c: ü°`n"i "S° °

E "" L5 ö"c i cý ' E

T ! S. E¬!

8 ü- !

ýY ä ýu L _

[S Ö. 5 S ° ¬6° ¬s 3 . S :; .. e

f. =fi tä <<¬; ýa" i ý° ý " °3ä äp:

"2o 5 ', 3,. =gE ¬e ýv i

"ý¬: ¬äPrý -. £ä5 Via. =[>'! ¬E ýg 3i3" ý C" S oa % ', S =YC ° Qi

y ' Ötiýg ý5ý °g

sn ö sa .c

P Fi" " i. "` of o _c E<`ý i ö

Y'ýýccSýe :: a° f c ý°`° S7a ; ° 1 nZI

° c

E :

ý° :T .yi °Ea e " - a ý £_c

-"` E '

!ý : oý _S "

F. "°" £ea a" ""v° "

?F CSC ;

it F O Söci.

g n f"ýe ;`

- E " ä9S QSE1'£2

ý .°

ýö

c ,: Lie+i ¬ ` ý. Lc ° Tb Fý

c `

e9 : ei "° °i ý : gip

¬ F "ýb

äa Sp '¬

- "£ - ýiw E"ý <ýi.. _ .E

' e `F: £fiFücFe` : ¬ "s° °ö oýýLÖ ö: ¬: :: öýäý' " '=L :: ' I b

¬ { Q` a_PlO ä§ý Oo zý i &'ä ýF F Sie S° ý. °, "

q" ö. £ 4jE" £Ed S S ö F t

ýSE" . 55C

öiF "ci _ ccr3i

eS e-E cL

`ý F - SY °ný: ülýý `ýý6: ta i "e

"°£si S¬ý° v EE c FöýüýýSýý c'! °c£;. _=gep f'"F=öý FýS S

ýi öý ýS säpc°¬iP FF

ý

. 1 . 4sýSCLi rBE

gG °

=S Dý i °

; Se iý 8 p°

a ö _ :

"° i. ý. iv: ýi_ S . ýZ

P3_ J

5" e `F +b OC° 5 v-

i ° .

°lES iý¬Lfo ý"

2 .

: Eii: fý ýGlai

ý

9- Y(iiäa Fcc äicý `{°ß ä. E

°c E Eý SSL °¬

Ö, ýÖC¬bba,; :a E '. Vi 'ti

'j " 1, U E

i eýt": =3nW

4 5 ° F.

*LR6. <,

"' ° °¬i`¬c°jE üzr :: " 5° "5 vLaSi" OF¬ý<:

s8

n £.. ý .

cS týScý i: ýö2

=R£ýý" e" :e

F": o LLS TQ -_ - SLS ;

E SLB-: j:

n ü ' '

Ic 9ýE °:

¬ia =E

:L X5 t:: ° .9 .. y o ý .1 SF " , e

< fi °

il

ý °'S

C. I

° S ý"

F '

öEo iE ý! öBe°>¬° " 'y :; ¬

fo°¬ 3g¬: o^; 7,

¬; [" p'='e

EE ý°! e,!

, ö i [ifis "` ££ .

°s` f F°i e

N " !

Eý:. SF s . . !!

Eý E ` t ý°

' E:

° t , i±o; iäi OM . 12 k ! y x ; ý- d f 5as5 SaES o: <i" öE. B¬. ýSS: E: ° EýSELnäc. Sly oS t_E. n kö

: Sg t Sf I. EE . 's j S 1 .ß Zýi$ 22!: Ei5iC6ös

ö csi

ý. S ßäi m° - n

üFÜ ä° 4

BäL m° ý` ä-c°aw

ä S° : 3o°md - oäri : 3ýmd ari : 3ömý

3 ä ö

a

5

r a

äý

5 _

o L

ö: B _ gzý Sze E ssa f_ f ýF

ä o L

g

c e

3 C

r-q L-1

C) L) cz

0 w

P-4

b0

cs CL)

w

N O O N

iC

a

7

o O O

N A

N m

m iö i

m ri

m

mm d°m

Y

b

ö; £d eFý $S i Eö ü °CFýä$"

e o[ýiäax.. q af! ] ! °Y ä

f °

* ý' 9ý ]

. uý7»Y5ewöw Öägöo''°° oit°iý öý y"c Ym 6°

! '° a : Wü= ofise v

äc

ao C4S r°ý°EOLö

°

¬ý:

c °ý

ý5 r@ Eö3äoe F{ 4y£ e2"E}'°Y'U°tma üF

" -°3`+ý `oitý°ý £"]ý"ö! ý

:; pY ö UPS S =FOn °

fig-¬

$ 2 "ooE` öt3 ý°

1t ° £ET `3 ö ýLL ýY aZc . ýýýýýM xf

ýýý "ýý ý- Pý . öýdý "9°"° ý ýýtý

°g r e ° =: g S R ̀ s S°°w: ýx8 ýr ; 2a Ste ': : ý'"" ýa ýP°°`öv . . t e8 T ° ýý¬EW -311r ýe . ä' °£Tp' ¬ n . ýs ýf6. iýg¬Fde

dE LNsýx

tlOZ: müB ! `° E°wEn°ü oi "

u ý`

° ; Yb °ý¬° Y°

-ý[ � F . nßä ! EE äö üi, Eü¬Cüý-c '-«ev "g $S e. c@" p°dý Fz

£ £C c F$

F _ä=ýFÖCF

FäÜv6 ä Eýr E

. e gs-; ýF°eor ýý co -

': gig¬ý° sö m °v]ý vä c° 5uü: °

'°Ecäýý ae07c oai: Co ° ýöCý«rv

- ýI -ü° ätl_'. ý+ 'ü `

n c

_ EswS ° e NN .4 S "üý q °C<°u_ eE äwNe°

wä r°iöýö2uýeý°"m zw oc-Eýx° _ °ki_° E'E .F £°Sv -ü Sý°oc ' u c

ä_i qF 'x'ü ý' e q

°y"c"ECi, «uF 8 " no oc ý

6E9ý S5 '. 5 U' ew sFö c t., c! "°ü

osö_ vc Pö..: ¬$S

q "ý ä]eeu. °O . nwy °a ¬ý s °6""$ý °8

° ý " x öa

¬ ; - X . 0! °5£ «"_aöE 'c ýF aý. S ä5 a C. a

: ä` ` ° ! o e {yo S ý ]¬W: £6°Ec°°F s7< ý

_ ýt

r__ö " aa4ý Eö ý rä

-ý°: W `EiýýE S°, ° ° ýczý" , F~ 2P ý

_öE ö5 :u

ät °f öy"4iS o: ý°e' . ý

a

ögä y Se

ý ° : c iý. LS i b NT ö`YC£ m yýaýt o"

Sýc x ETB a: ýýý, cý

: °p8 " = Ce o ;. ö! 59 gaors° ý=ro 1e s ]wý"m°of

Kä. Y C t°2 ý m E

ý 2 Q " o (' eX 3 C E

y: tö't: e$, °'ý öcv gypW=vV a°°ý' ' j:

t£Nigi -V., G

° . 188 o

E " 5

. öIFF °'pq.: : 'Z_ "cöäö"e6:

S : S° °-n qme4 °

" £ ! ° " O ° "

°

¬wi

yö !E G] ý°. f 0

Wp "m yyt7 Vy°

Laiö°q SES cäe EF oU $ t

gy -° xU f. n öW ýou F t-Ö L °C ý. w ä ä£cöe

s°Eiö! C£lEä c3 ýSäia ° v ekoL a äý

oý! SE5 iEötýei cüöETnüYEE < S°niuöiF£+a o Fü3£u

qcm o

Em m ci ccm Fm cn

rv n rv OR

ýYý ý msý ýdsý cs msä mss YSo 9'! W m nüm nUm Öm CnUm üYm °Ym

{ g

22 o

ö öX n n

K t V

ä a ry ° N

n n ö n ä

CW m S 0

m° U

m 'm U

7 2y

° 3

'' ý4 ý

=m Sü m

"L * c5 c° -EE-. Tg

äEö E 29 co

m m

'ý5 zs`ö°cm" g'3c maü EE OSE

. 2111 ý48 E e"m9ý°s -E ýöEý m

Uä. ß N°aSF ' Ut°cn eö

c E 3 ýS < ö E c

= r E ü ö "

Z. S C

K W t

LL

ä ýi

ÜVp

N

N

cc:

ý¢

c

Y t

I

s 10. Nyi

J O O ýt N 4J

Lp

Nei; g

y ! ýE

U ,M

aý ý-ýE ä - Vi=i

d U ýýo mähe

aE p Q i3E,

. Qiýi ä

; fý ý 6 0 ö 0 ö 0

mi)

6

24

: fy F r¬ 8

S3f`s

iN V_öF

-c¬ d

ÖE

p3

ö 'of c. ý d nn to e o

° `

O° ` E ;

, ýE' 3ö yý _ i Fý

ö ¬ý'ýSEE Ü ¬

2 [.. . £, c£ °

e sLý EE:

s ýE_ý T öw

Y3 EüIEýO n

äzý5: 8n ä BS : ýP

"

ý a�bbg¬

c "fiEPý E c s

Fa

y

E'ýgL: ý . f" 41

T-C.

Oýýfö °i ;n 'ge; ý

°ü z q! 5ý_ 2

lýz öE; °` _ =L ö. °f i"ýnc L ý. °

ro p sS =P E $ ei Ei ýi "ýF' of no"

ý

n e _ ;

ýe; a5:; FE5l; t, t E ý E:. EýL"ý:

i ý ý :°

_ `E° c y Ys

P<o. ¬"e; °' P "" ý =

f ýL. 5 ä: F9a$= g gý ° it 9g

PEiäSitlý? ö "c

_t ný'L7 " ö ýE°ýrc tý: ¬

I üS: ý ýL iäýäÖ-iic'° aýý

j. . _ ý 7 "LSE: "i'_8=g .

E °E. ä: 2

cý° ö=: '¬

. ýt i gi.

t :: öäi; ý: `> EB:

p ý °° ,F "" °

cg"r: ýeý0°sg 735. ¬

c- °b- ". ö 'Er ..

f .° ! ., tý ýý¬P En6Z

fiý 5ýä 2gE !g

i °ý 2,61 11

p - " ctü: Z

Yins= , e¬8" - "gu 4 . .e

s g_ g8?. - `3 Es `. fn 'ä

¬ý¬ : i. g°EB =-. a

öe L !.! 2° ýgý L; . p. äf,

-! ýý ý . , E: ps

°ýE °- .

ýf ;

er: g $

c S. tdPU °bS rY °f"'-" äe { : °

" ý< . "eioty-n". ö f -a"

`: ' t

, `0' 5FF: ý"'E: E. "

j .o "¬S ! ý. [ c E- c ýi P

C " -f

i( ¬i c"CBfäc Y ¬

ooi: o" ." e=E

äS eý°äýii=°°- { - et3iE C_ Sn

. r

fin

ALL L" : t6. ß, °°""

its!

-c SiLF ä'E`"E f"r. "

5. ö. ,-

!Y3 ii ý , '° °

, , ¬ ;

E f E C ý

eö§ ýp i"° __S' t E3F

L ° ' 3ý c' y ° i°

2E ¬°°S-°: ö !ý S. ! fir 4i pYE .. E : g: c, c

E; -. Sf ° t

°" : F-

E t4' ° o K

, 8.

S E s ý

tSö'i: c"E=P: 3iE' °. -ýE¬zü¬Sýýc° YEýyiY öä gc S -° °'? ':: <°"¬ErfeR .. ¬ jýE

E. ° ýg1 : bn yY";:

ä

° ° " >i.

": 31

fi ` ý °' "

t 3ý5°.. tM I E°_-"ý ý " "n

° =sý

ü"n tf0° i`sn fteE. g¬£p° fý

= - ,° äe.

Ei ; 5-ß'O -. o

F a

Qg H. ý° ' '< "b i s - ;. ýg ; ýý, =t M: a

: L" :Y

c E oL.

eYafi q ýr"c LL 'L °

qi i ö

ea t^ s: Psý E v_ d. r i g

9bi L:. ¬ = fi

. o"a - , "_ ;! =°a.;

.;. Eg°8. " cY L E ö ' . n f

E ' fi ag: 5- g E° Y " ö

"° _'- °E " P E ' .[ :E °_ E

, ne"_ " äe° ° . bl aýca: 5 f. ' ý O S cy ! Y

"E EFý2 EEýö " S `Ee )§P° e

¬ f ý

. ii ä ü cß o iL cc 2

L ° c yx = c,

, n- ü "=.. 'q ¬EaQ E

qZ L 'E'. Pp

'

_'ü b°= "$ S

ýi

ý`-iýiiEvi" ä"ü ýü` liü: i, +BEe!

y y fEi iSEii SiFF (C ÜÖ üýf öeOFFi m S ýi

.L SSüýi i ý

pÖ Dö ý ö t° 'pN

t° S[ N Sý I S g S ö t° I, tö r

" Sao$ i_ "E öpL°

"Äýö ö ; i ° i atp t

iI 3yp4

OR 9G

° .6 i39D1 ý3°m i)"m m

i901 °iD

ý ý p ü Ö Ö D

Ö ýC

°`p L "_ _" Ö

r ý FiY

A n nS'

° o L O

so.

-' ys ý0 5i

ý°

F

ü ü Ö : ýý ' =

;j $E

iiý c_ E°

" c

r b °" e ie

p ý EöKä ,

Ö. rF

c SE ; gý i0 g

E ry ESö =is

c Eg ii

ýc

ä: ; 5E E

g

E8FýBS ÖS SFi öi E

f r

p E ý 3 {

n t ° f $ g t E

c f 3 5

s z

F ; 8

U ° ä $ c

Ö C E

i

a

ý E

S

E -f °i

ý 2ý

¢E

C 7 1

.

° m' L Ö 6i SF i ö0 %( äi

ö s i 5)

a <1

M N-

it cu

C) U

i-ý GJ

it

b

x

bo

0 N

v V

.. o

1v

eo

r

18400'_ W1c

°E$ E2 --4

tI " 3

N

g

<

2 o 5ý °Fi

O .. c

'¬ q S`° � .

'qty oG °e

Rä

gilli" 3ü, ß °°'ýcCq a '¬ °° ae

�1zaK s f1: 1 _ rn-

. -1

Eö'ý-BEY

E °L

.°d S. is Eoä a° oYY3S9oLöö i° YE 9S

6ä £g5 j n

d cd ýE l Uö p ý, s i

lIg 1 til q c ý$i ' rv wm fv`° ?Tc ,N -ýý vö ^° uy. o6Ä S

c ö vÄ

E" ö �° c°

a 'o" 54 1^ ü1c ä° E° A ö .J

gmc lot ö, g vEmný ° Sc Es °ý'-

1! 0

n `c t,

ýde vý`^ mEcýý 9LdÄm$O Eä rýE Eý'ý

. t°o ^ EL'

ým oü 6r acýQ4 ? ENäE a cý'3w - ý`v'ý Mai

ö pN

ä0r ä Nr ZýC�a

d lag aye -L tow got Ed gong n. &A

0 0 0

a a ä

ýA Cy d$ Q

ýc ýr �rný o, Eü

rn Ä2 2

w2ý - - ýEä ýý EE

.6

d s n s

°. g

s c

e ý

c

d m

ý1- n

t: t

9) V

CC

i-ý Q)

4) it

in U)

0 U

. cn

'C7

x

bo

c; i C) E

1°

0 0 N

U

Q

b r. aý ts.

Q

n n ö r a ö I?

6

w

G.

5 ^-i' ý°ý r

Er ý

C

'ý

Hii!: I � vý

«Y °Sv ELE °~E ýq N

d

'6 E ., ., «xööIsä° Y TAE 3eü

ýý�ýSö'-38 q'8 I'rp

ý"maYÜ. ° «ý 5 ýP8

ä rýýö '

1SA°aE'nýä ýä. ýcý O öE°ö«ý°". ý"

ý o E

v

gm Eööü ßvä ýwr c

°s k

ýý Sren ö^

ýr 10äaýü. z k-wý c ý

10 Eö$; öi° .

n°

c wE ý EÄý

cSQý vw - c w inrn

"E - m, E c c öä

E m üý Piö b-

"ý'ýs"W äz3Ä= L B

ü

ý8 E 5-

ýz, « .. g

~

ä 9äöý° sý$Q ýyoYi § B

ýCV Epýav0: gýq s ý g

bu ° ý9"'° Ec eýE°c °öccE_S swrcD° gEEc

ý a ýi p

8üsvcE=i

ä6

ÖE¢ ?6 '°ý^q `

3L8$m$äS"'rýýýýö ýö q vu -E

« is Yo 'c g0 !! ü -°r

ýü ä£^v

Ä6ý°°°`"n

6d s PE

m 2 °i" .r tf w ö3ý s2C ýo L. b

co 'EN2

cz '.

o E i°E' m s8 ts

; - --,

ES ý= "

YýL°Qcäýg ýmwy

ü v E

° mt3H r ßn e. 5 , c ýý"

äN°. ýap3n

Ln CYC ý ýv ý ö

n - ö ö w..

öEä

9 ý oö ö$ci c°Eöý« ö

ýs C7 ö °äßc E =8

L° ý"' ü o °r- '" I CC $ý.. E '&

ö

üi' 83ä ý 6'

s ° öC c

ow ýý; E wE

3n ýs°s E c

vn 'ü`wä88Ääý"äýw ö ö

.$ ýý8ý°E8w .ý öE8 öp

e iýS°ä

Emö ° cm 6ö ö cö ý, Eo °" 6m ýýN

wý¢ --N b i

dN p Z°N

üB rn

ý ZdN

üs

Zd(ýy

üs rn

a Yry

4i Y U

i? P' E

v>w öm w 8m 89 öm L, 8m 8" I 8m e ct

o ae 19 i c 1%

ve r

e r v d v

°N

6

w

d

N

ý

ý

6

N

6

o 0 LLe a ýa.

_o d W"ý ýý ýÄ

CN Yw

5 5

b

C

p, .

7, a .2 9

Cz

FC tC C I 9C

°ü, QZ L e

W5ýrg uwi °r'

Y Gv E',

a

C 2 S

_ä. s $3

ýqö yt Sp =3 i'1.

1 or

EýSýý ý ä`s äÄF .

E$ östi ýiw

ý äi °

Qw öi

`v vEE

cv o E Eý S

u. mm m_ _3v S

ýk c Ü

.gc Ko

ýS i+ E ý

c vc E

c c v

ö ý ' z

ö m $

5 8

w Z K K K IýiS

C

LL

li IL

VV fJ LS (S

L N r-I

n H

cu

a) U c3 2 w s., U

a)

0 U

.. r 10

bO

C) E

,° N 0 0

U Q)

Ü

"o v

J

U m L

Ne

LT[n

' U 4E " V

va > Qs4" "

X? 5gý

. 0. E

QzSi

-Y n d °' L 0 0 C 0 G

C NGi

=F !

; f¬

8

Isis y- " e

Na'5 `e6

s¢ ý¬ ä

ÖF

ö¬ p d �

" " ör"i e

3eEFF:. anR a: ý inn cn aý F{

2F

, Z{ýý; ý& ýL RS i t

jý" öR Iii

LS 3E' ER= °ý6 P

S 4;

f° ' <F . ; ý°,. 1. _ : eiF 'ýF

! ei Yd

E�! =; <1. «dý F( ýf lt

ýEc n(e:; e

c ¬E`ep t Eint :

F': e ü

v° :i it 3C E ilEE cn 'g a _ it. v.

g ; "lFI; t WA te" i ü ä! " '3 °( ö' fi -! 8

" E 9`

°. f: Ei n !Ete ý ` °

ý ý gýYr ,,

f " 4ot Eä EY"i Sip 3: F ! lF

, doR.: ': g'`= e t

} pat !, £ 'S L' S !¬!

"E ý 1 E' j'+Fi

ý .

pa! o S.. i R t p

E F ii ;

°e ! p fc

-ýF i1 - ý; ý? Si4 F "n öYliýpi4. FttF öý

ý`¬4 no ! ig t !! £0 !i

.! ' tY Lä lii3; Ej

ýiti tit iä ö. i ý ¬ E $°t eSi

c F36'y Eiý ! "

ý "ý

Sf 1"t 'f F

l I :

tE ;( äii °

,i ' : ý °

g 8 ' aö psi S? `. ý - . ý. ä

iE R'ý YE: ' ý° ° 1'f `- !° Ei". s !, E° t t .E '<Sf ''

2'eý i i ýS {n fd: ý

!' ý8Sl; ' " °

q; ý°'a : i" " S S; e Sýga :ö

. SR ., , gis-- :' ýt 1 ö! $"oc"RFa cE s ! -ý8 5 8 L""

.; _ l. 1 ti ° . 'E. °, t".

it ýi a ý

i , : `r £'. iýjE L fff ' '%d ° E

! ! " t ä" i

3 . oý lF8EE

o

roll ý"'ýý£Sd ý. ny

! "ýF° ° ýý S: °: ö!! i

. ä "

p ön eöEr Epýýc"i"-ei! ^

°iE i"Y-e.

E Lcoleei ! EEtgFEp3V Rz=

iYi l

,, . .: eä F'! ý F£i':

? 'F F°.! e !

eS'eý !

x eq m F cdýi' 9 "iýýEFc }ýýdýi

p di ,i t

4 i. Ei"3 . o! "'

i s c n ! ma ", . Se. " "iý' °L

6i! d-E8 '

tü st os Eep o? rE4 "- äiýeR

. 3ýö:: lgd nS= c R«"

ei t ý+E:

. £oý. F!

_i°eE F°it s e!!. -' -till `!

i'-i" ut D: F^c°

ý

tö . L! Sa 9R

ep. la v ^ý iF. F i, = F F FF i S

, i39 >'nc9E n!

cE-°{s'ä E ee, ö .i lF g R

Ev£' ES L, F -ä

esijn h e"1! S° '

''Y.! dý" Y. .. , c Ril S! .e !, " !! R ä

H. p Fy ö-v u ;Rt uE. " s! -ýs¬

EER, t<aeL@oEä! EIi 6'

"R "°' ýý5

.o E'! l7"", CE -W . Fý

: flsi it L S'. °af

L. "' i° "' ° ; a: ii n's o F'E

C , olEERt . "! e! y

e- FE ö°}et- " i °

e . eo" EF nr p EýS Eý"y "s. - DýF . R£uö tf ý

oe t Sr". F : - ý

Fý} l . tF ý2n' i !{ LEA °s

.; "°S"'5 FF EIN F Ein E e' [.

F. £ pLi: cq ý"!

t ä°a p WF c ED ! SS@x S S§S ý NI

E l rvuE tFW_ o5: : " i; iF^!! At e !! d ÖE

!

3 FiEi

t P FF"' _!

l ° CS aip t Ei tS fat C.

i" F .e m SS n ' a -CE "ýt p etr

jie:. sýiR°°'iýFiE o, C . .. "bY! iieý "¢ R; lgý t 3! '(t: it; (lie p E liý"l ;: l

! {;! f ¢ Rý6{

SwE Ete.! l 1m FSi!! m te. E_nä. ai_ ei iF. iBE. i u: E! l9c!! ýeF+ö.

__E : i do"n eo. ' . °t_ ai F° ¬

'iö e ^

4 R Ri g Fö

lein 1! g .1 g F°

ýS F"

31 Ill '

pto S1t

j -1 do

"

E- "c 3

i. eiä a l.

5E19, aide

Ese I t 3

P e

3 e d

4 Dö 1

4", i bP 3S

4 Eg. iP e i 4

äs -mLL äi O

m ! 0

i e3nm :ä

i e !3° m °R pm R

äm . : #s'

t d f t S o

E f

f ä n t

m f m

mR3ýz ! 3 0 n ý, tý

. 51 q m °

m C ! m. 05 rc

i; a! o on [CnR°

.r 3: Eý F v "'t D =e ý4

R SR

. 24]-f °

.p g E ii: R

: t"EEä

! !i E ¬öý

p m ääs z: ý

9 äF

Fý

o'

!ý R ý

p ý

týý

S

`c

i g t

'eig rgý? eqýF

l ýs u

ä ý ý.. _.

Y

L ý

ON i;

: Lö Y SnL;; 2 5; :M

6

!8 n ýý!

-o' i Y. Q<3ýEa päj Erp äi'! °ýE ý 'ýý'. "=

sEtP: S ýYB"B °: äýý-`E ¬ 5" "ii;! fýf £L E F is

ES. I F o. üEE - e1e

ö! R': B, o F

et iES DE

I c. A, BS.

_. lý 3p rü ge äE

? F& : E£ m`äF

E 3 3

ý S ö E S ö _

p Y

C! o

' f y 3 R , `" F

3 ý : 5 e

" to E

F Ö

E2 FF

t F °

` Ö p e

ý mi 0 j

(- l i

n

00

co

d

u

N 0

N_ b

Cl) . iý

.. ti i.. l

N CJ Q7

E ir

110 N. C C

N J. J N_

UU G)

V

b u

Ö

Ö

N

rN N

adýt N

pN

65

'"° z'

v.. caw

mp 'GO °a mvm cu

pE

w °a

'S Mq Ö5

ý ° «"cä= ýý «ýudýs 8$ p^5 L^ ° °öESý ; ua'

n aE uäÄ s ý Y ,

" S gE «c uEýý nr E " '

I K c ä ýä Y =uýýý � üä üý E 1.1

A ' 2m '' ° ý ä

�et ¢ E 4 T

L° EÄ2

» " EE nvg ý ÄE

5 y

2v "

E s "w. 2 cY ÜO $e °ý, q , ^-pN

« Eööuýö° ` o va E^ . -- '" ä ýý ` ' oE

ny`-u Eä

m5 w vZ ='Säm

ä .r rv S 8 a.. ý+

a c¬ c3ö_ ö c« Liig °q-5

ä a. $Y

° E« E{ c ü: « E

_ 5 93t e'ä, "L ý° =m L'c ämýö'. t

u ý'S°o

.6 ^ý`u° °ä>u E it 2ýauq

AO 2° EE ^cöÄCE, E, ö-

E ýo d e

1! y^ m_

88EE`. °' c° öö °L'

ui8- 7- o

^ E$' 'SSE«c

cö '^ ° t_ ä E ävmS-Ä sEE- =d . ̀o . ýd . `e -- u«°

ä

9N .N

lNý

2 Z. N

vAb niE vmo v'i öc u) nýö m^t/1

c 0 0

d

0

Y d

LL

IS p 1T O _ Q N 9A

3 N9 Ný w

öm ^' ö `w

ýEü ýEü f 15

C'ýE

y a

E a

u 13

K

.c

a

° g

ö n

rn N.

Q

aý u as E s.. w

aý A, aý sr 0 N

0 u N

'e N

bo c Gn N co v

x. w

n 0 0 N N_

U v Q

"cs r. CU a Cl.

Q

ö 0 0 C 0 G

sNvE rn m

qqý

ý

` N

V O dýS

=Wý

oE v Yoi

E oEýý 2d ý,

ý No

E9ýo

tý3 vp ý' E

H z? z° z° z

Emö ö

2

ö ö ö °c ö ö °c "ä ö

C d W

ü Ip

d C

ýy gd

yý yýj

üq yýi

4 O'

N

ISd

1

IS d

LL

O '7 W m O NVm ?

NÜm

? Ö^

mpN ca s

a) N G HVm

fV ct CL Y

d 6

N > N

c K°

6 6

NN

K

fý ý-.. 2 A A

r'

E Y

.

°ý

Y

SE

O

w rj S y

3 eý ý. ý sCS cm Y =t

v... öc cN

$¬ md 5ga. ö

G

d E'rv nip 2. e

Eö c? Ef dö

cö. '- Y o: E

yl 1. Ip . .5.

S öCG q dd

ý

°

Q

Ä

s ` '

ý ý: v"äß rn m __ =. ý c vex C

ý E ü ýC E y 5

z 92 cc w Lt 0

Ü

C)

C) OD r-i

v v cý

w

-

Ö{

No

"ty

CJ

yi ýJ4 O

,r Nu LCp

m ý4E

ýoöö yd

°Eý

Cj

0 Eö'c

Q <Z7ý

fl P Ö

o O

Ö /ý O O

Er`s 6 d 6 d d 6 ° ä8

: fg {

fsf" äý3ý ä

n geFÖf

SQ

`s ß B

ýö'cFc i: nn .n s

äi yFÖI E'e ` i'

HIM,, g ; 8sg1

oä" Ei n: i : . °° °

ý7$ S UE"FF, B $.

$ .° irE

e Vi C : s`t °

E EU E::

° ' M Ru '

l. cilE °: i

srý F"iiý F E_"Td°°

is " . E$

°1 IM"

= Ze bOnöEff FPE y+

n . cp"5 biU i S

ý{ " e° $l n HIM ce", E E iu_=c

{ r eý J .° Mill

ö

Eg 4? " ?

Lo" -{

. $8+ bYf E nv; ý- n

e5 ýz: öö V

fw "°tl ý

. cE °i wal"s; ý . "EEY EE a ". $ ¬ßf1°

_

ýE: = : 8. y{ °ý$p

. "öE cö ` °s _ ". cý°° {! ° . eEgS"! »E a Ö O01

I

HIM Eiý ia

: ! U

% ; ! 3:

I ;

lSg EO6 :b j .q

ey, 41 u

; gö ° ý _oes°: `8 °Fý'ýef'38 $y ý5 : e. ý. "ý=$ E ' ' °ý7°ä°;

g F: "

ý }°ý eb0" o

Fäs' E 8 °>> Hi$E

3c< ý. M. 1s : mc.. 8[" 5 c°ogýess$ý

, a E= ý i. c :

: $. ==o :ý E"i

< '° .o . 5. $.

°P@ b3.. ý{. ä

¬ i" ý f

_a '+E° M. ti " ý

M.

ýý trfý im ä Yij aýir

_ '"EE

. " öz

-°gE. q.

Pý " .. E° "

: BE : 'C ö E i s;

Y 5 t8

$E ! qi"

öo äF , $ 8° ' w

ý "> `e ý"FF ea: ö» e F WN '"i : $°o: ý c"

ý _ -:, 3

- -M ýý":: 6.

em g ýEii < ` '£

i " z <! k 'ü' " 8:; ° "! -j9'°"-, o " °°.. °. g"$: {g i E L'

! V E

<<"b ý t° s ýäm

°t5 < ss

: 5lß °"e " F" %i

fi =

" f. eý.

ET e! oý, "ET 25, vl C&5

° nä7. °°oý= rZf if, ý

E p

ögý=i, a o ° Fö 6E

ýý. SEle' °b. {t; ; iss"i'

ý :E S"m iFBf i

m 2 "S. E' cF s yF !! ýEc

E ý o

$£t¬

ý g°cca8: ö" ° : ° °Ei°U¬8!! EM. ; ýEmýE: °" .

=ö$ nm 3°%T¬°. E: ei

of

f=»>` c . Y=ý. ESS»

; ý$ 8 'S`O 's"a°° rS'is°m : $"`ýý =3° wEwwww-SSHE ýs `w° be $;: Eý 3$Yö: 'oo: F: ýýc$EöCF ; so9es[s'p8

ýý:.

{ ný sill ° S giýn {¬ýý r1 ý4 $i I C C

ý H

1; 2 ° U. Öp. ä Ui nsyý

b ö ° 5p-. U Us. °

" Va

a ä< »Ö ^3 öm fill

: 3öm to flu

e311 n A m D

i0 ý A A n

gF o o

f cS E¬t fý ä ý

s S ä_ r

+i"i g ö« öý LL

p öýpe äý ; jog : fýg Eý$ý

" äE r F EöýE E: öieÖ EýU ö5 $ E

7

Tä ö$- $° H ' ¬ 3

` f £

E % E"2 gi ää i. Eý $

ö E%

ÜEEgi Ö°s ___ _ ýeFE öE$$8 ö0 Eý F

RE

_

E a c E `8 p

b e `s e ° Y ý Q

ö

ö ö i Q n Y

s Lö a Y p E t

7 f

6 l <_

l

00

N 00 r-i

U S..

a, ai IM4

N 0

U N

N

b0

C3 W E

i-' W

tl% C) C)

N

CL) U ^Cf

b z cu a a.

ö ö

" "Nrv N

r gN

ryä ýp tO

NN 6h

Nf GN

ry

E' YY °rn ý.. ö

^c L

,,. .., 3` c ög _ ? e-6° f 1t CE

°, °¬mY Ec"¬° i Eö ý'üýnLL 'ýöwEgEäg°£.: öÄ°°. öý- ., Y se s , q

_e ö: -ö$V^ T. 1-. - 9

YÄÖZVv ö

ý S S üc

¬- ` 'El wc Ms W

` " o«°äÖ E ö .P öc öP

ümö. f ca

ö

cceEc müS pyö

@E

e m« rnFvö9

L ö» E c` T^, wEp °vY op

w »` ° §^ SEngöic 1E_^'Lv S' 2 ýQüü °n cö

u° Y

Q^ ÖT 1J .

,c äv$ ýE öYi4 ä

c L °

EVF w öö Ym ` cäEp $`

°c 5°+ '

a. °i .

-ý -ü _

` Gb° ^ötY ä yLLF ac

°Y

ä n4c ° E: °

w¬ Y

Y°

mc 9 £övä 4

o _ ' q

eöýöE . 6S o _ ý2 äÄI E ä.

` EY°ä ý

a öSrnEuff

ö ö äö t ö.

ý. wýeäü ý Eü«m°ä `°pY

or a; dd` ýY -ý öge£Sv 9ý ö: F3e Ä -" -

ö n

ýc_Y^ "-°x ý ööü

- "E E

-E IE E� w dÄä , L' cQ e

L° S ¬iE

i-w. °jDCäSE°c fr °

3göö-

S ýý La e^

rö `yE ^9aL^°

v° ' f . c ÄFä25a ä°

wmü u$ oWävMS

ä_ mSZ "- cM`YýYä

z Ym s°öÄýö-

oS«er S- Yp $ý_ßewý °c29O

v gt', ä s5vv S«^ ýý`q Ä

°ä ö� _ .^ Y_

Eä ^cäu`v

^ Z ýi v

C" Yýö`üEd 'Y' °ý Y v_ ä¬ °Y v ä5 «¬rý°. Y_

z ' Eý °

Cw gýýýqC^aa, °y`

"° pE -c ý ÄE

q`a ý ý`ö Eý öErq -

Ve; `EE » ýö `'°Eý E0ü

S Y 'ä ýS: e^

c ä s` w e=¬ýF Te« cý

E -^S ý QöEv. EQeääEä 2

oE5öV. - ädE 5° i dý ü° Eä

ö ,

öPi m L' öYHr C°

y ä

s"ý Gvivö-

gq& t=äff= ä"

d ä U

a ý vY>w a3ömv 'n94 A '3cmvi

0 c

n a ä

s N ýL p

Q ry

ÜÖ

Y °

s d

N ýLý W y Ö

ts

c« t $d E Em öE ae

N ýscc 2

2w m 6m c 6ä m E.

9

3 ce

z e z ^

g m

$ m

ý w

Z

M 00 r-i

19r 00 r-i

a) U

w

a)

a)

0

10 . C4

b0

. yea In

c3 GJ

E W

N. O O N 4,

U

U

C)

l 7 M

O O ý q ö

S

tý N

Y Q

ß

rm

2ýý

t; ßä^ -ää ÄIý, ö

6 d

° ° ^ ,ý ,

»

v' °=fir If E' eö örnms

s E' ýs E !

6 a ,$E

ý"`$ °'s moo v

ý Ev Q ='ý°»S Er i`'L`'gE9

EL'

m°'

g^N

w

m N

ý

2 cw

dN

d ý

ý

9Eö WN

$

Eö 9

dz N d

ý

S. Ep

N

Y rn

ý. E ö

dd Q

ýK ý

d m

II ýc m

m

Q rn

='^ ä üm

Yßg

d ip ý

z Scd

ý°am

ým Sc v ý3m

ui ö nü

o

n

O

.. CL

o v ýr

6

ýup M

6

N

p

d

CD C7 m

Q

U' LL P... :: i

I3 . `ý

.p .40 ^= d

Qf 3 äý

" 'e Ly ýýdCad üW

En 5 ü;

,Y Es ä

öýc ýcý°ice v'ý5 s. cs EE

- tS E ü

2 öd ö

w rn Z

l: ý

ü» YY1

G W

ýi LL

C) C>

Lf)

00 r-i

U a)

U

Cz

0

Sr

b0

i-. tiO

Ö

Nm

U IL"

ösg

0. ah

.<QH!

;, i 0 - EP äi¬

° d oo d d

eF 2F

, 4$

E 3 F

a 8ed; d ON'Fei

of

,cc 4 ýEa n o: -

ri o öý

° a

" tE qi

; tt'CEäe> ý?

t d'! ' g4

Ea "g rý( ¬ °ý

"'

@' FgýS':: i if° "S° " r

t (

, i.

" giif

6d !a ý: ý;

"iý;

faý"'"

, iat` iä ' ti ýa

i° :: '

lij8: ' g_ý. i , La t ! c

sýni SS Pj

`4d' ý g2ý¬j#" 1L"" ; 3ý¬-$i " i. ° a °

i°i Z"i "i ¬L °ý! f S'a°Y ti ;

i ( tai i! £

EY ` a ¬' ![

! i .9 sä E"'°", E

i"i, " a". a

i ä i. i. g"" aý!. 'ad d1ýý [L ti. ý"

o4PýEý 4> E'iF'. o n

: l. g £ Egs. ;.; ; iý it "g i. F"ia °

a dE 8gga: S 3tö F" i i äiE

: e4 48¬aFid'4{'lS a

afvq'9agEýi £ 3 " i``Figa

3°ßa2°r'aiý £ ` i

} _ ! 83 ýq [igF'a` ý'i .5 4igj¬ { ý :; i:

r42E'ZE: " Eec. ý6: Y °Y ; ;:::: i ;

8 ! ""c" °° E

: "¬c "a, s + tYf !{

tec i` !£ ¬e ý t "4g t., + cpa i

i "a ßi ä>

°t . ýEýE

, il4ý: i ` a` + E d t£E'ö ! . F¬:

c 's-: rä >i""i; iý: l;: i (jaS :E «i : 6¬r; ýýq F ?ý

. .; iý'

er

'£

-. ýg?: : o. i " t i¬'

d ¬(ý"> ¬ ' ýa 4{ ä' °

: E: {8ia¬i 1t F SEätt !t "F `

:EE E p ti 1+-" >W gEl:, -' aýE¬ ý

[a! '. ýý

i; ¬ ý e

ä" °E£E iC i: 'c

iE !

i£ EE" °i¬e'`gi !t

ý" 4

5! i, i i

' e5 Lrt Fj£r

i"a°6E S i1

' ýa '

- ., i "¬. >ýSq4 ¬f; ": ¬'Ti4F. '` ' l: S ! S, i¬.. ý, F °2 ý ° .° 'si. !

3: ¬ü, a sac#ý> "Et iE ýý E't': i = f'a.: '' " i' ý fF`L9' q! s"t¬ ý !: i`

°r 3 'D o C

''v a "t ¬ =qS `aF `a "88i YE

I ¬ q a.: ° ý

.!: ¬gý¬: E

: iý ' i4

ar i ! ý

E ¬arý di

i 6b{ "{ 3-lE" ii¬i 6' i ys t ' ý ,

gý br ¬q. i: l re ý' "ý! 3 ' i t

! [ CÖ ! a,. E sEät y¬ýf£(p '. EL, 8. l: "F ! t°iE. ° ý¬ ° aEiýli ;. '£SE Fila'! t : cj

°'"S4 :r S .: g :: ßä: i

i , .

ai °

d :: ýt

t! % e - aý 4 ici"$ö. a.. °, °¬ lýý" " .

ý £' "

. L£ £aE a. f !

" " nýp 9 4 ! i i'"i-i ! a¬a! i "i.

ý: ": _iE

i >.. " {

`e qiq ¬6 ; ýti

ri3ý '°¬( ýE¬. ig: ' g l'- ö '

cY ii: it

°g`s - : "qE" f, Fif x. 24 e¬

+° ! !! ýF i a. gf.

x °$ ! ': aq ä'i'62Sd'. s, es"t: .;; `aia, "

ei l: ti

i, E Sjs"= : ii i

jö' 9tc9i5

£ tý

tf 8 "i. " a +: =ý ýýý E: iý : ES = ; t ! g ; ai¬l8ä8 ''ii ¬4e: 2E

t ° t> i' ia` °ýi ýt ý2¬ - ¬gE ¬E °'! tt ýE : Ti'6

¬° ¬¬6

f'i `Ea'

'" °

, liE i" e. '°'t aýää6liýcEiä, 6qý ýý Ca

Eoi if i" p: 4 *e iFPjýý i ý +£y ES

¬ , . -dý°! c°yi9

°1ja eF t]3

ý P6 iy ¬a IIFi ý

0

" ö

{ý i a' iü i° !¬S : sý;!

° ¬tiE ¬ZEEcý iýiä goF¬ýý¬i

°

, .[ 't : iF' "¬`lF:, £ii,

c E? `F:

: Y"

"a cF:

iE {,; sa -s 5'!

° ia

! S : t: iEý" ¬äöý`i. °iý.!

ý8!

"j, p ba E Eiti l

°, t "S ý$:

Sap. Sýitf'"^': ce

4iý° 'ý

; Edi? i 8 ac p °"iaia.

iý$iýbaýslII t'tdE°'ý ` g` t

° .! i ? £ý! iFýii? tF: tE:

p ;` . a :

S. i !

i a i l ; ¬E¬8i¬ i

iO ýiE Si `SüE üEei! ýSÖý: Fa[ i:: c¬E. 2 laFViäät "eF :: ¬öi2 it ii? iiE: i i aäFi6i: F

S i ö �

{s 4iýn lijg fl p S ý 01. cýg E Ei, P

aä ?

0

S to ý, "

: aybm scti

egge

ý"iý a

i °pb9

, ti g

4

.S £i

31 , iP ° a>`i"

v SP"ö °i " i W¬ m i0 i; am ý90 i; 9D 0

°0 . i90 390

n b d Ö d

1 .ý a

Ö 0

+

ms; QgE !i

ö d

"

.: ft : ';;

°

ö ý

sq die: iC 0°n

2 e¬iav

" _

4 C

g ° ° ' E P. Ed °E°i 'Y ae

"i

a at: E

"Eý: 'ý " t i: d

b ¢e

E Np esýit 11.12¬ ä

e'E is ý° qt4

ö i2 ýü >ýe

to iF:

i ES$g Bp4i ýeý ý > ? 1d> iät egg $

"bav 95.! a0. s s: ¬{0Y iT i? a? ¬c ,ß ýi! '=_ E u F

! >Sfl=

¬qF 8ýý3iý =ýq' aE ýý E jiEF Eý¬ ¬S a

_ E

o

3 E

3

P

t

L

ä 3 °

E

rt a

2 E !

PE r

1 j 3 cF c E -

a

o ?4

e E 2 y

0 F

<-

00

n co r-i

CU r. U

C) U cli

w 1.4 a) Cý. a) s., N 0 UU bN N

bO

N

U E 1° N- 0 0

N_

UU U

10 aý C1ý C14

ö ö 0

6 c0

TmS ü

d

m

^ E E ° c c 0 Y8ES , ü. q ü LL q -. Vq %

Qd W ýýE°ý'ýýc cater, ä"ö q'=u° ° °Y u? E

'ý-° cd'°"ýýO

T., iEfqý, e

`n¬rvmý %Sö'G° E 2

äÄ öL

43 - 9 ,5- oöc E 0 SJ Q ý5c £c Öý2, . `ý n

-c is -r 9 e! °°ü cý Ö° bc ýý

yüä c9 ý m

öö ýýün .6 yý q ÄÖm

. ý- ö Ää

aý . - Tnrý 'ü c .ý s K`ý

u ý eý

Eý° äu c_ Vý Äy üS _9

ý ö

_ 42

y i= ö ýE äL2. R fi EEv

1 W. -P ýE I vÄ2 E EE ` fr

e

K Y Z, mm c_°. qý m E

äK

=! ä v eS 8§ä bsE

CQ ýýw= 'e .t

EEö `ý' 8 MUM ýO YcE

Eo g~N O --ff (9

,6 °° r cö

@öäo' "ýö

o $CäN

'p W N> Om fn h9m fA

ÖmN

O O

d p d

s LL gý

12

öä ° ö

~öä

E 5E3' Q2Ä Ä2Ä Ä2

5 .E

ýe K

5

a _

E

OD co r1

cu a)

0 a) v cu 2 w

a

a)

0 U y

.A

N

C. 7

40. t"

CD

"C7

U

4 c ö ö 4

ö

4

n

EL

rv ý

Q5 E

u 7;;

i Eö z° z° z°

Em° Q .°N

CD

ö N

°c "Eö dN

ýs ö dN

ö c"E °° d

c cö dN

cm dN

Y ä ts m 1

ýi Yd dmäcä m$ rn

ate, vw pc e' 1y 97W Öm Üm N

N °° m

Ö M" GÜm OVE as m

ö N

2e o2 O

öC ý O 1ý

e 1ý

öe r

O d Y

6

m

d

N

v d

ý

v d

N

ý d

v'

U

Um

Om

v)

0

r

LL

v)

(7

.Lr ný d 12 Cg C

3ý Cý NN

y E N''L 3 .9t! cc c

qt Tý

Ö L rýo ý ý

c. "S g c.

>, c c w- öÖ

=c cý

yý Nä .2

g. u

2L S u

Ä'ßü

n .Sc cvo

EE

m sif ýE ýNý¬c ýi E

EE ýý

E Y t ° E

Nq cý' °ö v. tcN L� ýö

"ý! Ü r

y

Y

ö

ö ý Q ý

E

.5

ö

u

Q Z C & R'

Y

w 5ý W

p lL lL

!7 C1

rn OD r-I

JQ

ýC a U 4) cý Ly

y ! g8 8

Ü2. E

TEE a

Cl- X

CöE

Q iýý ý

N O

Eýy a3

N d

0 d

03 Zc

0ö ö8¬ ä

c7 öý

o°äää ý

N äö

MO

'ij üm ý

O" ý Ö

' ÖCA

r Yl r^

r YI

°g s' 8£

ný2. E° ° ý X $R -ýE

ö r' Nüo m rü"uEN ä

E° t'i c¬ mörnvm3

° 5d" E" :.,

$ý ý i äLgn T ý

ä° w. C ¬a

^Esö w c" 2PX p uncäNü3ö

ý g W� Sü » öSä e° EEE cmEa: öä °og

n _ apKä EccgE kv

ýF :S

cääö

v$ :

_- t 'ý c

"c°m c E

9 c o@ B1Q§Eu üY

b00 ü

`ß '6°

s ý$ Go

° ° 0o

vg p säg ýä

E

y ý -S-8 £

E ü = ý o ý e

g `" E Iý ý ö ýse p g=

öE äw . nr

ö v d ý

ü° c" E E . »At BM iö 5

¬ö b ö

" °' c ö °° " 9 ^_oöW Öý'cwy E l $n e

ü ny

c

Em aE 8W`i sö Yöü .e e ^g

m ýumw ý$mc°

E o E aäa

a sö9°n EEL'p

$XýR"9ÄmEF5$$i": öy8 g X ä aön$'BEBý»ti. ýE cw"i YCO=mE p icFtoEýge9 zb

°ýä q°

jcäKS°Eacw" a

äßä. l ööö$r ýw

`

Eoý8. ý

g

WgZ ry c'ýSSö ^ä'i ýö ýýac ýEýn

c w ý{ fi g .. 'C. -'

g" ytüEtöau

'. aý"ý`ý: E

ßz55 3öZ

ý ~^ýý y1nüg

u"y { c°ý

B 5 tl c°

ö° E öw3

cä3ý ýý%§ýw

"`äF :sý? ° ööý r" Eýa$ E8m öw

U¬ ö ö9ý äd; nL

¢ Ssu ü °`,

ý, " 'ýä cü R y

z p 8

öb i ETie '. E5 SLL «SüX ÄýB

F'± g: 3 ° 8 $'. wý»s¬

ö5¬? . ý9^

Ä° Eg LB E ý¬

C E : »5cß

8: E = E ý° EC° ý: s ýP ^ §" ab ýo; o-

7ü ö ýv^

x y 4

a ý$ ý ä ¬ä 58

ä: e Q Q 5>° `^ $w,: "° 'd E 3. °w .öüäögs ö: ä 9CöSESwEE E" äEFö

Eýý c c

ö- ý U L

mýý LL väw S rv

om ý �Ö

n v �3°m" ä3öm

_ X X X N N N

n ýWc

äcý+v Ký ý C -1 2 U rv ^ Un

¢`ö

F ý2ýö "ön

gý 9ý F nntmü-

X38= Eu

w

göý

=

äö: gýE g�«qSn

m övä

n . Nä; ß E9v$ 1

vnEn ü' Yý E3E

3 EEE

a E

a 2

ö

'C

E

ö

¬ ý

U c ü

m Y ö

iý rEý

%E

yý y Q QL _

o ; C s Qr

d_ d

w

s.,

. --. } V

bo

l°

0 0 N N_

U U

. L"

U

0

ö

bN

ý^ Ö

ad

W~dq~ qý ý? In = ma u� ' ý Ta 13 $'Ny

.2clÄ YÄ ä

2 15 ö ý ` 2c ö

. 5! cE¬ ýc Eä =cö

-R 9 ý ü ü, göc°'

EcöEc°ý$. qýtG

E c° rx aEp ö 2.2 -

_'" ý° °w °"vE m $ q ý

Va 9L ýa m

E 1.1;

CÄSv. ýSý°= EE S'`t ß'm°Ä } c ýI� äi Em m°ü 3

Eýö caö c' c

-N f"gät°., `°c,

äsý g °cäýN ,EYg 9W

ni y1!

0

N R1(n 4!

°j Oi.

N hei 9 mcd . ß'2 dd

hV mN Ö

Ö

ä v n

41 g

K LL

N 'u g

cd

Na c do

79 'S

°, E

m2 °u m 2ý

i , E :. E ä n E

d E

+c 2 ý

rn E

a m m

ON r-q

N O\ r-i

aý

aý 0

x U

td

W i. ý U

a aý 53 53 0 U

"d

. `l. y 3-.

b0

vi UC

G

w

n O O N

U U r. Ü

b r. G) a a

Ö O N

ö ö 4

G d d d

r °? S hii: ;cEü

E

.. öa mSc°i 6ý

rm°. z,

9 gcöý :

c 12 IS

vt ö'ý, ý°,

oä, ß2. 5$ ý

Sý° `rte%öc q cg`mö . F. an 6

äEý°'wö

äS xý Äeß3Fd E ¢ö

C' StEm {ý' :Q.

ý ;a Ä°c4 m vö

Cö

-2 Tu ° $ c

tS sU 2° öEcýl öä ý,

Lz ý ýN 2Y .°o

ý° E .Eýý .6E .üý. öam mrävy Y '3ý 7K

äý-°E '3 ý °

u` l

Lßä '" `R

ö c9 z ý. L^ö ßEEn k° Öm Ö Z

goo °E+yý

ö N

ý'm

ý6ý tZN

4ö

gýý ýýrn

ö s6ö Npp.. N

ým ýBrn <<n

dN

yßä ýCpc. dN

yarn E cLL

y a

. ä'" d cd <+mä"ä cý S" Ä

vw öm °n $m Im öm L k; 38 °. 8o ;e N

öe o

öe 0 oe ö¢

on se n n

0

vv v a

ro

d

cD

6

N

ý

ly

6

N

ED

N

cý c

d

(9 LL Ei cý

g' " 2. CY. m .SE EN

:9 . u

° ö. E 8R=. E

ßäc ýýE äß. g ýý ýý6 Eü

ä N¬ cs

EE E

ÜöE ýn mEv ° Üc Kö

E c

I

r ý

ö

j

" Z e0

ci V

M Oh

, d- rn

f Y i e r

4

I

E 2O

e

oJi

N ýö +' L4

as Uq '" a) gei,

p=`i 0'4E

aEic Q Qýg'ö

xeg.

U 0"i

Q Qiiý

8

p Y

e, -

HT n d d

E o

o d

o d

eý

P

z£

F#s - l il ÖdF

ll A

°oco o nS öF

iE

uF

n rin

3S I' "¬ "-! Ebii: BFS. e !. S

_" 6 Ei: fi

" SB F". `

tc ,i ° x ,Y Y{: `: S i¢ iL t f°- F

s{i;

E

t Q'jý`

LEt

n t 9g! '. "t°b Ctcp w

. . 11 ! tS FE6 Y'T 6t!

c" b 4"'orE`Dq?

"b" iö EIN 1 RIM

"! :

, EreýSFO {

tý"' Y Y'^ý b+

ý!. "i l ! 1

F. . o týýt: bý "ýb `ii : `, p F. 6

ce "i°d=Y"" ° "4 'ýEEY !° , E"

{ äS6 i'i. EýS

SFa jl

i ' 4@

D! " äe

YgySi 1 ¬ '

� c

" "Eg E! EF äYSSäi 2e

ýY ä2¬ ýYt`ý:: ýF

! {° t . Stitý'F"!. ¬ , F o

OE l8ýCm3 FF.: sä o4 Lg i

fE+ EefY Fiý <ý ý ,e i 1F, f t. ¬

! S "týlS "

S Pf ¬ý- S. lt D : Yt

" E

_ EeSe: Sc E"ý: Ö tPEi S'

P, Yi ii i'i Eg AYE {"P E-' c

ts! °

ý z ý4ee Sdr!! e. S ýlE. = ý.

"D i8

co ". Bt: S- o. ° ` "- " e

Y -sl, i rf- all

. 'O S:. 3'ý

Yýaie ez" y¬SiSt ! { ý ° eP '

" °° "s . F. i{! b; e : S; °. lF if Y ism

i ° 3 '°

1_ bi ný3` y

cE ; ¬{"E

Y° äOP ý

- ° F

öi !.

ý} ý(

' ! `F kC=S- F ! '..

ý L, ä ,1 "

{ e YE'iaFE {

Ee to 3 s

E` LY ¬" tIM

V1 :1an

FiS YýS

tl ýE': i

¬° ¬

FT e e öE. 7F ° , griE ?D SlG o-°. l

e, to'e+ . F ." !.

f ! cE 4ý' i

e-ýýý`e , iý 6ý¬i= Y ' t , [o $. ý"

n ! 'iEl¬"ý ýtE

[ " . . Fe ýi i a'S EY Y ;. j {: ý: S f"¬ e

ý'ýi! _ F"ý

'. E "'i FýfE i p° ý iE! "

D F S "ta no" 3. " f=. -, F E.. ý4ö "E{ fy'tie. ýle _ e °

?S p °

ý: Y o t[ " lYFt.:

ýS"'F FE E 4 ° ,

F L n La

Fl ¬E3{. ' S ßiF "i

ttlp °{°i

o 2 " t Y {rrE e iý ' fr rS' E E, 9 SG=z"e°, 1 iFb. F

`ý ýö E S

e, E n' °. bm4 böl eO gt°Se ýi

"i "ib. i '

ýC E Lt i ¬` {YPa{id i S -! " ýý ý ý ; { ý S 5

'gi . iPi E, ¬Y i! L`

[ a ý " ö. ý8 ; lSE {! . i[ýP ý. ["{. m¬ Si:.: E"; -¬ EE ;i 'e ý F b : >ý £i e '

p a P. ý ieS: S4ýýb.: . rto3 YYfoý . ° ' ý'0 i

' { #

ý S> c: S {l ý' F:

. EýtF! '°ýý`ei S'

p F' ¬a67:.; , Fý . ' `

tE{ 6i" yi e'ý

( 3t! EP¬IPe ýr.. "" ý

S tF 3 Syfi

" i _; ý £i¬ ;

, ( tt Es -, ¬a E S`. ̀ ýýý: E ` ý. : tm", -'!. ý. SEY( {! En

F" alYt z

"3 Y. : i=

y# SEr[fl".? ziel: e- ýy '

i ;£

ý ý . tP F" ¬' E .[

e. iýýý£ý°'rif: F "¬Y; E ` ° `

ý n i'

; ee °

S4 'ý¬° pý ° : ;

ý`ýs

t ýfi S° 3i-i"` e! cýEý t ü i'eSY3YöSFi E¢Ff e.

¬NS{F {¬d S °

! -" ý`r: °m@. lSwFý ý

sE b$ Sö - ýf ti Ei E t ESc. ¬i`ES¬Y

"i 3c t Sý. 6 ¬ E '

f{ 5 3 tFtýY i4' " ;

[ i n 5ý

"ei° 'gi " "F'5335°°

'_" i. ý

iaE¬r `f P,

°E:: GiE{ c i . § °E£

MAN Yü

.pt: iPi ei ='cs `

[; ) ýpi=, -E' ? Eä''

. S£`"r`i(s"°F' LF tOý

f " e 43 YS {F S

D e!

¬4' q`

S: ¬"E; i 3"

` p w {ä ".

't {Z HIM 't

° E"-[

o FoitP oc["rS r E- t ": 'YEo °

F .3

S`i: " J Fý °'z£" £. E'

-

Y t° 2°S

' P 3: ! tF 2Fe , °EE'

S ý:

p r E °E ' 2rä iýaD ,ý f' ": F ¬

o": Eo E, '" "' >: §2 _, ý3 : ýSa; oFF

H g_- £cý:;

ka i7" i e: EE Sý

lE r: `E iE 2

ciý

'; { e f ! "¬F y=P: eý

3S ¬'SSý Si. 3. f

£{ö ". t a {ý `iF2lf9 ä .

i'S i34F eib#

S °7 S

ý : - ¬. [¬ 'n¬: ýF-_ýe ý: SYýigY:: m: B[: i E". s:

H

. tq O° ý; 8 bwY=Eö. 'i? EYS° ° . leL'x

x '!: :: ýiaiig:: ý¬ t{ P.. t¬i zý ýýäes i; ý(e:: l. _iiI

id's träg 4äg "

! f ! 1ä8 fill 8p8 p

-'-ü" 3, eý %191 Ei=S ; iii S

E ES aýTs ifýS i_ ä[ o :ö i3DÖ : 3DE : 3äm e 3Dp "3äa

L t g ä t b ö a

i R m a

mSopPi ° ° ä Y, b n

m m m

t : FP ' m@

ä

m S°SC

oil

0°n 0'n [ p

Hit;

aP ': bi tEl

t°

F£re

°

E9iJ ! s ' s 3 "Lii

s ý ýfDYr't" RSS"tS .: Y ;

Eý: F E E

F`s

sI Y if8

oS ýýý

p.

3ýl: F u ý"E

E ES't EýBi '{FE° ;E ^ ý

dill'. iäf . 6! 'ýf¬ 8i¬8 q ö6 ö

=: B F.

ýä¬

! c Y { {

p

b y 4 0 4 _

° 3 c E P

i i s ? S ?i - 3 i '

s$ ¬ a t e

E 2

H ` = i

o i ý a - 4 S

ö s ý

t <ý 1

Ln o. r-i

rn

0

C) U

a)

a)

0 U

bA

ti cC UC

O O N

U

U bÖ Q

.Y b aý a

ö ä

°mö

öön ý

$ 5! : z"£" V f':

3S 3E

e " !

eve o {F¬ ` `

H iES13 ':

c

S "ý cö

to bg ee A

ý

"E !"t jk ý§ýqý S. S

L L ! s" : fE$ . SSi: q eF `

ý ` S ̀EeEý ES t'' Y! :; :ý

vFi t "- " E, ELS ' "c ý : E.

: E3"_"- S S7E~ £EttEy a- e

ý° L f! "g gi`E e

e MIR :

t5 Ei{! , °, i "E `Ef L f

c -i 2r. P °: i c8°'e 'O48i

s: °`ýS "i` c°

:I r fi`ýöiý E° °

MIA! $"EE`IA! i! p-S:

iö ý voc ! xE

. cE "ýy

L °; E: c S: £ E

ý°: LLäa "{E äidä E! Sý fý i: fE "

E ö "S T!!. 'VgSS

2Q `es e -°x ° !°

-, 3ýlt iEýc E°"t >! g _°E ' `

Y 3-r" R N Y2

t Su lB pjz Sgý=ýYc"g" u

¬ . uSýfi

fa°t¬! c25

"! "" .c E lö :

iH St '

£; E ' a

sä Fi iy EL dS ý¬; E ' °e "ä `ä" B

tý: s is. : EtYO t ýeEy

2S

U ý. I : f EE _6n Eä Sýääi' $L eb"Eö ALl, E Le YY HUN, tl;

ei al ý B ; lt"; B. x;

$ v -e v2`i, f E° "OHM E c: S ¬

. t' '¬>: Fa Y

c , i <: s" ¬. B :

, E m l= ät` °

f¬ RUN ! C�t - f-

>"e . ýEYoE it c mý3 ". °¬.. i a [

... Z`s Bsä

ce"ö, 1ý

s lagl.. t'

ö ö6' : öý " °

: E F F -`!.

IN °t! °

ý ißt S"i :. SF

2_e.. ä :

f: ! -a5' c

: BEeo°°o" E: "Cj

p ý R i_y4 ýe LfSe t ý'L t

¬ HN -IMIe

.S E!

5t§ ° 5ýý: f' gi" i1 a'U

¬ £a a> ; 3ýý", c , i¬

ö" Qý, raýo-°'ýif'i E'ýe . ¬. 'e v r. g El Cýtf

c

nc 'aý°B '" v ' ° E

"" Sm O "' eusg

cý 5: ef

" , Fag: Ei: >! °`!

Ei" S ° t. E

< "e" E[ i

ý y -` : j:, ý Ei

ai' ! "F 'E- oe'Sc £ ""b"

: B; Eý_ ä:

If_ £: E E a-- P. e °E, S l,. ;ä °

r ,E : , g! r" fi; 3 ` ýu'#Er Bi! off

° t ö t , "°_ t¢ rs. g, :; f ý 'e

ý"; E . EC

"EE EEf Ei °

:.? i! 3 E °: `. :' ý $°. a ýe. $[` !4 o;

`s: B!: Y 52EE"' S$`0f 4 ; b° tE E $° o S '

. p. i E4- 8E, tP " ` °

ý n$c y ý "` ý ¬BS' ?

c E d c5 ý

'm; "E P ý: E i z Yf ' ý: °cE: FäEilö-`m

. i@Fý¬2. £c :P ° iliY: Baý

Y ° t °

: öf ýeSE[v . `öi. ¬i

f

_ __ yl. cremE ý°

° s E" S ye'E: j=i! B ! ýf` E? ̀

E? sV'E`Er E E

ý-SSýE ." S i ¬f

EenESEa t E ýý - c

ýs ö $ ý : - EESLtn: ¬ Ef äS ýS yl

e° !' e Ü-B

"'f fJY" ý F

- 1 5525beQ ¬ flý° ¬[ : °t

S>. ! -" Pc

" ýr Fpä'"LEB

of 'cý"

°t ,ý °x3 rP; E¬ý. - O; F:

. "¬S 5ut . fý n°

ö !ý ` ý

fS fýCrO - 5

ý F' ¢ f SIIýi f: i

>°r.

ýF . °F O r a

i äFx4.4": iS ' ý°

ý' ' pp Ee°Eii. f E` r: f !

v :! äeEiE M°

: 5:: °> E E.

g, E vEE ei rg .

{>? E= ý

ý

S; Pf ý,

lf : F'

' . EE 2 :; E

"" ', lS ifs Sfiý 3E r°

` " iv E: S: @L ä

° f S! `>e 'e . b- y !

iii%cv. B: E"iEý oý f S

°! °: E B"°

.E FrýBiýr4"u : °f eC, QP ö gf j E ¬

B BL _

a.: _x o, ,O

"s ..

ä T. ° seiSlC öeSi . F1 [` =s , _ iýsi. >, pa

E$g: °f` , sf 5¬

F'-fi $ i i: . EEtre"S ! £5t'i

N xSä. "cy`ic iF°! i'"' . '£ IM Sa 'f°1i1+`ýýY B mýýY i ý

ö =eEF`$; ý ?6 :L °ä35'=E : ý%. öF° ! P`

E E

.>BpL ° !F e ýrr°!.

y' -i '"

s °"v ¬ý r"t tr i

` t Edl. r,.

: E tný: °eE

ý B, 53 ý"

EE: ti' EEi_ "eF¬iýi! ". v :Y v>"

$ýi c

YSiPy: - i: SE" f ^f E: E: 3 eýiýi°fi"3

ý G S 20": S.

ääiä_°¬¬EL. m69Sý fIä! > . Sä°.. °2" °ti[ e$leEflE! <e iL45nixcE ý fE¬ösf" Eid>EiS r _'liE3:

9i [ : ý >1E: E ýýöEEi: ü

°eö

' ° m Eibi g

¢ W

-p ii m ä; ämtl

me am 0 9ý äýÖ W

SE D* -

°c" ý" f

Eei ii '

ý !

E x ý 3

t Q ä b

F F p

ým

N. 0

00 rn r-q

0

a)

a) a)

0 U

do "Mýy

h

i-.

t

CD CD

UU a) U bý

. 21 b a Lý.

ö 0 4 ý 4

ýp N

a a a a

Sm

E b y

Zý o r? E

Y -

- d ý°

öü rnE

vý ö

E

U 6ý° U fl

c

.! 2. ý YaýU' rnTýiý

Y w ýö ö Ec

cE ýE a ö

ý

dý�ý _ds

2 q

U6 ° p ° $ c z in 5. v z

vm ö .. ö ý. Eö ä c. 9ö c. 6ö ý. E$ QN ddN

2

d (V dN -N d

YQ K

-E ýý,

d

4 rn I'' N

ýY ö. N

ý4 d 9 rn

LL d

1 yö

PP

Y r

a

m

a

N M

6

N

6

r N

ED

r h

c! )

E s

ýýc . °ý ýý

ö coE:, g o

zc den

pz5 c

Tde

S .: 5E.. a. ýd

SE..

-

$Eg ý C

-eN>

ý d Y i O m cý

= äc

n °' r°ý

äE ý `ýc. ca5

ß. `v 6 8

Ä. 6c E

8

ý E oöE ' 0 of m 5 c Ütc Ko

.ý a

E

Z K K W Z

Y. lL

LL

0- C)

rn rn r-q

-ý4

v

C) CC

S-+ G3

CJ 7.4

h

0 U 42

.ý "n w tZ

0 0 N

f i

jF Yy

SQ

Zýý U-

a ``"

;

6E

affpig ý

, is

[

Eý4¬ 3 Eý ! ? s 3

eFSF

iP

3 ý[ýi"

ýn °n

ýýi iiie ýEiEe i Fei w'E y'ti

g'yäi öi ¬ yErF ¬fF°E ¬t.! E

iEE fä°=' °t"¬ lSij °

t¬: r E1

F " bi ý '

[ E ¬ Ei i. ¬S`¬i're ii

cý . ° S°¬`;

DS E'! S F'i: 'ýE

. ý. S

i q

,2 t" dGEl E¬ ° ä2f

t °:

itS°S. Z .: E. i5 'l(. f `i¬"'

e rS oýj. a- E

E ¬t R 1t

:, S tt{FF LEt°E° "'

ýO SSE ! ¬°

'aSii ri [

. ý ,

%6 [ { ep : 't ° g " n E c. { IN " °°

. cý BBB f-{ i !SE! E t" ' ° t! t

. 'ef F ¬¬¬ýf a¬

°¬ '

. : Lf eEFSfS

`

E} i!

¬r c °t

"F f tE§: i" i

eä E. c_: _¬t ° E'p

¬j0 ° ¬. +: S

%S-'t F itý

i 1

¬j EFS a S". 2 : E

°I tF 8(OgE

°j E" ,E Pi f Efa ii : 4i [E tEc'¬

tmS C t { ' g"eS;

. Si{: ' ä L , t- 'E `p 'iFE E ,

r. SE "6 ¬S Srt ¬"' E_ ¬ý'0'` in 7 .¬

FS. r. : Euy: t-S'«°FYi , I c ; }i ti°.,!

E ýý ötstt i

"ýEf¬ S

{ {t iä[!

"°$! FcI Y" E

7 . ¬c °'. Eä f {a E 'I' ¬ :

: '¬ ¬ýFF¢Stýd2

' " ýFlS er

F ` ¬¬. ¬.. `. £E; ti

E °c . i. ""Et .i iSýEttS¬. . : ° '

°t E !: I E

py i ?F E.. "c t 3'- '1pp g EE . e S ý, f E°

! EEE f° S: ES 6 it

3 "erl o. `t t! { ýß .3

.i° sE S

Yä iii: u¬t % ¬"!: t¬

? F! E°: f¬ 5" S itFpt i:

¬

ýi`.. ".: °,

" E

{ ; ' atEä:

EýS! [öýý t¬c" ° t

t t F° q .

4r'ýf äSýE "! "' E

E EcE ¬F ¬f. E i ' '

P Fo'q;!! io p3 aI'

p. t °{IE E°ýSc'. F tcS>'! " p £ ¬E!

fEE ii `e c¬ IF Et °2S

¬ Si

_ ýYý, S[ F `{E¬; ýS¬

p¬t? ý ;E ! eä

,¬ F"{° , r¬e" ! p "r ru El '

fps¬¬. "! 'ZE l F'6 E E iýp ¬ä{

[ t !' ¬ i; ec i¬; Eit E"E .. g "tý E

ýi EC °7f" Sui" e : EfE °"f .. ¬ Y

:f t4t aý ý iFj-'

E6ßS is E S! °FEP¬°t{`. ¬° ý¬ß;. gE: ý :E

,

E ýS'°

{'" Ei icßyFrE¬F":

, -t. 't `ýe¬ tS cý "ýeF °_°

a ¬as: t ý F E' °ý E EE` t, r : t 3 ! ¬`. S '`i_° E FE "

ý . F

p S{Eä'{ ö¬ °-{¬ .S : eF ¬E;

. pE" liSl: .. p: cr t ý ¬¬

? 2; s tF { ý . S! t

¬EF E"'!.

, , t F! !! "y"°"" ESi ̀ c` tý' "L '

; .! ¬'. p

e Fi`"(¬ ¬ S " ýý'ý, F! i°{

r ! 'f E. ' '

ý° ý"� Fä s. o ýiae " ¬:

t='ý t! Ep rt= ý Eýi ° S

eSej E¬;. c ¬°. ý Fßi. r

T 'fE

P°.

-ä

i ' o "i t" ý f' F ýlE ¬°l

s¬Sf :; ¬e"! -r_

ti"SEE _ "r¬ '

E¬ ! ý"ýttc E.

ýEE4: i ° "

. ELf jr, f'

clF tii ! Fi E ! rli qß

E ES "e, s 66 "E

°ERß ýEýSäEjb° ý

'E {c 3 i {e tE

e; a 3 "EýýS

6° ¬ä E{ f E: FFl;. a B 0 Eit:

: ti ? t r:

ý .,! _ E'2t ö"ý° "

9t q I'{° ý S yL ý'ci rjE° ;:. ý ` "ý'

p ý, ̀ +. i, ,¬

e ¬c. iEE E b? f- , c °,

[ o' r P ýýlp ° tt; ;¬ ¬E ¬gt

fe eý ;. %ý ltts. 1

' . c ¬ E, , ý i

; EE . '; eS , 3' - ¬ 4! ' "

F . . c%. ' ° t ;

ý ¬'. a ýF EEFF E : t{i. 'e¬; E r `u ¬ °

ý t 'ý"£'r ýu

ß A t o"" i'ie % [°. E_

_`S°! c'iE. j"Y ¬¬ _E F °}

} .. ¬? 3°E°°L ±ý , .Yo

, [

S Ir t.: E¬r_Eý ; °p

`"t

% S S-"` "E"eilS

.:. ! .

ýiSy S¬ EE¬ ¬ ir `

u !r E¬ . G-`° ! '"'° ý

E. t cF

¬'. c{ ' u

'fi 3°6 °Ez ¬bný

- r ý'"

E, L 2 , ýýa . ._

SE¬ä S'i ßqS "S

e Eö¬' °:

3l %-. -

: '{, 4

E eSI S ý, E3E°ý !

iE: " . ý;. F SE{E ¬L `¬

E 1E `lcf. S!;, ', ! "r Sj Sj, a

cý' E ' e

! ¬t tF; iäS.

L -F°

la . ti tpe "; ¬«Soi .. e E.

'. e r°. Src.

td ' °L

"ýj.; rýEte F cE {Et E. ̀°! E E %

i"$"i ' ¬} t-"Ee 'E¬jý{S¬+ °

E : ° E2 ° ýý Fýg¬ýS ý

3f : ¬E ES fd$4y ,t tS t: E ¬

° ý S¬E !

ic t :! ¬, .C {ý>%

t c

, ý . ic: { S' 11 fir, eL E"

ci ! lEG. i-`

°3 iE° 'E `

,t sf SE '°"Y¬ Syf ý rc E ¬ ic:, ti eD °

[ ° ri{ E" ¬"

!! ° { E Eäý"

E°Ee"[ý . F. i E¬ ¬l

_tE ti7 Se3iF:. 35: E}[".. ¬ä': :: ¬f F°lEaf+uf.

"! S pp ???? "SE°u tti aY tSSf ýf ýý8SF7E

FI ¬°S °. 6. iFi

( ¬t p" !. ,, iýf ! ¬Elffu, 6 8%:: '. t %FS«u sED

y{ E

'¬tEý°S6E 4°e i! t'týai y

o°¬e . 4Fýý7

s -- t i'l. M c1. n t i ý" .E tip'.

ý S[a q

e 1, '2L

o! S

3

3 Epa"

o=5 tPE°

[-i

° ö E

4 4

S

`¢+{ ao. [

£

Ep4Ii

°{_ i c E +ý m : 3ä B'm : 3äm 3 a : 3äm °üi

e3am R f p

j0 {

S ä e

ý Ei 'Ehl, n 48

ýEöE s

11.1 N a

y iýY m n

is.

3 at: E E

BtcSiEE

iniEýi

I c{P -Z

SE

Ea Fý E cq

; ýE

5! E ° 'g p EE

°3EýF ä: 9s: 8: .i4 °E2EB 'D f f

_

y ö E n q

q

° a

Ic

E ! b

° E

3 c

a E t

ED c

_

S

ý [ S

a

,

` i. z <_

z

C CV

a) U Cd

C)

0 U N

N .. r

bo

"ry

N

CAS C)

E w n 0 0 N 41 N

U U

. L" U

LU. C) 04

O

m- ý dý °c

2E ¬veJ"ý" - äßg

WLLÖsüýä

"ýga Em-uEäN

E° ö Äý 'ý° 5ÄaE

.ý ws tEw

E

- ö cyöu ö5

9 Uif_ .

-So

° nEö9 ° L^

ý ý^ öý Sc

c 15 -6

cti

N "c ý

E air g, '[s rn°'

a aý. °S ma IIý ýü

m°i ad

d H, ü öý°'

v"i W nýi vmN) n ýöaÖ to

_

wýö mf/1

o O ö ö [i

ä f7

,p

N

ö p

cm y 13 Äý

ö ým

`° 3ö v

89 ýb N; 9 z: i

V E 15

E° E

_ 6 Eä 98 _ .3E6

ä g

. +c o

a

rn

E

o c c

N O N

Qa

aý

u cC

a)

W

0 U N

'b

bO

Co cu

w

n O O N

=ti U G)

U

a) a4

° ö m 0 ö ö ä

4

m n

"'

0 ýN ý� N

°

`: Ka : EZ ^r

p

ös fiEcE 9

ýý

6N 4iß

c« EEý

Gý

c ä'gm g$

6

F

r 2v ErE. ä : -c i`öE. °

°i. F aýgr "ý" . ö öý. s. = c °` i i3 ^'yým P^ mýiö °

ý ,ý Q 5`äg°ý£,

Lý E-'fiä tlF ývöýu'ýLý

e-62 i ä2ý:

o. ýý t

; ýNWý

eLmn m " 5. LL

rpmüHa^ ý3öEi oo

° °s^. vCFcöýc ö^c

E$_ýmrý°E' m

,ý

#° oytQ

e ýFiýöEq ýö ý : ý�"'ý sSý: £

°o ý7 _ =ý ciJýctnEý°ý

: ýýb i5 rcn^ m

m=g$ EEc°mrv3-Ec ä ¬ý` 8

cýc

°ý'aL E Eý

Ev. °

ES `

cý 8ö5 5v _r=WýXýý urB=dL ä` -. FC

8ö g6 g 6b ° `

E"ýY ^ "r äý

E EE

Eýö = - ^c

c e :o 'Er ý'z'o: rvo. em E

c c ý ö_s SEmg'ogä '

n o e

. .. «; E

gý3m°p"°ä. Et

ýýc c ämäc 2 r ö' -o

" cFý:: C pX

E rýikýc°g8: g ° F

s2, 5 mö =EE

röE

ý °ýi nE nc 8e 04 i-$8 Y bo8

x q 64% t gc

.. Y °c

zgNOý° ä , t ° i

. pEli . o£ö. ý.. SS F

Ö o EB v°ýüsm `i SE

cgvg c $ýü = " E. _ "_E°EEäýoo¬ecýý :

ýý -ýR°-2Y sü EQaýEEL :: g3

ä k ö'FV L. =noc' g ; eq ýý m tp.. ,EC

°ccL-^-a .P ýr 8m

ý_ , -yf ^o

5üyýý2 -o

w.

o 6 c°opf L, sý^

p rcö -1 - EEs ` s =

ö ego Yö Fg e.

es°jGLL v 1 ý 2- 8 c °_Cr Y

ä^° c 3ý « x s-

eö

a 1 .ý z Cz° `ý ä °

; °

Pc F i po

E 8 ý ýý i« ý oE w

4 äorvC 8 aR h

iiE : äLL Ei" EöSs_-

" ýý

ý si SýFeE ý

öý-2 o Eä f ý.

@ °. cýnýüý, ö : sý'

ö^ ä: ä6 qNýc G.. =

8°. ý Ec

gE =E týýc°, E ög'r §

`Qyý'aýETr i. p öiý °ýe w� mE

ýgnýVFr, - C.; wE .

üeß? a°, c o« äEe

c .°

eb b ýe 2"_x=öoS=cE LT'"ME Ea "

mö n. Eyc 5(7Obýz{3

ä° u <oýýC °cr ncm

mýiýý ^e4

$L°8-c ENn S FE. `eý8 rý°'ý55_ s`fäýErci °WýL. c"51:

"Eö2L F� ÖKOv '°s , 4 i8 I

ää Iý". E L° im

@m gýN ýiv

o °Ee Q' n

C 5

°e Fm N YCO

ä ýFö

ýR ZýN

ýEö

ýýH

gFm ýg

C m

ms ms v 1}4ü

_ m° mu

ma

Y'. 4m _w

mm ä

Il

m ö E 9; W

m 00

g Gm 080

S v nsm C öm 4eYm C ýYm öüm

m

N O

c ÖH äR

Y Y Y

°

K

ö

K

N

n

N K

0 ^

ä

KÄ

N

ä

N U)

U' "

U' m (7 'S E C7

ö -" ä m

me 4ýE Ecgiý =ýö ^ nýF 6p

Y8 1. &v. cü

Lf ý 5t 'H mCý: E ýäý6 g2n

ý'ýü $i ýöF M °8° ý_ý ý5 5r

o°c c°

g cs'Eý' ö3ý 3ü8 EE

E

ö N^^ 5F3P ý. G c Üc

Kö

3 g

ö m öF s B

ý 8 i

I Z ^

M, CK 8^ r

K u7 C ä °

Ü

0

Ü Ü

M O N

0 CV

C) U r- cz

rA

w z v v cd

2 w

a)

C) ß

vý e o$ Ne

rZ bo ý

cu C)

Nq O Ox

4 0. C rt

ý E. V N6°:

LýLLý

VäE. ä

GEäý Q 2ý(ý

öm= U;

n 6

= 6

ö 0

ä 6

ý d

ö

wßß

s' -

zs

v ;.

ü i 3

e ö` mN C: of

q

Ob

t, p =F'ýEm

Ö ýCm ýý°0ý

ýGO.: n.

°. NA

°N

ev ýrv

NA GIV S

Nm

do

.n Nn Ný pN

Ö"

pry

! : tf s

""s-g8 'ýtl -:

ý""amt ý"' u -i (

'Ee

?? 8d E `e po at-jai ý" gc"F . i.

>=°3 vö i. ° - $

fsacp-; 8 3 ä'6 ij

!P ; gym 8ý:

sQ

E $

" f S ° .

8ö _s S t ;

ä i

° SS m'" p o n

: -SrS i8lý

8 gm S of

! °-e. °isätýFý ö i `

g ý'ý : 82

: Fcý a T a. m. . . ý°ct. E lS: C2_ BE: E`R c. p y 8' E'

b y E: g2ý8fi eam °

: n. .: S"ýe ý g '

8. ýeöS

c ;F° ný" 6. ýi i 4f c o_F "a Sm

. ný"i 2

F : ý: ýS

°[>! m. ""fi_lEnE t'"Sü

By _f! d

ö" "F. °.! ". °L " F , °"Eý'tF!

cn m{>ý'

. } ` ai lmF ý° E 2. -3ýy -

ý} j. mj . 8"" , p ° :: b _.:

p : ai fE 6g ..

°.! ° än m b ' ° r

'Ut 8 8 =

e .= E::,

"s 8"¬.; ßt8 ää ýa¬ `! 8'S ° ' °! a: ° 2 dfm " ; 'gmlE

ö::: gi 0ý sF°s" mý [re

"

'°ýL

m 8 E c i cOy° f

ts'ý`

ltj0"

cy_ ýý

w"" ce: ý: v"-i 'a !ä m':: ( F. S`

'" p°! S

" B

E: e'Fe <" dE' °!

li . r. ° p

PIS ý L BrEF

Ejýj , iýmý

ö°e E: 'oF aBmý Föel. -mF : ¬ýe: t> E¬F' o ý.: ýEtliFF

t« Sbmý".: e 8! f< iz 'L "

ü! ° ° ° c'¬S:.: t"

`

8ecS4 ý ýE

mFý: ý äýeFe

s - aýF°` :: t° gý ýr"` e0

m ßg8 mg äf Öi "; i" E, < E i"ý. 8ý' : i¬ ;_es. "': S: .f ,

ýý rf net vt`" ' F

e: fm ai filaý'm

c

FE ý. °8ö tStý

EF "! w ¬I

iö%°i'C cm" iiy W: rtF ° k

- "" ° ým ä $c ""i Eö". 8. _¬SS

Ep.

°i 8° L c °a 88"'C ö! E, d"

s Ö afý8 . B: i° 5` '

{: mY

.v .

>ti=§E ! Fpm g> . ;Q

cw'°t! u i "

äma4 ? - a

ý ö Ly °

Bn ! y E'. m° mcc ý c -

m g: ýSäýzg 6 Se

a 2 q! ' .(

ý°

v c RF °°p08 R J I U

c @Fet ie " '

°lQäOc 1S g5. ö. ýg

iä tý : m"84 m u iEEC 4g"_. um ,

° mi; F m

H { Fi {¬3 ¬ T

21 c E3F m p - : ýd gý'W mf -8.8

: 6 = F'°

"tb: 3! F F-`F 8i'! "t me., al:

! 8t ' ¬E FFa ` 3 m ; 'seý'°".

f P '#

H wF 5 týý

RSmlg p - OFF

g r fl Ea4 ä¬bp : e. o ,t p S-E °r MIN ". ! EFL[ "°tý8

F1F :. m[ 3E: F: °

° ty

,S °mU; "i ° !! "! ¢"

8°"c 0 8c_

0 iä8: inign d`mSEFääüfS.

äö wlý: ýs F88Y6fsif8 il: fix$f4: woýoýeEö

iE E

ývý A

g

N

Y

t9 i°

imö ý. yn F°

Sm$ S: Sn .m

?c sn m° g

8

ý 4 ° ° n

Fm äö 3cýn

Fm Fo°m 4cEn

mL ß(ä

; 2" 0 im 4ö cDö zmm p j Z!, Fmm

mý" c c om ä 3äm e3am 1 ö m om 33vm nMo n Ä ö ä ü ä m -

ý at t

rý° mi;

7ö ö J

m

g ¬rä; ; E 4 iiýiFf

ýc3i$ý

d N

. N

r N

+ýe: m g

S. F

Nm fý N

v p

. Eýcö

,F ýäe

Fä g Q lit

°S4ýg °p

c öeý

tF

ýL

ý

q i

_ qqqy=

cCb cg °4 ¢

ESg °

ý; ` 3 ä'i¬g-$ ^c4'

to p`8e

m m°

' f ö[°ýÜ E

g

f=¬

m U 3°

tp! S ° °

° . ESS yam

Q2: 6 c "ý öS ̀i-mä : Ng"r, " ! E`

rS :

äe ° ßa: 5 tä. FW

p ec i ¢Fý;

! fei m ngý

a mKE ý. o Ü + ý

go E

& cö ým ' g v E` Y8 . f. äý;

; g ° m °

° ` F g . - m °8 e= ýg aý gýý fä " ý ýa

Fs8-t_ g 88 E -! S °s !i E ötým

c Eg öV ö "@8` 8, Eý °

ý iäem - ý

s ýe, °, a - E og

ý5 c'

I Emö üFE .B ýi . md_ 3 0 0 of S& tBý. F Ö

¢eeý _FF L E

L 3 g e

s m c s

'-ý yam Q

Y w 3 8

Ü o d ä m c i

S ¬ ä jg

6 i a 4

S ,E Y S s

mm ý

' Ö `<>t fF F Ö g Z $g

o s

i 1

C) N

z V a) U cz

2 w

0

N_

N

rZ bo

G) 2 w

O N 42 N

Q)

4^1

b

O p

nN

'

rr r

2 =c £ ä Y r :K F F ý Se; =E: c: ü e-ýöý ü°. ý

-s d E 4ý 3

.1

Ti5 ! oe ig. °"ýöwýEEý'aEö iY~ e"Qi¬2 _ _ E 6 E °'

f `" " ý .c aleý ýö"S

T°, ýo: -" 3 ev°ü"` ::: 5"ö ° oö: ý ýý c 5ý ,E

e° F "G ": ESl s -`: clýE: 2öE-"02ý; °£ SS "i Vßä ä}öi°"Sýýi-: °° ßößä ý'3: cr "g 8` m . ý1%° °9s°ý` -" býä: äöf= Fý` 5 <: s CES s` `" Y°F ýý` "

ý£8 : 5ey¬°g t_ $ ä üý'ýýi>=Eä äiß

e ýd Z-"ß m6 DeölE L: E! . ° e! w f

. ie�5 .

nügsýo-ý'.. ö '

i°>qa'£: ii ööe"mýai :ö

sl2ýýý7: ýý' °9wýc°? °`l: Cn: ýiS

ua`3 y 3E wÖä. ÖE: öEü65

E ýEF5 Oýö^öo: $ý ' e öFE .... E. ü°c°Eö

'öö ° Gýr2

" vä

"s. c"ýEw: °; 5ä' ý ý

: ö'E-ae° ö'° : E

° «F cEý d: ° äý: ý

ý° ý" e cElgs " ° c¬ý t> ; S ý E ö" w

: di i sey: gý°. s ýn "ý - cý cý ýYäE ýr ýe: ! TEl ci gi

"c E g: ý£_ý OEZsEO dý: c° °

o ýi _

c

EE ci.

c SýFi ý y" EP

"" ao. ; ýZ�ýQý °s .

'.!, °«öE ° ? 5l £

öý ö: 5'F

^! ':. 3ýe: i° ý

C ýW!

2E uö<¬: v . m°

%

e aß;

c ö

Z w cä$: ° E. i

E E ä" §"5 "e:: göL. Gu äs«rE

°° 5 ý ^"_ ýw : öäY£ ýý °. CY 2Sä

9 s`ös =ö : g¬<a"eq2- 2 o e%"gip ¬s f: 8: öy ý; gýäö

°ä E g; 2 t¬z".

°^ö2ii°=- E° ö öý°ýC; ^ ,S ö = ,

ötE L

e ` `! O t E ýEEýs°'ýELE: ce: EE EEn c2ö i :7E iB __° SSSEE" `s9

EWÖ £m ö

fCjN

ö

CDN

^O "ý

ýaü u01 -moi

öäm pof 2 3a

°Da mmc ä°

97W S> öY9mf/1 3

o°6 > eivmN

p ö ö

K

g 3ý

°ä c c

ä $

n öm F$"

cc $ °5

E, °, m E, °, m � ^E.. ° ý£m ate

2 ýc SE

ä? ä c m 2 -m3 i i5ä fEä EE

b

e

a

-ri s E

cc

m ° ä ö °

110 C) N

N 0 N

() U

i. +

N

z U aý v C cC

b x

bo

-° 0 0 N 4.1

Qti

a

4

ö 4 ö ö

ö yý

cE

°b ck ý

LL q

z li e z z z

EIS o t ý'

ä

m N

° '

ýrým

g' dN

4

ö IS

-82 rn

ocm -o

NdN

4 rn

ýcao

p0 gig

d4m

ýc

pö d `1 N o rn Y

ý rn rn :

ö"iW öm wüm wäa ca öm ti. n'üm .1 812

wüm

öe N ae c .L ;e öe O iý 1ý V V O

6 d

N ý

6

N

d r r

Of

co X1'9 EO

En

öa a d d. ý

ar ö .2 o aý

s . _.

W E Q'

Jt Ö.

Ü

äo ýNý;. a ýqsc 3 ß ÜvE N `5 ý3 °

.° Ü

.gc .n d' o

E .. c

Yd v

ö

ä ý

ý

S

Z = K K

Y I p

Z

LL

y ý lOi

Ü- Ü

Co 0 N

cu

C7 0ö cu bo

a) u czi

2 1° at ^ý L

0

Vy

b0 i

t0

w

N. N Oj O N u_

y Ub e . ý4 Ni« rv ý NOpý E L4 s_" E

U E"E U maä: 0

ýý a8'E E

ß. ä° : EE

Qz 2

C

;. g n ä 0 ° °

EEP öý8i

öI

Os zF c

ýýs i,: JoF

n-öP¬ 8 N

wöF EF

ý¬ 6

OF

4 "ckö

et. joo

mý".

6r G

N

6

.ý

4Ö

e.

Gr 6

n

L S

c5ýz'

ý

° ýl

'

`e:

''

ýQ "

"F ißýýC`3 `

ýei I$$ $+

"Y+ ;F

is-"ýcc £8 2

in%' $ : "i¬°"£ °F Y.. $va

c`i i

;: ^ S - IIý£ ý

b r3 ¬Bdc S!

S °L"i ° p .Fe:

.: ji"Y`Fj: ". ýY

ý E 'ýýiQN`o°rý

$$$: ': t p

$"OC

;

:i

i e " i äß I äl:: °$ ' ; o; {

f. g: _ ýýý{":: . 5ý: lF°8`: t

" : Faýý ¬ aý. ed6 g° :} ý7h °

9t=. ,,;, ai

¬ :F

-. Bg s , 3`$- öp: ° %; S »5:

_ "°eS° S; rj$E t

s°. f 8j 8

` . !8 ccý !ý sL'°""ýý: ýý$cj =Nf'8'3: fiý °SS ä"' ° : : " c

n

"

ýc$: eSii$ "ý ° F "

4 Ü=e¬f

äýýp: sSEýýý: $: E"ti "s °öei

üQ$E: ýacýaää`( t ög. y f'' F( t 8 ö: $

. "F5j %[Q ý

j :? { ý C

¬$ @ ' ö? ý': f ;S

ý ý$ý mýtf.: ya p ý: $ cää

ßlä ý°s; ä" ßg5 ,ý

i ',: $ý;

g

Ps> a t' ' j :

e °ý(`ü»ý: P 0 #£-?

. öi

ý : aE

" Si"Fi ft p"! : ° =: "

"F `iýý. i eE. C. C {e`!

Ec E ic! lüýi" ý : "

ýc $bj

F ° ý " S o " :

! x°

. : öä" °f p

``; ' L

-_ : °Ec 5£

a f : F'u ¬c`E

0` :: EýöýBäý""°' y : äEE$ l: cNa. c°. ý 6

°-ýt i°£f" ý"$ýýýlF ý i , ¬Q j! ` i`

ä

cý°

-'Ený ;t

"fi[ :` örcEý` r°ý; "t i° :i °F

;s _ ¬"

Q °f°cca. iö f : 4 "

" . 3!

tq'rý$e ' j` vo ö 3 a:

' i E ö Sf E %

i :" £'E s F ° b' "° F$ ° F : 6 $ i

0 a: $' °CýS 38S& `ý'E ý ý: $$ ý F ýF l: l ýs

Em° g 3 Fo ýFg c$°

°o0 8 558 ý2°

E °d ;:

ý: ýF: e ýcäö ýiyn g i ;ö . v- HDä

Nii. L( ÖN rN i>Dm i>9m iSrÖN m ON iýÖN ý>9m

n e0

0 N Ö ü Ö l " Ü 0. n

3

0

i'sE ý m N

H ý iiýE

N@ m^ a

q ý2"ýý[ Uý°r Ön ý x °! äS6

C.. 5g T

ý E

p EFe : y%. F8C $ýý

4Y"= sg3°5 $ ýc ö

o '" Y B

g F. ie

: E " ý

$ ýC

181f .a aw

äß: S ?:

Bng: ý: ice::

ýggP

E", Y $ý

`r ,:

'fe Ys¬ý 'ý n ..

»fit 5Pý ! 8: iS

" . £ýýy $Eüý 9öf£ ý£ öää ELýi it: °t°E ö°$ °SS" i "y

" ° ° ý fir. 'Eýo" .ý ýýýp ý°

ýF F »s ýq

pý EoBrý iä z E: ' ý ýeýsi : 5: 2ý y 'ý`

' s ý c

iEi'° üoäE:: ý: SFi i `=ifs &___ ýSc`. _ .

ÖS äg! e s mid

! e ý

Y

E i ¬ S p I E ý

i S ; E¬ E y p s

0m " Ö d i6 F 3

YO s" '< Li

0 S

c F N

rn 0 N

0 N

a) a)

C7 0ö co bo a) a a) U A CIS

w I.. a)

a)

h O U N ., r

Cl)

bO 4i

h C3 v

w

C N 4-1

U U

. Sý U

v v a

d 4

a a G

öN ý, L°'=äm ý'Ed

4 Lv ý $äa3

a üý ö ýC D

" ýýnäý ilä

ESR u£

c äýý v nrý

öEvýNý'°`N- $

ö äm ý

üN -¬ ý' ÄV ÄEmv

6? ü tq-i c2q

u ä`Y-' yý g Pi

t u

ü (l L' 5

LL i° m 2 m_ axi ö

2- ° cH ý L 'g-8

*7 9 3ß TSt

SE °cäg ý"

, ý6 E ÜÖ

ry ZL ýö° -w

w- äSm° 2ýy w

vow Iý - aLi � 3ömi -. 1816

c= :C

ö

ýo

d S2 .p

m

6

g a' lL

Na Qý

-$

S

Cy °g c

8$ B$ CO

Eü E °i S' E

E5 Eä E

6 Ö

K 5 E

0 rn

N

a) v 0 0ö

ao

W U

CC

it

'b

x

bo

t10

CD

21 IID

G ö

o d ö 4 ö

a

19

a

°

cL

ö

ä ý

Yp 5t

2; '9 a° iy H

p ' qC

g rv bz

öL ° d

E"^ 2ö d. E

ýý ý- E C äý s rý ý ab 0

qg .g. 9m

aý äc vs

T=F

all.

e

Uý ' Yn ° m it

S

�an -V

, S _`s°ä3H

zj-9 i 'Eeü Ln

ö

a a öý-^ E

C °

C ° vs4 ä_ äßs z ýÄs

ESN ^

9

ö

m

ýö

dN

II d

_

o ýö dN

Ö

ö

Ö Ö

VEö ` lV

p

°'6ö 9N

4m

iE�Yý

W d Im ýhGä

nüm

ý ýýd

8w

ý4 ý

mN°'SCV Ömd'üm y

:2 ý

nr w8o

m H 0 «um

0 n ýv e

ý

d

r

to

N

6

rO

(n Q

n

N

N

CD

r

N . -.

a

N

6

O

N 69

°ým ßä,

E

ýý -" °m

.L ýöö, Y

o

cÄ

v y_

ý$ý d

mm

° 5= m9m

i ' Ev E2 m" a a5m�

F

ýn2Ln *15 mö

ö C$

mEö "' c m

äN ä$

v" rjm

a', iJ

6

Qq8

3' 6

a E E

ý"ý1O E tn°¬°- ý E q E E ri .

I vein nr°-__' ci ýc« cc

n

Ü 6 r y Y6

'm ý

y V

u O ? ~

m S p

U A N dY

E

LL

m lL

z 1°L

C> Ü

N

N

w .,.. b cu b

cn aý v cý

aý

o

"0 2 NQ

rn C)

w 0

O Ox

N tý

.ý xF

C Ngn: U ; -3 ä!

EE

$E QýFe

afg` Qiýý

7bn

n ö ö

EFP o ý ý ý

Oil

2E

_Sy F

,

cc gsis ° Hi

" m

I tl e n

l fl o n

ý M

:. ý

ryP llf

r 4ftý F LFýöýEý 8 B-d. i; { I{ its ý8z ; °i {ib {ýsFF: Bö

e ý8{i p t''mBöii 8°§ y°[ nmE i8 S :

n=

? EF

ýF6 ä{ac° { ob SD 87° =#ýi{fi

ý i

g

Ö ' E"aF `8 ti ä m . f

Scäf ipey"i gis{¬poý n

e `

'FEe°Ei ýTý'a{ý 'i{i ý8

ýuýFo p ýW0 "

ý+ ßg8

n m d{ : S. Cýý'¬

_ vicm: a _° °m>>d: ý`oP mm ; L

:ö ý; ;a

= "e°i { 3 6

::; F {{ýsr 8". P t! {8 6' ° ° P' " . Ft ° `

e ;

"_ P Et

° `° 8p "$ 8E ý° ytbý = $K8i°8i? i eE PbE E 8. j v:

.° >ýý''s ýt L fý

ý; op a a°

S ý$d" FF E 8 ýrj

_ tF2-{: 4 ! tae f $ý: m t{{" 4¬ ,

° . 6s. aS{p oei

' _ r me., a , 4 "; dPýti. 5 °'»$e FS° ?E 8

:. tsö: G¬"" E

eia , Smi . F; egi " { ö

F! SP" ¬"`, i{. ý; jßiF " a ` .1 Fö

: [" a65 iL: m3 iim-cFýe

ä[ac äbs_^o-: ý

!c 0i"a ý

[`sý8:. lý ýv @Sn F o: i ýý8 i S'Y E

° °t88"; o5ý_m O: S£ö ; öe

F ü: Q 6e

° PE: Fýi Fi: [ 1585

S ° j

in! {SF' E8 Fb bpB [ c

°ý`OÜ%F{ 8 {" ý.

$ E

8=1ýFEe' a3ESS Ei- °°T : r'=d6cäaä ¬- S ö"°[{: d : "

a :E {: 8Bý"t. `- '8 , £¬

. {{ '6

ý{ iÖa"1. [d8ii- 6! t

tý "iiE ýý{$ ý 8 ' ýý öýF

, pj. 83.

iýLB"ý- ý'i: mca

-E Y 8ä EL

` 6TS8=£; nä 0ö"ýn' ýmý ;E

FI { . 8a 5f 8! tt

: °FIfF ¢ i:

"rý ý " lj0 j ,f fý F ! ý

{ t[i `F8i P! o " e E c°8

SF '£ i 88 ° em'"S F. =Fcros 4o

=$F ee ýE Fm Fý ýSS {äf . {8:. 3 ö F

8.8F°"a F°_fa8 :"

c° ý9"' " ¬{ ! ". z < S. Is ýEý t"

F

ý c

ý inYlýfii' "

a ` ge-° gf idWoi»F ve4°c>dä_. t sö `

"S F ä: s

. iFý``: eF ý9iiä

`

b ; EBEý:

, lFý; s; t;. FB #ßu6". P S Fm {sý- "-=t ; v

o ýa ie

[ @ ý -0 cmt 85HC E-E P. f =mia

°jý =° a

. a..: ''i ' '

- !F !t of ;. -.: °

P ° s"n< { 8

m p s c vF [3? "S" [ 8""! I `ý "j v. F .° 6E 6= {

E tö ° "i F o? ä5: g-c

°I8mc9ým _ ci°°ä°c '

i- 't. :.

{ß. 5 t6i

; a. °"E F $ S ý g 8F

.1

° ° < gS!; BEdv`f

pa. '' iii ýL{ý - 1. m

'°

0 v. "i° i`i äý ýä8: mý

ö MF FsicF to F! °°;

"S ä y j2

a;

s .ä ' [ ýev! " ?m

1'ES 8i8 dii E vSC° . ýý mim wnU

kü ä:: S äE aS[______

0. ýE -S{ I: : FiF ý:

F a US ¬dj0 . 1. all il: 1

_z Eim Em

s Fm A.

Fö iP 4: $e n

Fg ieP

°

3 i8 Fö

SF$n F

YE$ E

eFö ý i Ji Hit °

3FmZ Ee ng! a 3 äý g mýi

ýp$m Ea$F

wýý `<e öm e0 03ä Ila e3ä0 m om : 3nm +3a

f F F , Fo e ö e

m

n Ö b

i° e 8 3 n ýöz b

4E g P ý " ý

ýx`F8g m oR

o sa LL

° zp

: ¬fl4a p LLp ý

'ý

d _m

, ýr ;a

*

$f '$ s ' gg

ö

M J1 j

ý8$1

gffl 8 g 4 ý i

team Erej rF °ýmm o s' e: c e

ý FBu7 g: ý°_ {

oLat ° L 2dF

S dgF

0 mo: w 5.

g8"! BE : Rgs". s" EL -- -

Es I r ýý? ¬¬x " °

$ý °"

ö: sý "

;d = a £

w e: ̀ ý

. t ýg. ö $m yg 3r¬mv

u m

öýogZ [

8fl SS

!! °m e C

E "ýF: 8E8g ` i -

Yi mf yý £ý84 i

Fi -cý c{ Fj mg :Yg

S e

111i öy" l `t FE= m"F_ m tE _E 6`vS üEEae F

! E f

=[F_ EcB S. FSc'ýF J Cato _F8 3GE

8 3

e 5 S

g ` S

e 'r

Y ö

: Q 0

2 S U

0 _ o m

a o

:

s o

ö

= mý

-

YS m

PL

m

E E

p

t Ö äc oZ

r

,Ö 'i

v

m{

ä

M r-q N

d- r-I N

w ... a b

r_

U U

td

t0.

U

U it

O U

b

x N

bO

N

Co U

t° N- 0 0 I

ö 0 d

NN

r (pes N

r6r

ý

üÄw°Äpý °�ý +ý+ d

ý rO

N 6üNv

`q ö

G

dNÄuüv6OrLü . F- c0 oN

. m>> i° 4üTäQy,

pS E

16 '- .7 co°

, svm. - A

.27; eö OR 612 24 ßäo öäß' °c8 aru °E vüö

v ngdZöä r° üü. C2Kümö v O' w O ý N

N '

5ý vý

:: `E9gv ._

c '°ýOC aN üý`m E° !

; - Y°e rn

nqvý°äS°ýý 4°=iödäs

ill - e EHI&M 2S.

ým. Eo alma ýE °uý'i°päýýs ý ý5 ä i°ýs ýS°Ä 1Oa S°äi z

.C

ýý ý cý Nr

2CbNý >CON

U °. mD Wd da druoa Am dd

P. w N

-ga 4 . 3-9 Z(6 1 3v`mv6

°e °e ö

p ýy.

6

E

Eye ßöä. ýE

v "¬

dN ýEE

9

6 "3 y

Y R

E C

ý Z

pl

C N

Q[-ý

= N

'ý

ö

N 7 ýO

In

w

a Ici Z.. cu b 9.1 cd

u ctf

w

a)

a)

N 0

. -- U N

'C7 N

bo

N

C)

4° N- 0 0 N N

V

b aý a a

Q

O

C; ö

ä 0 d

ö d

Ö 0 d

G

a

d 6 1

Ö AE NEü

c E@ Y°°-

Y

E5 ö

ü fl S

E°a 5

ßäs»°

tai ýS °L 3 z z° z° mö

CN

v t7

ý

i ýý

N

'' i

p CAN

y

p NNdN

y

cw d

try

-58

tV

i 'Y

u y u H

4 4

M

4

V 1 ý" Qu C

W Öm O N UCD vm O Öm

Gw Um O

ä

Um

S d

N ý[11

e m

öE o

öE 0

e2 2E on n

ý n

v

n ý'

K 0

K 0m

6

K 0

Nm -

6

K 0

2 A2

It E .

E. 2ý

EQ c2Ti

ö i'

E N c. $' m

.E ". 6 ü 12 ä t1 Eý ä E

ýn Eý °¬ ü s"

Ea EEE Eý; Yý

ÜäE viim" `c_ ' L'c ÜEch ¢ö

yVI N

C ýG y

d

s

0 12 aý a

C> C)

N

L

b

0 Q)

U

CC

110 Sr OJ

au 7 c9

0 V

M

b0

i. ý

co

it

110- t" 0 N

-ý Us"E V LE: d

x; e 10 cf Eý. ä5F{q

Q <Z3( Qii<

1 n ä p 7 ° ,,

'BE

a3 Jill

o! ` z 0°E_ä4

t. E

P E. n in

'Y °!

PE °! E? Y { ga° gF

{i G 18:

", q}

Y °i i i. E{EE ý {'

ga .{Ed glF. i{ F ýE«

c. iýý ýY ° g

r[{{ {F { ¬YýýjI ddS

: ý ° 3y. {fä {Y

E«ý' t'tF`

ý"+ d{Yiýýd a

ý°

' Y

°

{ "6 'gg{; ýE { b E F g ýýý$aý-°'

t:

F

g{' °Eý «Y. °. «

.! {. {. ý iF {ý"{

`"°Eý ? loop

ýE"yYS tFe =. d't"I Ei

={Iq: = " ýE{

hh

9y ý OE E SES a ^ pEý T E. E

E' FE..,

ý

111 ýY. °j "'{"' ýýýý3ý 3ýý

" ý O: «i£jgd {

[ ! FE : 'Y FS{i ý t ä ä5ývt

ý ýt E.; F3 ySglp,

: YtFE-'ýýSg3g ¬F"°.. : ýY°g

i 1

ý. y

ps :{ý pý° d

ý

E °{°[0 14vE§f ýjý

{ ' " ýý{g'$ePEý¬ Y' ! °

Eý. ý°Ei{4 iýYg t

[ pbi Sýd'YýýidF 4°YýfF °äF10 6{ g¬i a{E'F°

f iiYl ý;

¬ j c1 VY E SSSp

F

: f" 'S { " "v E

ý

Y Y« ' ¬ bR ,;

�ý t

$ E. E Y°'° "g 6 ý _ ýý"n'u R [tE i

ý E.

: p. {IEaý4{ d avidt' § g' : jtggsi{

'-g.. °! ü8 ý ý

« t gFg`t °{ .

ý. iE°P 4

`V" ý( tpv Y{. dt[gYä

. E{, Y . °Y-E [0

; ¬a i°U{ '

° Y E , '-, {f iv" ýi¬gý

ý3 Y , t Yf"p°{{I [7p! Xa ttý ; Y°

FýFf ý Et2g Ijiý Ea di°ý ý

ü < F

: R`t. ýi(«ý; dS: ý °3p c 3igv d'« d{

. '; 6g'E{3ýt {! f , {gýt Y°ýd{fbYd: °« ; Fg' ' [ EF',

d E'"'Y aýý d i ` äný4ý {Eýt {ýFg sýý4: jiE

t : iig E"{Sgýý{s, g .{ 'E Y` { z `' : 3d

_ (: ýi«{S

3, ä :gt e F.. t.

° g3ý6"t'ý «d

g Y°:. l3E'. ,s : {"VYý{ý9jöý a`. 4 E {"° { ?! 3°: "' '

i , syi3. js Yp td s t"' 4äz ° ý

"d Y. a s. $gýFäYf«

; ý°ý! '. g`ä' � s118 d{lit({4 g Y IIf s! s° ", i " {tö El if

°";; g "t3 . f3° ° t$'s" ý°'r {''E ' 9d' F

ý Egý . ifgä: E:; '{I . aY Y1 3 « ý ý' -{' `

jt {äFýYII { giI° «

ýY tE ¬ý«: ! dt s. EfY{Y£S] is d r t

_ ; :G °gYýr. , Bdc

F e' FFd g {., ° ý`

p 7' ý E p° t

° iýt0ä{7Pýýy. t

.S Y ýd'iöýi el{ "ýsIS". 'F°ý, ![t

Y 3 " g. { a{jg ¬«dýýF

s [fg

p{ EY F1

U

g {[{ 3 Fr

g SHd 4{ { "

11 s

lu! Qi'gS d. I{

" . -.., ¬ "i

a${'D'ý{ E ¬ oll 1

. ýý'

? 3Eý °

1 Hill"

''ýEEY. YYS i `¬«ä{l. i6dS d

3EY: i'P «{«Yd {, de1Y Y° ýY4

by{ El: YE3{ . {dveýY ,. S . i «'ESYS. d 1 .

{_

E°a B: ä 3

4' f' f'

Si{R S«ý° f'

Sý{R "

S "

{ g

ß? { a ö

Cg{R

3P

S E

f 3

f Wýý 2U

E(' ýÖ nj SO

°.. S; ÖÖ i; ÖO

S°. g 3; YC 0 ýi r( 3; i[

; >r{ EE

L w nä t

t{EFY 13. y

1 ä

' w ýY{Eý

' C g ( C

111 E F öý ° ýt

: [e°

-°[ Rd fE E

e E y '4 i ýgiaF3Y f'ti; `3"ý ; ¬g; P ggýs [ ;i YP ; F

: Pgy - ý: ý: '111441 "E' s i8ý1 sbfEý g g

,ý .

; ̀ ý aye c °¬gg

i0= °tEd s y4 iý38

o ; aj >3{E

E ýPg: E

n «Y g

«8 FE ýE

° u

; bý{ 89ý 9 . { Et :. tES; E EHg « Y :

3sý{ ý' '

¬ ýýapyE b _9. Sfl

t, , S" tg'd t

: y. ,E

a EY

_°,

3E EEBgt

ý;: ; Fy tEE EE: !

Yii : °g ̀ . { FEE ý'. °E ' g s[i cf :! f ýS ýg öp-iEý E StO ýt ýyj'

E'{ýýEy ?

{ýý s i4` { 'Fpý

-E3ä Ve EYE 'i S; i 3Y '{Fý Eg 3ý F. F. F Jtu [e! cIE ä6E

I ! E "

S

ý 3

°

Y F

P P

ö

, ö 0

o ýö E ° - !fF

g 7

3 Ö ý mý

0 ( F

z t1

n N

00 N

10

O ci

u

a

n 0 0

U

1v aý

Q

°

i£: f"

4 ±¬ 'ä c:: ' "LF

ý"g3 {, 0Y

iEi`ý; 2° Fy °

! ce: -. -

i2° öf f i , Si :

"a a'i

Sg i e l 2Eý tý

i !ý ;f! -z d E¬

iF a

i�

! ý{ }ý

ýFýö! :o

¬°

ý° a E

i 2°ýSpö .

£2P''# ý

" l¬s

y "E'm: eE --ö '

y Y ö i

_ ac ý: v

ESý

r: 1, .; t

y Gd SEsä ý. 1 K1 E¬ S

!I

@ý _ FL L cf :E t £

ý;

.;;: 5:, ö : - IE .p °

ý ä: ý. ýci ; 3 o a: e" ; ¬" ö`E

'ý faF ý; ! i"i

' Ei3 ¬L o ,a t Ll2 ý"" 3°O r' a E Ef n 5`L :

a¬: °

EF .' a: SA58 "Fý`4¬z: :

e"- z: 2 ., a f l

E. 2ý2ý2EcSäý" " , n¬F ° lcES°" a Of

t: E;! ' eEF" i" E° ° ýcZ E"'i 4,, aa ! 8e E ¬o. va

52,

a ý g e'! °- Sä f2ö-" ýý'a°sýF = ¬

- "L 2C EE`: Pýöi'' .i Ear FL ý: """`

° 'ý .

°_ ä c _' ¬

4 ;" _i: c - ýi¬et¬"° " "i " i¬ý3

$S¬ ý'

äF: yý `eý SBý - E

-- "e . c6 f- - 2

° ¬; - °i .

E"°% ELI!

,". iF e '. EEýe¬ý°E

F ý . g c ý-

3E ° ýs '°i¬¬ °{2¬c

; 'p^¬týS" "¬ä 2F ýö¬ e,.

, -, p_. °i S: t : °ýYg,

; � ti - 'E c

ýi`s ö ' T _

öýi eý[2e! ö -

. ; £ßy0".: v-Fi° S °Fr°: °vý. Fý".

2 F ¬ýY t !:

p2£Qýý; F ' " Fca °i: ýiiice

" [£_°_ .! °cý -ö ° °: 32ý

{. i. ä '°"3 fi ". " E""Lt " . cE6 S: ö!:: -ä _7 o_°

pn2..: 2 . ür ýw" =a°iü: ¬"e' Släi¬c iä

d 2 1 ii2 c I"

F PL{cP , S, _ " fpp ä3jt £o

, o Er'g'`ý"M- {ý; E; ="gifE _ _°

°c : Lf°'Eeý c , . °ý:;: E ý. " ro y °

: E ýsQ F ¬L

¬ . -. e l ý4

{ Eic ¬Pf'eiv. i ii°Eý°aE`i- ° : 4 9

3. LýeiE tells u S..!

: "uPý°"ýý¬E"°E

"ä 3: °- Etoc(n

_ "¬¬e = " o " tc tyE=: dG¬£

!" " t

FYj. '°"+2to . ý` ýi '` ° eS: `^ i2c. £eäý'"e

"" °>. ¬3' ¬Y" Fä : gyn. °: E. Eýoo3a

' -" 3 e°

spe _ Eý ° "ü2: E cGi- -. °'- 4! Qv ý cE

ý3 4

: 9f

_v f c""-'f 'S E E " i

öb Sý a-°

. iä: 'F' F c

$ °E'f OU3": -"2 :

ii Ilya. °y öüIoý2° t °p°

n ,; v _cmöpa

E" !° ¬:.: FöL "t4"y - i Ja -

. y 2

: f`'¬" n - a£: °° £rS 9"

"y m ° ýF 5 - ; ""` ý: c ivFvt-E on°-

8? °! S - 5lö: i! 'E":

E-. _!

�0 yaö"S-. L"¬"°. 'c $i-`` _° , w

s;: F{ elzfööfö. °

S ýi` £ a°i ; ¬: iir°{ ý O ýS£ý .

"

g v -t i ý . " ci! " i

f ` l gF :si ie: %°£

äw"fp! F " Ea tE', °8 w£-"icýE °°"t `

Fät o "` s

av" F E; ¬ ü"-I E ems

t lo 'E¬ý¬ö°, `° 2 _"`cf

8 s ä5 ".

° ¬ F c

!- öý° E - ¬ö lýýöý" ' "

¬s': C. FFP''"°" 30"=

. t5w° 1 L'" yEF¬i`

2 s Y¬ ý

ý! P': ie"'E ""¬° f - :i

- ;

"i tP: ° " FE ° i¬° L ;F p-0"° Z

E °°3: ( _ YP2e°_ iy y e° `Pt _ c

_ gg¬"SeSý¬>Ü e_r i=O ö

E i. i" 4F. u° i

2i c"ý 'ý6 "f-y° X2 v it ýfl fd E ä ö: Ei.: fElo-¬taE u p . e: ý t

": ýFES£62 El25. , ¬ 'u .F "lLE 644nG i ö! ýöFü PSFaP:

f i 2 c. ic, i£Eöä £e6 2 : tS ESätoö.. °P

! ýbE 2j"3y SBF. Sa

EO° ý '"

EO YýäRý

Fý gýäný

tö ýPän

y rý nQL=° -j 8 y as- öiW ä3vm0 i; äm0 I MO

Z

! 0

ä f 0

ö m

0

4

N 4

- y"

°

eö 3Ö Öý

fi cä E

:ä 8 36 8°

ý ý ýg

EI z E, °, iE !

s I

3 y

to ä2 2 H fEi i5ä iI

° _0 ä

Y fi _ C

0

a

p I

ä

ö

'

of ý ý -

rn

C) N N

"Ci O a) v cti

29 w

b x

bo

4°

0 N

Ü

"Ci

ö c ö

c 0 c

C O

d d E

d d d

n fl!

o Ä c

c C C

ýEa

Ecr

£ E F9

öEö ý+ ö r' m3

B2

5 z °2 68 z

Em$ oN

tC

Ö

pG. ý

cm

ýdN

ý4 rn

wpmo

ýý5.

+d N

u' "

m ý.

EmO

N py+ dN

ý a'

cp

ýdN

Y -

8 cp ýý

N

2B ßý,

ýn `. 8 ý'

Y

n 'n d g rn

'^ rn S v

W öm uni

8m üm öm

m pm m ä t8m

Ö 8oß

O O o Iý p

ý 'R

d 6

N

ý

ý N

ý

t CCD

m N

((D c

N d

CD

r d

U) d

Cal

. ýr

rn m

ýý d aý

. I. m aý3

8

i j

tsýg q üýn 'ý'mrvs.

ý 'C

V tý

Ö

c6st

N-" ý'

d C

. Q

r

ö

E

GO Em aT ýý ä

E�

m2 ý

; xEQ

c 2t5

Jd

'O ö « . . . i a

öýý 'ýýc cm v °c 'Qü8 EE öö

ý'ýä . co a ti ä ff.

ýEý. °¬cs

§_ 'c°

a EEE 4a .4 cEE °

« ai S=

ÜvE uni m mýý ý3 ' ý .t

Üý cH Kö

-

tl g'

1 ý

ö

ö

4

-7

ý Ö L

Z K dC

UL

K

CJ - 0

N N

'

cd

R. t

3"' ä

Na

i

N

Z Gn cis

c, ý s

Hi OýE O Nm

v' U

.U °rv

U U": 3 E ý; EE ö

s; a

8 <

Cid 0. °b i

QzDi c

ö

° " N 0 " Ö

ö °

ry ° t E£Q ö 6 ö 0 ö ö

s` ö

2F

_!

S

2.

OöE

fs`fs` P

ö6

¬Q " 9ä ö`F

s==¬ H m ma ö 8

äö e in o d- m 6: ý$ d- d: ý Fc .

" $ .. !

"` " 'E£E= 21 Fö 4 5''1

"

'og LL E s ; Eý ° 6 F ýfý..

PZýýýeE¬ä E¬; {

°ý%"Sn ý@£ 'ý! -' di

Ee! "- ° 8>c . fi fýýyý 5-°'E

°F!

° UE: ö° n ¬i! E "

5f E`ELEEO at

'F'ü ¬ca'g ECc - vF; "Etf "alEz önEC , pC 3c "2F:

pqo Pn¢v8

in y

:E £E ° 1" ý: "s e°ö iý

.° E"t fZäp E : ý`f i ' p

%. " 9E¬ EdcF ýöýF i°ý0"°«

e ýg8ý' o : F _

ý ýEýýý S8E E'eý Eg

PýlEE: `'=4

°r" "

! y : E`" 3. @e

g ý E lý :ý

P « .ä bs°. ` F 4a8:

, ¬65 E ý E : EE`, ' r" `

ý s r E. £ °. 4 °ýäý:

ö oý: ! 'pc

"tý£eFee ° "

: b° ý v ýo .

ý: c E

1 ý

c o $°m ä. F4Eä" c 1 ¬uý E5 ö: Ö@n ie"° : °'PvPE ¬. ý`! ö°ý?. ýE

ý 5 SIFý ýS ý: "- P ýSCý+ ": ýY Y. ý

°! 'P 4"¬; ý Effl! l =OY¬¬¬ sf ö: o" CiýBSa "tLSi "ECSB. °F: : 8S

g3S 8 =°

sý ä¬ö : 2 öE "y'° 5 n °PZ°£Ftic°. a" $. c 4" CFýP"vE F

=" ý sgFE °pn,. °°Ck n$ °- ý: aý= Eäm ° g

Eeýa$c" °'" 1"" $ xß ¬c ý4

diX"`"'i"a« E. " " rcýE5

. ým 2; äo

'x _ .n z_ t £

ý C "-o"ýi: "E' F£ä "

ý =o"S °Cý .° . B¬rS=iB: #`'"6'.

" ' . x

t p ý rP OGf""ý E, ° :9

E ýuB Eä-`, S. E Si pI$¬_¬a8 K.

. ! : "- EFo FISE y ý °E�°ý HI ý: ¬ýn

, - LE ýo

. -U Q i'a4a ý0 e@ý

ýý° ýý E FB

'i ` F :5

g -ý " i E. `g

cö _ . E_ "F-: tv : 4L i

a Z2E . zP ° .¬ °s b` FEE `o' ¬

ý°ý;.. Fu aý °.. ö '" ° ' _

¬: e ¬2

ä88Fiäg ¬53E l 5ý

Eöýý3Fy°ýE ¬ EjFE $

ý W>

"rE °ý ö

eE@xc`=' P . ".. ýE tý "2: ¬8" E2.

EEc Fzä E_E 8"¬e `: O i8E°E

ýnr '0 88

ý 'ý8

p. "1 ¬ý E 1Fg

Eg " 8�«! ý! I% 2 F7: F 2 Z EE Ed'¬ p D¬äF FF E¬!

iEü¬ ýEEE iFd iöaÖöiÖ iEEfE ß üSE . 1: (" Aý a

r"_3oE! iýF

ö Em$

e= 0 q

°

Fs ýF°o° 2 21. '

iö f, ä«

"

Fö

än "

I. iö

2ý$' Fö

. $. ""

; e3E

e v< i _° i

U: E^

" 4 "" L S9 n p F ä'B

" ä4eß

4 ! I-

4z f'pýi

4 ýpH

411 ý

IM U

Im 0 3öm ii

M. HI. e° : 3ö0 : iä0 n n ä m $ m

y: ý b

in N h ý ul h

Ax

°s_9g` ä ýi`: n

ä "

a a öE$-

IpaP m p.

m p

m p

ýý: ý" ý oý p

a pg

° T gp ýi'Fa[ 0° On c S8!

° ý

8 .4 _$ `ý

"" = ¬ 8¬

c c

'iö3fý E£a

eo 5

r E: °v UR

ff_g8 74 " BEci E c'ý"§ EC! c ä`« E'E

"2c "ES ý c is« =" i p" "ý_ý.

ýý: E "° ý P"

EYS "c$. öö c8$ý4zC pý ¬E : EC

a ý

y`

ýF ¬gE

f ýF

EF£ ÜEERä ýElýöü` Sh_E &E6 °EE ÖS EF: mRE

E °

E : E S 8 f

Y b ?

E r

x _ B

i ä x

ä S

F i t ý° c E ä o

öI ! -i

. j 5

"W y f O U iF E O Si 0 I

a d

ä= 1

N N N

co N N

U

N U U

Co

3r w

a v

0 U

b x 14

do

U, cl a) Ei ; -. ° wn

C 0 N

_N

U . ýi

V

b :i

04

a

ö ö

oN d d ýý

"_ ör«E. 2 u4 mý dý ,, mý

md2W mssi möcm " ý. °zE =OÜ m`oü m m _ýEgEä^E

g`ü

v ý c £ _ ý %ýüÄi`

4 cam°3 °° E8ýl5 2tg8

-"ý bc ° 8

: c: ö" L ýq o

ý'`'f °E %: `ä. °ß ý

y cýý°7 c'

mcýü cv£s°cv<ýcä° ý. Ec@ ü

pme ° m5 u': cm Q ýä ös2 ü

.U cEý c E .. "�m

p: Ec°ä « ,.,

c_w

rýä°'m %E3F g¬Oý

o V $ oüE ö°3 a" osä

': tT'ýý9ýýE: y'c . v . C`mýicýýý 4 `H 4ü

vc m. n äE m-Sy . ý'moö

ýi

d ýý`o°. E öa e, ý49

vo ö Ä

ý° Sgctýoý. 6 ö ýü °ý'° ýýüe ü ö

-: ýö ö , s1: ö `°Eý : Soýýý= . "°'8ý; 'mým$m'SQ

; E l

�ääö öm"E¬8$ ý°En°ý ° E ý

¬

ý°=agS e Y y": ü ý£cE ýLL QT m- 'c yö«_$r ö!

:° öö

' z" Fä °ýv u :; i

, :t o°_ m

nR Ei=. cV q f : ýe' $

ö ° soU. n ". 4 cý°a 2° 2 S Qi öe

ß gý

2: °°: °_ w °

ýc'b9e: ý: ý¬ýaýt° e « m

~pýEýGqýv3ammm '"m" 'm

ý°my$ýý°viS a; `¬ "

¬ ý ßbfi14 ` c' ß

_v °{>" ýý s

.. u o ý- -Eý%+ ýäYýzýE°

g°ýcýöf ` °

ý "o .n' 3-

$Scý 'v ý c° $ ßäö@ý; ýýýv g R y uq F o4ö : ýLL m

c

er . ýýßý

g : °ýo :ü

""CZ aýr_S

ýb ýu :

: 22° -

müL äýö c äb ü

-tiW mo avi - m°ýä

° ;e üý n ü

S: _r2

ýC Lm(ý°Em `

g9'Vý°g rv

EStöýýýöy"ý¬ö ° üEw3fm, o°22oa a. ý: äö

N äEv ü"YYÖ"pmemäöa

qßg°n"g öv c'YV Öýý 3

°

2 'E v, öüSý

E «: böýýo °GüPý . =ä° " o5__oL

` £ ¬ ec ýmc° tm '5

gxqäv Yßo. £BÖýmý"6t

'i' nEö nü- 2, m¬ " E

ü' Dß Yä"'gý " zý .P SýzSýs = `F E tL am. -x Aý ý=-ým t U

} [EgŠ `

ý umto $ýiFýtsýE. e: -" P

mýfi= o §ýEÖS

-ýu=83töäeýLeö o8m. ndEF"öý8ý ýrýt . <tm -

Eýý ?"

m°j ýaýý

üö £ýaNC

, Ed Y««

cý ýý 6 ýýdW sý "q dm 'm m 4

9'miW n3ämVl ü; ämV% "3ämf! ö Ö

n r

m

°

3 yN

`s F2 m` g °m t r5 - Em E ct

6 c Ec

ýs g

g28C ý2ý om2

c Eö iEE

E

y 6 E

o L

ý ý ö

" m m

174- N N

N N

15

a) U

Yr

b

x

. ýy

LSS

i-I

n O O

Q. i . ýi

b

L1ý

4 ö ä 0

ä 0 I?

ä

G rn G G

ý

sdýöý ggýEF

42 Eg g 8 !

- Z z Ei$

v N

ö ö N

i3 Y

ý"Epö NN

N <

N ä

... ßä- N

-

ßäd N

.E :g

W y 11 1 -n

Z. =4 yw

$

t u 3 2

n$ Ný n$ m

0 0 o r ý

6 ä

N ý N

(

(. ý

v' Ký

C: ) w

6

K

(7

ý

n yG -

,n K

(J

ý' 1O

To cd

_Q ý"nc ö 'v9af ö

m ;

m

qm

N$

c

R ä.

m

. ýýE@vý`oy

äEr Eö ü5

. q ELI

I- Z9ý EE

g 2, 5

UvE in mä5E "_ ' ci cýir Kö

ý E 2 Y C

_

E N

U

rn

d

t

ö S

ö 5

c ü

5

Z C K K g

tlý C O LL

ö LL

ö LL

Ü Ü

ýD N N

z 0

:. i

w

ä` a

o Ü

"d L

tin

bO

N

E

w ye

co JF

Op äý NL äi U

V12 ý: % sE

ýýpö El.

. yý mý gS

Cc' ; ä ä^=

ýCý+ O. E

fN

Ö

° EEP

0 O 0 O 6 o

NU

¬a P ZF

[ _Ys g

¬ 3ä

Ssgs

Nmö_6_ a

ý¬ a

ö 'o ° m " .$ n` 8 b. °? L0

pFCSn G ' 1r G6 .

6"tY

Z i

va£ -Z °ý n 8cg86r $. Sac E

`. i J 8

,8: i ,

ýj t c£Es . 'r5xq ° ; °ia

¬c¬ °Yi °

6: q ý i- <Fin i¬ tS° Y°te

lS t wtl. °InEm Eaw £m t£ Y°. SýE:

rný`i a¬ ]ea [e ° Fý° " 'Ec °O- e " ' ö "Hal ` -

eg pO v F ¬" e

e . . piii i te f °n Q 6 ý8='F° " yý^ iii : c°"yýof°° $ Eg 'Ei°

.a° ýý{

ý ° S rv "ra cgg ý: °r5

E`°iFa t8e ag cýc S'"- '

-i m- ii E! E¢Sn tný e

_3 S°öf;

ýn° ¬i

E

t°ic c° "' F"r°_ 'F w'O F"53 ý"

S ¢< ä8' ýx¬ ýý ^ S " '

" ý£ý" - ýFO Ee a .ý : '

¬i: '£Fze¬ 2 Y"a°iii p- öýý iö¬

ý n° ¢" ]e', ý°cS fwý F S £. P]siF £

a' ýý F. ýlw° °i-i°ýix S`O { gý'ýöi't ýýt! °gý<v "

E sß . cý i: "¬ß i5

ý 8 ý "ý.

Ew

as ;<]

" Y

; p , E °¢m

ýT"a{nm Si¬wi¢ 0=058£Sa ,° p

; tn]Y a , ie t

ý" t ° ] ° ¬°- ý

° e ° , >; Fgi. i; ? 48 c , z5i ý s. ¬ e A ö 'zs

w n p a . � p< öo

°S mi¢ pwý fa .:. ]¬oE:; ean

r r

E°a; r

T ¬qiÖ [ Pý t

¬ e: e l tcE -" F°E

ON =YrliP.

FRF

( p rýEyi8w A: e°ßýý f¬"i P ýea$+E" ýt

5Es wi4M t" Y¢¢ £

° _ p

i Fo

"8! t%; i :=

ü . N..

=E ij q.,

M L 2 8S. ! [i' s` gszä2°Ecä: n"4crc° °¬s 80 t s, ofil

° sn8° ý. " F¬. P8 "' g g i :" i °4 l

p°ä¬°e ° a

: p. " °z 2

: ==_gosýýa°Lc"s' : ist c° ' s

if .° u t i e; ý: e ý . r .yj LS ` ýs"g : i5""°¢ i' Slt:. ¬o¬ <:

wz''" iý [gis'°}ý<° ýý, :8 E °d ^£ i X :` bý ° ý

" `¬y°F - ac r" .: iwe &. """ yc'°. 'n li : L@=r ä f5 n

]ä, :, ,. gý tc ! ; nn r° :. BF. a]. 6ýr' n'. 'yya" ý ecliic sw, .¢ ¬^ ¢ i. i gyp ¢2 E " ý' ` as

"ö g5 " ¬ _

g . Et c¬t 5L vii

°P"miri gr

5" `

] _as , ". 5 ek i ja i0t°j. 1<£c" t1- uW {

xi t i i i E"" i3

oiE iýi S c

! L4 " ti it t'. ýYTE2 YY ! ,° z rIr_g siý. 0¬ i" " tttY

äý 3i g Pr, ai ii i[ `G E t c¬ iÖ 8"ý ä 8 ä°¬ ý¬a ý ; .,

'ä £

sýE s . g ¢ ng s = ]e wr i ]". ¬£

2S"ýFig:. ötc: ."

,: - " " tb ý w S iH ;n " " b zSg2 T. r i ý

°a¬S '

L z5ösc$EýSi tOE .

¬: F F °°ý, ö S

1°S °ý n .& E! w wc l° cP d e Ö$ý ýän-, ým', °. e 6S`

8ä=ý ým`td ýE 'fE ¬'i yý " : ýýä: ¬ ," , E ¬: " ' ( t >: f. [ 1Ias p. 8 m ä? ýC i ¬g ° °

I °ü iS: zing:! E e':

¬¬n iiýti ¬m57:. ̂ in: 5 öi äP ¬.. 5 ", ýwSFE: CiSnüe ä är .wM M i!

4 Eým

ý ö 3t F°

ä £°

Siäg F°

UR F°

° ° p S S F

Si

ýF - o

M. o

I. LF"" c 6

1+F'ö

ýüps

a E <w ý

r. a' ° aý :ö

apýi ,_ : 3ým I py:

in 8403 : 3°m

Ei_°5 : 3eß

2 X m

m X o

N ý

X

mt°3ýE n n _E

S'Ci Ei

"rýEä£ N y y

_ "`ý "ý"8ý

' ýiý: ¬ ün uý ¬ýBPä Fe8 Y<

.F c rg 3 Pi iYýg a". a. ý$

2 0!

qaoö _ig' S eF$0 `

-L `ýý " S8 °m'py h

g L OEi E¬ai

"

x

ass, " S °- e, t 'i "n: £i

E E cS nP

s R i3y4 !

-

i£cý : EOt f

N E aS"ý E{ öl

Ss r . r

$ F

E[ i , E£ cq

, 0 n öä s

:

E'! gt

I

i ¬ -% aýifö'a

1E tE" 2

° ii= 'hilliü

unE. ,

ck :ÖSo3F

E 5°_E oEc88ý

$ 1 Öösl: S: n. LLmx

4 0 4 E ¬ S4 S as E

E K S 3 =4-

ä p of <E sc f8 füg

Ö ¬ f

3i- = L

< m

n N N

co N N

z 0

(V U

w w a)

U

0

U

b

ý4 y

bO

.. ti i-+

Co

a) E 1

00 C) CD

aý .C U

1U aý a a

ýN

rN

N

O

N

d °ý ýýV °= Yr '

a. Q"L`E Ö °

¬° ý

E"-Vy e'"N

ý °Z

q ý4 E

rqýLL ü ü=c K=är, 5aNZp °4

E° R ü e ý äSý w. - m ö.

o tý °c °2t Z6 E: 7_ E «

SB 8 w$

u ýL c E ö

Enaua wgä ýN HC(W7q

m 2

' ° og ° � Xö Y dröy Fü Ee

cE9 C i c °° ° p L i <aÜ c_° ö

ýu u n g cý U4öý rö w

33 . 2 ° z° xöeü ö°¢ ° Y' 1uY EÄ E¬ c c

pe ßög jmö X °_ jv v C@ 't m3E

c .E 'ý3 ums ü. c° ec nWUý

. Eö ý öÜ e=`L

8 L' EZ °c ý oNZ 2

ä ýg Eý

g $ý eý'9ý

6 e

6Wli Ew r'mä58ýý ýdYý 3cü

u _ -yö gqAE�p °np üW uä u S ü

öT ec$Qm9 Ya ES., nQ ¬E Ü mý", Cc cü -EEö°ýcöq'$ ä^'°c°V S°ä üm°p

c gyrr ° ~ '3 öF

Cyýöäo

g v3ö "x, Sn

ýý' °" -^e ve -w

q2 ÄC

ýL ý-. ;`ü. i9ýv. öEv-K t. "öq Bw

ä -u r. °+6E mä g gý `ö_ýu @ö ? Y ~ U z. pm °°- ý'

I

0 `te ý 2e =Ec ýc° ý ýq ü3

c ýd d. vm Mý we N_ ýc öu

E` ' cöron5 ü°

S° Y`° öü ag g i 0 §vaiý m 'e $ý!

wyr V'ý 'ö t `° ° °a E c Väü Y öüü

ö ýg o °a2

ü jflJjt'IUk

Y° :

I;! tHhII1I! IIW! 1

öý

lii

tc 1t . iöö

12

N Z

Y t( Zf

m

s

M

{n ý°cY

ý E

a c_ öa DOE eý°vm "oESn

E. n ßäE

A£

8

°AJ

Eýa 0 z 0 CE

C Sý'y y.

ý üc 2° ý

O8 Y D6mN

c ý 'C

E

z qY yý" ° u Ü

Öý W=ü3

rn N N

Q

C) M N

z 0

v

cd

2 w

x cn

4° 00 0

. Li

21 10

ä ö

Cý

3wEöý 12H,,

e ý `oA

G

ýn

ö

,? qö

ä `

F. °.

SS Göýt'`

.EE. p ý`wý co'ý

vac

Nä9 ý. -q dg t2 o "may

aöö '"c 6öC<üuW E

2 v °q'ý c ö - ^üýc ä

"G ä°

Q

i" ° cäöt

Y ccnäö tR . °i

.ßn 6 � EvdL>°T dv

öJAE öc

Ä EEöüýü Yý ýö 2, yc

ö a Nc

'29 :V s-aZ"sE

E N ý £ i <äro w ý`e `°äEEE m7 ý`e i- in v

ö . Eö . Em

a " a" ý, d S" öm wýäm ä3äm

ä a$

S2 a

Ný LL

q

N LL

ü ' CCcý ° ý L ü i ö o ö °9' TS

5Cü gýý N

¬ 'E SEL ý g <n Eý d3ü

T !ý

4 ä Q

a= d

M N

z 0

a) U

CSS

112 GJ

a) N O U N_

M N

bO

N

Q)

w

OD O O

N_

v

U

21 b aý O..

ö

nwý

_S CüýE

ýi ný

1,5 ýs F- °11rný 1O

49 o öE°ö. Y s

vLÜ g

z9`4

12 EL9 ýnY4'¢ö S°ä

`n i° mRE qc ýö `w i°"ý"

cýY'-E+'-"ýaßi

oü°c °mý. `v'

b° rä n« 5° eiö 'ac° `m t_ i°ý = d« SmA9ä

g =°6 ýo dG

v aýi ß ö- YäY `° °° LL äm ý' t

Y t9 'ý g ö Y , a . i°

6- C- Yw

ýY°ro1! ö E

g " °äoü iý dYätÄpö

S. Z eöcý iýj»m

$t

sg "ýýCÖ

c c s` n° .ý

9Nmö° s ü

ma 61L L°mo "ý3'Cý ,E 9 ek 5ö "

qýv `aw pö öS Y`e '"'m Am üe w" ö s' n

ý üý 'ýa °c°NZÖ i"

Ec'A ° "

q" W$gqqý9 a i° Yv E cL ? am

N

ý _ -

Y äp "619 ,

°,

- i°

c cä ý

öa _

Y � '

.: ý" w 6m üc

cm

- °°°ý°müýýgmsýsr _

ENg, ° ýo°°vv-ý=° -

S

1 3

Eý

-

Eö

ca V,

"ý" o aY' a

1y 97W h9 mw

O

E.

c aý cN

g

ö

ýEü

yý- i6ä

a#.

K O

pl E

7

N M N

M M N

z 0

v U cC

2 c. 0

a)

N

0 U N_

N

bO

N

C) E t°

00 0 0 N

rj

U C)

V

td

'L7

U Q4

Qi

Eýrýýt `sue ý . '",

W 6; sLEütäe

ýnÖý26 sö

°4 a¬i c öý ý' ýF

dg y-"21. ma

uý 3v äJ

0

.ýö Ä

E°

No

Ö. ý r

d'i

"C2t

°ý .pSE pe oqm ýu

ü -'- ý

F� ö g' nöZc

rvý öý vSü

" =' 'k cüßv '° «c ä '2

o S 72

12 ä" _ss äb¬=°ý

EvK ,e aýi

5nÄ 4ö =d ö

CNonnpC9An

L' °ycCy 'ý Cý .6

m¬ c4Ä

ý=äg dr a "QE

va 'ne g2 iri

0 h

öý

ýö cý

cE Ho d "c

ES¬ cog

i° 5° 5°öy! výE ýC

cý

EEL V. = Iä

, `°, 5$ RE äEESgö

aý EW6cE

dQ

a-s °4r

rO Eäýaä

5 ý'ýna

C-ý°5n

d- M N

z 0 0 a) L) r. clý

w s., a) t3. a)

-} 0

V N

N

in in

CC a) E

w 00 O O N 4,

cu v 04

D.

ö ö

`PgEEr t n "F

6 öm: 'g " ]ý ý i

°f °" ! g:

P. ýtm: etz; "' "Fx" g ý. ä! fei ,$ ý"

ýF' '"'f E!: f

ri

F eii gý i" gE

oiSo a of $

' " ö

F S i S3F `¢: S ä! xE

: ' {t

Ff F a# F_

_ý ii ! : ä ý T

t F , ,p , itg° E¢. S. ew.. ° ; ý" "_(

°[ iF

gf ýE

e['g°=`">i Ei ä °jýFf ý°'E

¬a Sö Y f$ gym'

#[" ' $" _ ýtF$

it ý ': '' ý t^ SýE.

S° S Pi ä °t $'P'S

ä"i tE I E" ý"g

tja ývi zYFp " $wZ ¬p

iE" 51F .

E ä lFj Sf $° Pc

:f ?fE iömf P "

{c SE$

° g'-: e¬ý. t 'F "jt"':!

": Y j°nv "

ö" £t1,1°"

x ce"> fg8'° t"Y g. {F-gf.. i

" '

$ "r : $°¬cý3.. F! 6g¬ t$: if $"

'

< ° 3 g° ýaný.. "

i! o. iF fE , ýP$y $. ty 41° ! t6ýi "Fai .

" v{ ä]giv

Qg i

, a : EEL1ýý ý: 3

t$ °"''°S gýF Ih '> ; ýe { "i #

E : l5: i iä ' fi§oä ei$ :.

e "p[E'° ,

d4sä ¢"i- !id{ rv["ý

° O. e

$' iö; >ýiE

"7. °. o, "ý: c.

g ý Y '

: Fg. ý"6sFi 4"4

i°`ýiýF6ý[ ¢E"-E, t ýS4x

of g"6ý"'j .ý gfP Fi

e $c° "g$S ¢g ° ' " ýCi: i$F

i .< gS ! ýi `oý°

f$Yö Ec L"

; yF ! "E> t

ý " , Fr$

$PF ,°'

°"FSi : i"!

jý; : a: ý: g °F i"c" `ý ;

og g `

i ¢""g<'°E Fn öaäf"ýýc! £06

. ° 3s g ýýiiLbSýg'" u ý

Fý: Se i'

f $ý t :E c "¬t

F° - G

i i g ° ýE :i

i y 'i: {°gigii; StE

$ tö ` S: i

sS vFe$

tl. EFi F f ° ,

r3 m, r$Eg: e"i Eäiý° i °

° {! " R£. gmPä° `#3x 8: t

° eE'ýi ;: '" n: : gP ýi° °'

E f"x £"ý f$ý¬ Cc! ""° ý

' 4E 1 c

( } { v[di: 3(t - "

cý § yQ3

"E fiES° 3° F'ca: v" "

F ¬, ýý ý

t° e ge F E "3 ýý ý5: E "

gö lg °, i tI:

[n u4 t 3ý. "3 ý ' i Fg

.c , ü$

" ý iýP" r [_

6F : gip e'ý F. ,ä"5 ý "a i i

L " "3g °. ii °e "> vv 5 ä °

gi $ "$ F", " d'iv. $ '

fig#e [ °° f, g0. '""°°ä"t°

""d: "{c t ý:

agf ' . n al :f : ýc F >"Ea: ö4 pSFiýc"" 3e°

£ °,

E ,. °i uE$ý'ýSi{0°°CgiEFf

$' ä$ i <Ygfc i' °

t . $ .' $¬

. Yi: E b e°S i% E' ¢ý 3s

. ý' ýý

°ý i

w£i. "E ! e(. >"a" f

i`. °ýE$ '! '[Ef°z gE y : Eev ýEyýL

: ýEP'f

a$-`Fü"x6 HE i g#f

o. ý"

:i iF. °FO ̀ °' °ö¬i lg. ae ýF °ö YE¬

Ctt : mýý

. '{c " $3

. ¬'r'" rlPac -F: a3° ýgi

f', o {r ¬S. a

!E °' týc

'ý: f ,

'$ ° BbQ °! iý

cýiF v! tS '

ýiP` d$

y : ýg HIT, f -c

e

eF:: F ý".. t ääi. gg

ý"05: t-°ýi F° ° E: c g£Y Rum- EE"- a" >

S . ! ý°; ýa.. c ga. c15 it p >Fii° ý'FýF

ic" '°`-- I f ý"n

E ° FE I eP{ fig $ g""E: ! F

ä $" ý_ $' ý

äbEZ $ ýE Pý

$ 4F.;

cý3 g 4 E ý ý 5ifm: zE äcý: vý ý

g> `

Sä °{ °{ : I.

E fS$$-g. _ýy

Y_b° ; .

( f" ý°$F

c 2 v°v 3t"ß

3: ý g3 E° ý ,

ýF°. i i7 tr ^ý. 3i° t3 `:!

°¢"ý F$ Y [

ý S F ý a°°f' :j ` ýf"f! S

g4ý S" t¬ a, 'ýgäd'yv

c 6 ßi f35 e ;£! .° ca{ ,c

`! F f" ° _F s ijm_v

` YF ! ý3i g ýý ! S-u o"p :s Y>8mt°Sý ; YbaiPS

"i fgöP ieIf

t ° LcF[gOEiä dýPt

s$'=°ýf° t ý $EýF

i: Y

lF! $° ii 4�ae

ý[$

° $gY4E , OtE

ý °

( 4 #

c{>' Yý"$c ö Y°E A, [ [ Iv, c ý mg>"ýcý

zýS g $3'; 5: {c >

ýsS;: iö: S F: 'ý:: z ° Y :

n m° ! i$m $.;: F! L

. `m Y °mf9Etc. i°"p $:

F iP 8 ' ! E$ Ixf ýY g$ ; : ýff

"ý 3° ° . '°! tF;

'ý°G: Fga :. a c°F { ; i

S>, {. $ p°E°°

ýL S°a.. °Ee "tc

;Y '-34! w

ýüaigt S. ° °a

Rr3 ý; Eä tS

e_e aý" =E S , E EY¬Pg auE °ýS 5° ! _t$°

<mY

_°i x' F-ý3$geig :ä

eý ¢ý -

°. $ 4: is SqS

"b r: 's Yät °aý: $ c m8 z "'"'

: ý 19C' !' ý

g. i

a .ia w fii E# gtl ° _

¢. $ g°"

E$: Sx

eYý; a l

gSýý: Fi flit,

E %dv nx

" , 'tt i gä F: " sEt "E$aa 4 fi{-r ý

°. f, Y'gP ga{3°r. °$ gaa Fi F "u iatY i

" $ ý a: c e $ L . acE°.

° ýa Sr -S L" i iiII ýý

e . öy: F {, c°z e. tý4jii $

-

6v lc !

"! $i "` ¬n <C Ea F g °

F f EY t ý$ogEg°a, """:! f i"i£FtS

t( 'ýjv tu $° .{

ä2°i ýP' ${ E ff i; S

iS 'i 3 <°f !b 63 eb S S 7e '$S " aE" 12

° Yr YS "f_;; ? -3°F "i_ S

" F -" E' i>': S: a `Y $fE

;Pý . "a"$a Qa gý )

c°' äi s: ýt 3" ýba

i , I iC t dtEc. g°"

t $ L" f, gý ,

i> S{°it

i. "ýc No 9 c'ý-

L" "

}t S 1'" itf: aýo>. ">. >a° i{S2: ýu: a.. °a..

S F i3¬¬ e$u gý: äs tc{! e c :: E 6 t: £i i ý. $_ö. if fi

.S f: o $ lt. t. l. týaF

b $i F'ev$$i. °

gFg ýI°

cce n' i ism ým

" Ym B s

ýaf; r, s ýE

£ ý6ýö itÄR

9r

$zF'Eg II

[Sý

! ^^m

ýý Fg ý" ic$! SF

äiä $; l EýäE=f¬M mýee S; n¢

3 3

" p

i u

3 I "

3 i'

M N

M N

z 0

a) U cu

2 a) Cý. a)

N 0

b

x

to

N

Co v

1° 00 0 0 N aý N

U c)

Ü

b v

° 4 d ö F' 4

d

N

y MM

ý'1 N br

NM

CL

: 3ýý, aS

eE ä, °'ßnÄ äötö °_5 $a ý° ß WOF 5

iä8 VEB p ES

0 ý'- VE H

E °u8Ä rE Yc ü v i äo�

EöE gö LWLä «E

2ä C -ö . «r84°o8 8 T° ° Pi Ä«c7 ' ° s :-N

ö Mý ö C7 �9ö

EC W mä

cY y<9änrT

ýv CS T , ° �äG Y u ýä mýýYoL S Zý

7- Äz WLtl 3G� ö' ü

�c

W i

E° `ü öc Fv q o c M

v n . Q c_ mp5U tt ogYüu E " q

° °ýä

C °z.

5 ß`c2

§üm

:! i ý

j '2 aýiiö 6h Em ýt ü `n t c

mýE ^ ° ` '

ý'

Hm SCý. s E'ýý. ý t; E

qý ä

$oa ýýý öö ry,

ý � 6 eýögnc

v �

Ch ° R

v ° E'� rü q ., e "ü n6 V oE ;e g- ö E. . v

`cä p°ý W= °cý gn rig eW

äC

E `m . O q Ä5°o, E t7 EE° öE

c °V

'E° EYYürý. c K n�

o EYa'a''C öHy Gv`ö. fi

Yj EC

°c o üÖ5ÄröEdE

�L'

_ U

gcä CqN r

'O öEE Q° 22

° ' ý - 2! b 75

=ßßvß ý

ö5ý ýý'v ý' Ed £u = °ö

` üüý° °-°Ä a'ý`

,c pý

y �° SAO $ 7°, aý ^. $Eö>ä "$ ?ä

°näE °W iý c cý"ýýý 5g E

° rý ýýE

° .Ee

üU °

CS - ýc°rE ö °'w°oü°üý

c V_ L n Ti ö Eý x155

rN .N 2Y N

Ld (y 2"

(y

yý

d ly

2Y N

4 ýy

{ Y

�Y �Y ,

YY b� d DCY h W

= Öm

ýn üm ý üm Öm 6N Üm

vii Üm

=n

x Ym

ZR e 0 or t r

m °m

N

c

N

a

N

K

r

K°

6

N

°I m

D

p

«ýv

.Q v1

'C m d

ýY ýR T

B OvÜö . ý 3 'ä ° 1' 6ý

s C E

c ryýý SY

öý E"°Ä E ý

cö= oM S2-gY ° 1.9 c oJ "8

c b Y

T

c cYC O1c ý

ý _° c

EE o2 II E .pc2 o

ach Y e. 9 Q E EEE EE

Uö Nmmä5 ýt

.tc Ü

. °t c

Co

o. E g $ ý ü Y 5

öY ° E

Y ü o x

E " 8z rc ýz

LL lL lL

Vý V

n M N

00 M

N

c'3

(L) v ce

2 ICI,

ap 1-a ö

u U 'Li b

I- .i b

cd

O -ý sg OY e^. Nm °ý

u(.. .

U° N oj

21 1 el

s~ a-U m 0

7 ýN C

tl

G

öf o ö ti

E d ° b

o

0

zä

! E¬ Ö ö8

°öH'E ýEa! a

T WL .. SF

Of

c¬c

ý E ýý$ °

tl P C

nf

.

C

ý°ö_ý:: a a a S a

ed Eý ýz3 r= Fe E6dn3o. #%ig

2= 8 d3

; e° ' ýi f SiQE¬. ' d: ii gE

it tjy E EýäiEä{a8 lit.

"; ýý`% ýttE°Yj`` l¬"ý.

[

i! {: Sýy % {dEi=

,ý {4 E3Eg Eü ýý

{y'gitZi nE%"g=ib i g ö "

ý BEpe Ei C" ; °'aH2:

"ýg ý' ý ic°S

: ýt%i HYi. E; 9 'H. ¬Fe

icttdEi. EF S? "°, i p{ ý:

ýEi''H . tii

grid{E3 E {{

{

'EEýi'üaS"ý Hý ýca 9oE =F it

d`E °, jFööiFýS EE 11112` °'1, i ! He ä: 6 . 'H DE f E i

E

f g i > ace , ýY' Eä

" .i5 " i

fii-ic d8 6 ; i: ntEai i{E": ie. "". E-cý ! f'" iý

i äji, i¬, i"Egvi

: cfi z

{ bgd: q f yin. ¬di t-

EýZýäý. E(qE{%gH

f¬ Y ¢! ̀ E'"fj%

i: i"S :E Si:

z i i. E 4 ! 7e S; a ý . i[¬. E

äeci"Evd ¬¬ c EEb "¬; ýnS E "

E y

.. ai $d: F. Efi tE3'

. F"ý°di !E iHS E{ i= ya a: E

_y :c 10 ds°7ýiatý dsi s % 'E;. F7E">°E' "

i " E llE . t¬" '. @^ýE: '_ýa

i

cr f; ýg ß! 7`g S e$.

;' Ld: 43$c. ib"

ib{2' Eel: E¬{ý,. H tiEP>eD°dE Eäi g

a" S E F:

i Eg"" {55Bi'=

=igEi "Et pr'ES EEcEi i°ý'$idt1{¬

i ib ad g6': E

.eE gE P E' _=

ä zEg;; °: Eý . E:

g {ä. Een. E { > _tE ES b `PE °? 9Hj3-ýi i

E °tf%3c.

ii 8

:c uSiýi'fg, " F

, t 3 E. . E E"> {F: ý % '

ES. >° Eß äj %"'S' { : ¬ägý. t2, eFiE3 ý

ý" i Z i E>°{; E fEn d. cii Ee, nt{

a " oýýý¬ E i °ýý a

Y [

¬= E"t ý9 t ä, ä teSF E'f EE`"

"4! °ci EQF i "=" i' ý

%s{ s SEL F . E"ý

sSF ., ý3p @i, cý .. : b: T ýý Eg : Bi

_ i. { _ ie _E f2%

4i"D c t ä "

;; .: Eäýýt7 E fEi i S

, E (ýi RF Ste. FeZ q : er, a-t. iEPE id i'"Ef t°ip"ýf. Hf

{D t_: rii. ° iii'¬ F8 $dct

pt E$ioP$! EeYi

'7 S : E 6

¬! "iý led¬. öiiäH ýi al '"i;:

lE4 yt ý ESi

SjHý: fb °S

ÖEptpz ý iii r

ai i H g4$dä

° ! iiid3H 6E.;

7ý d ý: 6ssEtf'Eä °

{ iP {_it E S F s :E

' =Er-

`ii8 3 E g" i +i ý

F ä¬e Y ýiFn3b2:! -..

i %F :ii !:

SEi

i ¬j i: r! ii 6ýE

, f}.,; d

ElH ; t

g Eri }EH. F

ýi

Ff t LtE ö! i Yn t p6N i

ö " i 6 '

i

ööi gý

E " F ` g " Ed{iý: "

i EH E

aý ýýiýg ýpe:

ýgt E

SgE E tt ý Edi$ýý¬ e

i

aý :s e .: % Q E" oe H EEt ý i i. i= i

{i¬ý¬t`ýEi °zi 3F ý

iESe8d8 ¬ sE % ii"4.

i%;: fot d ii

pEe d'g ýHgyF(¬

: '

E i , e 7ý b

. a, E{". g fld881 {N , ä i6 E_ Ei. 3 äi: SA¬

". EEi_8

e¬: c> "e° . H.

. te Yf

a lrE"f eSt_ E. EEEjo ui. lun u_ d

F Ee$ ö

i ,$ ca

'ßäR ce

'%ER c

J. ö R

ca

p '

E

S S" 'pý Pý"3

ti °PiB"

PfP, it";

bj3 9( e

i0 'M BN i5"EÖ Y; EO

L Ö Ö Ö ä d

i 3

iiE: 29 8 E F2{y Fg{! f

" ýÖC ýi%E: g

N äR

d ö:

t N e

ýý

S! °F>ä {g" 3'SU

8ý8ge

rd 8

Ej $ Fg i {Y

c °='S Bö8' !Y 8%ý > zEäEEE`e

E. " . E. t

ß

ey

t

. zy F%ta '

mS B ___

ý

VE dSiFe i EF $¬¬l

dý"g$ ' _

S $

ü

° ° a t t

ö _

ö2 ,E

! Ef

ö c ,5 Ya y, Id n ¢ , Ö ii Fc 3F FC_

r -p u

. ! E

2 E

ö s cn

t .

" t

ri

0 co N

0 14-N

tu

4) v

w a)

a)

0 U y

bo

C) E C°

Co 0 0 N

aý U

. sý tC

b aý

¢ý.

o CL d d ö

Wý

r

ýBrmEnmE f°. ý

aýmoý $° _dd

ýN O

Ex d@ä-

nr`öN LE«m

$«

N

äü rä 4Ä �

q° `« : 'O

O4C

c

r

mr cö`° ýý F: a`? o«ý

m°y2

E

Ö

a

q bd

Ö

d

E8 ýä

Ei E äý Oý°i `fiö ö: Sfi" ; &ý

ýö üö°c q e° r

- `rm E" Cý42 = yý ýnäs

- 2 9 ýy cg mQ BgS Em py

,; " E-ýEaW2E

-

c ý� c müo$.., ý E2 $' C �.. g 8'E E I

-Up ° "2°2

U.

-Up-22 . -Up-22 äL:: t"_ ° S Ný

Rö ýgcS:

a B°s o? r , te

Eu

, te e ä 'd 3

a$p' Qý' c n. ' s

Q örö $'F S vc cý ^ýS"� ® ä CE 'n `a

cc u p rG,. ccý`F c a E

"98"9ýEý3ýc°ä`a y$ý¬, g ýdemo «EöEýö

g °$3 ¢6cmý1 ýc7W sE ism b ý`E `ý_c°: °goýcö T°,

ýcmý: =

2Eý5$E56 sgýýö ?o ga; 8_°ýeäT

g

pýY 'c.

4ýýý _`

a ý2. "Ömcý"ý°räýma ý

I °mEYgýg'a-ö c ý

ä n S h3äLýi new ö2, E

8 ü "$ ý '2

öjj. 2 YÖy öR°°

$° 2 kg'`st3 q c rwö

ý ° " `

k

` n m c ý$Y E

$'ýö EäýüQýn°3 gý9

°

. ýýýä

sä äc °dq?

mmýýpýpd ö

°ö°m ür ý ýý°. a-Eý ý 5@I E1ýä '-¬ `

cý E rcýýiöF

s° D° .z p ä9Y

Q s

äY°c ý °

. Em a`ayW' ýt 'ý. N E$ m

mözq »t m56 ý ýd E

°ö ý`° =rä° =ü V w c

ýö F8$ýä'ý2oýý w

ä ävný `r °m�ý ýý 3 c .ö Fi

.Ký: ö E °ý E m ýe e öc nä E a y q=, a

`? ý a g

c ý ä

I? E m

a _ g g ß ö:

F I

` " ;

Egsý ö' » --"g

ýd 3Äe9 ° ä g"de =ý¬ vie s

`coo '= ýsE -

U

m

C. d

. Em

ý Cý N

A

ýmp

Ca ry

Ö ä` ß ýd�`dm °m

bd'. Mdy n3öm

EMd n39m m m m N

ý Ilk

_E<Z`25ý

A

E WOCrvý rý° gEýýý+'ag'i

m sgtsý YC

Bo ü ýý'"s`ýý J ýr _Dýößý

°n g EBýn 88 nný

p m m° 3 9

Q F Jr

cy t3

oeW ýýEqE

o 5

4

=

F$

2

4 n

C n

J° tT Dm

m Ö Qp n4

W7t ý ¢ n

, ýý

d= G d

N

N 19f N

cJ

G) v ccs

a)

0 u

bo

a) E t°

OD 0 0 N ti . r. v v U

b aý

I

6 6

r

an Oil

S °:

ý

i" t a6' Eßä F {

6t"1

> .{E

ry

G Zö

, 3t °

L$°ß cim rgi°ie"ý. °

ý p

Di

1t " 5. t'4n ; EESY c ý

6

c°1 E("

{ý tE" FýýýS° ! { " 3 "

G

5vg ý. '+'' E{ $ E' ¬

4f

i8 9 me

ýý{{

ý''Yý 8¬ 8 yZ S. _g

6s "¬¬ .E E{ , 8ü3"

ý' 3ý ýY FE'4 t ' Fd> ::

n 4Ea.. - g5 e= ¬, m

¬. c

F E ýýg

, . Frx FF. E 8{ '' , ý"E `a ; , >°{.; . s 3g ¬E` ya""t St'

a2 " g ýI E ElS. jf z. ý {§ { i {

E°z, i IES E '°

3 ý fi ;tF ýfi nE " 2cv°¬°ýý

E Eö c

"n t :ý_: i

I !

t . g; fi E ýNZýEit t ä { Y" H N

p E- 8 ¬-aOEgýEEy Fi m' s' pS

ä Y i2 " YtEiyni {` i . "t

c {E .¬ Is

t°e ¬a@ý°ZE ¬ :gf Ei "E

fE ". .

a "ýý¬'Cf

p g

g°. <¬ . N . ° Yto2 { . ° ý" ; E; '° 8 c c_ U ;° ¬Eý O _eaE °e fi ¬ E e s 8`g-gl2.

ýE_

v 8n a ý" C . ` °° . E ä_

"R fi :

3S ¬'

i F E>ý3i'Zi ga¬f SF ! °' !°

,ý t ý; _¬: :.! 4D

p ; g; E.. B}{! e{ '. fs° rein a° E

. Lb r¬ p° ý : 3 °, E {"g2 $- ?

ýýF=e 6 öý 8 i: ý i{_"ýa

Ö i £ F f<e -4mFe§vm 48 cm4£-8 ýES

, °a ý> ¬S ö={rý"°ý: {°'t

° Fý85 Eg i E: t i

" ý. "..: " ý ý 'i f t e t8 S' Is

ig a$F8 >rE E:: ¬"

gý ýsaý

{, 8y' eE. g °c "a ! ý s L

ýagsE ! L¬g_P 3 'is. e°. "ýE 4 --5ä F8 I m£ t n ¬'

im{ ,

SBý. oY"t°

ta ¬°, : tzs ¬{

3E Ye'°-" 8: ` t'ýln°" a£.. cZ

N

¬ E r8 %t

}ý, ^ E g8 ¬

° :.: E !. 8ý E y? 6: ' 3 " P" $

°"= 8 at öElE c ' ý ý il E_'

. Lä5 g ECga° g=£

9-YEta E¬' E {

f 3 tEb ii{ o°mF d ' ¬° ý

{; " . F fi¬ iE( ¬ ¬{ * n ° °

£e t i'

! ! -äýr

! avn pill

° s nle EC 4d{ p>{`ýES ¬°rEr¬

; >E¬

, . 6"- Fg $: HUM Hlii

i ý¬= 3

SY°Zta_ ! 'c. Eý g °0`tägoaý8 %4 -gmj a e

E . .% {i i! F Hui

`t¬ý0ä ; li={" ¬

S -". °i EYF ° ¬C

'8fi: EiEgYtS E. " °8 ä 3

. - tf c¬ö ¬! E°tý"

i£{ {"°"°{ Y ! lY 5

; teji{ "1f

82 {' jýE B

iö !E $iý£ ' ä4>UT{. : Y "EY $ -ý

_ If

9e E i° 0iZEee

¬{^8 ia8t ° ýs

=

g_P ta: Fýäm ° =OiS" m,. g`sn= S fit i

2 ýa ¬

_ tt EE f `ýý S. bE 2E"

x! P" E ý

s f 5 Y Z' p6ýýya! ° ¬t: 5.8F3"¬f

Y ý ° F{¬rE YZi, 3lýEi ° ýý ýý "

t ti°s ig S"192

ý ý_ !¬ SY' E i F° a ý ;y :ý F{ö1

F EýEZ2ýaEEEE" :S Emdý ! ": 3m ý; Z E° ýC eýG «ý

¬, ý ä. -: -gFg sg ým

E-, E 4°£c g-j ` Ee- .! Ea

"%ý: ¬E Yý ¬{£! °3" Ev°c gs t. do. cs ES

a. s{tiE""y ; S"%"

_ . 8° fi¬Y "

, >"% 1°ir'.. 'ý "gg. ' ""onc t: ". S

ýid$mýöBiiýEqýýF°8

!' iv. f f{{ { 'S l E

c ý E£ Sýf

aä{cä c t2¬=Eb c^ {ia "

¬e-ýnEg' g äLO m me i! '- £

. i-ý { a iý$°'O_ 2¬s°i°:

t "

F'ai y{ 3 tý

eE F

i c' L'°°¬{ýY_t 2'E"ýEF°.

E" E {¬ ! °. 2

ý Ea "°: f, ý

Y'"{: {{" äj .i 58

5¬3 8 , aa "i Z

°{ {E Et $ S i E> ¬{ i5

a öt Ef

8 =¬ ¬¬ ¬"Yýei:

`fi

iy { Ef¬ {

sEP FE

fi ¬y ,

4{cEZ"t¬F ¬FS-ý °

ý°e a . 6°°E°m°njS£° ¬E

" : 8

¬T E gi

°ý_ t'"---{ tm g¬_, `.. °. ¬ "°c {"

¬t its qE't tlýggýý 't °°"E t

. t ¬: ý°e{ s"F, °t'

. "Ea y q °tm5 " EE 8

" ý { gE OýEeiFS{ Hin ! ý"

g28Pi¬{ Rp ý S m]B$ ¬iisYyp °=aE tEE{S EtF y ti : i§°

M ! i'. 8ý i p E1: 2öya{y q {

. ýýfCiEF

na[tiae iFe Fe¬: ä ülPaEie S2 i: ü$ýl üF¬ i¬PPiF-Eii v{¬¬ Ii ¬S

'Si s{¬e¬Y YSi°°ýLl

E; ö £° £g g£° 0., : £g

OR 7C£0 fI

ý Yg4 ä° bý

ä°

Sm-Zi {pt'.

3m-°Z Eiö In

{9ä ý

i ä e; ämä e; $mW e; äm eüm i m ö ö ä n

m m

u

äa

1 < ! fl£" 54ÄR

m ie -c ý Z

p

is!

4E 3'c: . Pe 8 . gzg 4Ef i öF-

e: t e2g Feg t I . z1

{"4Fe E£E¬E tý

i LaE ! { Sa

3FE iF°E i{ dFI M ll mS$. {¬ ¢

4 c' d 4

i E ý 8 !

I m

M

N

aJ u

a)

u s.. cn 0 u

. ul

Ici

N .. ti

Sr

N Cc

E il

12

W 0

CD

u a)

U

21

"o u

tq'ý O

ý

O r N

O

O O 4

n CJ M

h

n

Öý

a

W

d

ýf

d d

eg RE

a

RE Sam ö°mýe c+ t eýö

2 t'mpcý5 ° I'

Edý m o

° °pa E °ö ä ý. `, 7 3g to m- m.. öNý

i

ý9E

! ý' E ý'°4

E @E ds: ýa

gz nodoýöcvm_ ýa mo

mm c. can tepä

Eatl"`° gi n;

° "ßüö

ßv2 dE=mý

^°

m ° öýsgmsýö

ým ý5tio2ýQmo ? 'W ö "

c. eýsg cemg "E eFý ýý'ý Z, äwB

m_m ý öý

ö ö EE

ß t5 L ý3 26

o . Eäß° mSp8-�m `ý ' a ý °ä c5ýý

' , oý rcm3 3. Ym$ Eü äý5ým'o °R E3`o 3 ý3 S , "e s

`E öfä 8ö;

0ä'cv5 2° ý

m . arýým + uy m

. 21ýQÖr5-W45öie-" o

22t "V5 i

¬ 5

` zoý

-m°y"`"'ö..

ä $ ES ý.

E 3°o88ýäýY t'"cmmÄ

W

Lö@ obemi $aýi päy =

8.: t a 3"aö rs=¬ýý e mm o "_5gßä

¬ýe E<<

3` "e°äýý

ý

.g C3°öo m öyEa= ög'-Z 'E cm

ý s yd N 'S Y.

°" aE°SYaým z 5

E $ v °

ý $uac -E m8'e C- o=

zq-cö " 4g "c Eö d` 35

.. ° $m Ec `ýo mc p m

eYýE "°? - "+ä3

Cý- WEE gG«

15 =-äür ü-02Es o

CE ý 3 ýýEc 8m 8eti ýC

0.1 "$'e PCS"° s8ý33ý Ä °ýä»äd`mä834-m m 38ý4ý1 88'3EE

318

Eö Q "1

ö 'Sp 9ý dry

Ü i m

5ö

gýiry

ß s

ö 'S ö

@5ýyiN "

'c "

ýýry 3 mm

°6m 9iý E

d y

vs sým

' 9 äßm T+e s ý'

oäým s o+ äßd s ä ä

LL ö-. w am omH3m "3m

öm mnH3m ý3m m

ö ö ö' n ä

n r

ý ° N M

N

m

r

q n

w

q

ii

W

iý m

N

Cý

r

LL P-

N

u= $ �L CC. .

N il$

ý2FPO ACS CAS ým

"

QSEmeö ýE� `-'E 5

sue 8° ýý"s ö2g gý

E° sp ýýE ö 5

s g

= "e

3

d

c- Y

e 5 3 0

öZ K öK o: "'

3 9 Z

LL

I1 IL

N

U f) ýj

N

Lr) ,; 1- N

cC Q

aý v ce

w a. aý

aý

N

v b x

bo a Gn cu

w OD O O N 41

aý 'C V Ü

"0

a d

Yg U u U, ¬

yE ýo ä

a E, ° Fa

X? 'ý

Dig

7Ö N

ö $ ý ä ä ä n E: Q NSt

2f

_f3

me¬

mý3?

3

_cxm m_

is

05

R. i

.' } °E > ä

`ýE 01

s ý: `

it W om5-° So$ 3° öb e`" am U o°9 @ ib

Who

°r ° F j8

_ X `"

a : F d c f'

ö " E3 ¬i

S` 8 18 - 4 te »?

gw fn =ý: 4E m n

ý: ýýd ii s Uýp II"

°iF Ft, m '} ;ý imýS

°cý}2EiE of eý a° : °S m Yi-°E L -a gx $c g"E

3 sö 8'ccrc m }-ec `mF°b b

; ät $YE ý ýä°S" ¬$. S Sb" ý::

m!: $ Fm ByýE 'F i- E

p, S 668 ýý°" c°ý`ýmöcm

yi,: 3ý b4`: ag

ä8; ld

, '°cm @ F ° ý`

° o " Y. _° "iT E 2oe ''}}E; '`{ E

E m aBYýýWCOY B`^°$P`Enn c ct 8 R E3 :

t ^_ $= m$: _ 3°v<. . e- `8}°FýE "bm-8t.. cmL }}g ö¬ rS2E{

a 5 mi8. ftýý i-"4 atp ým F: ý. »s"" Fý: ' E b Eý ' E ý

_ SWi x

:. ýS

EgB e 'at

L `

; >, Bx (r 8 °L p m>

E

San@i 'E= Eýiý`crr

.p $5 : gc °k mtlmg

9Eä°

s oým

. ý3; "L" ýjSE

ýý _ ''_

E Fi} ` ° E,

ý 3$'° P ii ""3$ ay '_ Ya<Y

E i " äs! _: Pi"i

3 E8Fe tlt b p5eEr- ýýq ngööm8: 4 m mä °

8? ý8^g9

ö} S . S: `: Eß Y

8t : ai ýýi: iE: °': '

$` a üd'ö"m

u

Snöäi` 7ý$Eý° ° $' '$i[: i :; eS

'Ea NMI

m . i"q

} ý

p-ä on4LLeÖ: _^mmF ýan=w $ °f' g`"E m

: - E`4_=. Y

": t@e 88 _: }°, ° x°E; S : e{ä8s! g

° e$: '"

m$ 8ö"" }8 freqi c

68 d$eýi5 iSý"-

m `°c 8 ' 2 m

¬ i ý= S»E

ýcel }{ -0'' 8 °°! °t4 S3

ýj' 3°

m - ° ä_ý}p acg. "ý_e: S 4

co mg gEE

.. i$ n So

ý - ¬E

s e Eý- iä=g n�mg io°. ms }

ý f "F" ii-: 8° ^n°i

bL°, -

ý8

$ u ¢(e° t. iFý E '' '

F< : _

E»ýd me@^y^_

¬$"Y - T ns_m°c" fS m+`cm° n]^ Pt`°ý L

$ i' sm E `} c C

dE Fi E}°Lt>SSj' ' _

ESn .uý SS ,

m"i ý F p $ E` Eäü I= °8ý` S ä m= Fý 8ýcm` 'ESST' Y°° ý

: e: 8. r5° ö}ýý S ° g m"

m söl ýB

O E ! `'n _' ° ¬° y ý"'' i2 F': E 'E. E84F F! t ESl8. " Y6 $m OFý 3m3 yf E^

ýt `-° gSýtC r E_- !t b ° m m . =O ' ei' CF Sfý x }L<zýe" i- ý°' `

m E°öý¬EFý op im: b °e^°gy 4: gg; }Fý=:

S E. g} _ $mE44 ýýr It iä .3 Seö °'ý j°-8ýýý.:., '. : ýý ýä

g "i Ltta " ' gB. Yý"f i4Sf

L: ES . imp

4`cý': 34 ý i

E S'týý; t t`° eSyS ¬ýE

ýFý;.

" SFF ý'F EE

-

,m - Fse `E. EE S ýý 84ogeEfrm°F: -°`. m - Eh cpö<:

" Cm"e° $°ie cýö i 4o

! ý

ögi [Sib Q°ii t t3 L

,6 Sý`F! {c

si E8 °e' ¬

z : a E°Ey°Ypc. "Uh

SR

ci ` Ea I -a r° `'SS

Si n{ i fS 4` m c tý

°b yumöiS: Q°k

53-

rw

cr ` F' Ee °

°Söä'S°L2 ° $6: m °"

, ff : $ :Y i"

8Y ei3S = gyp FEE E °3 ` °

'`t°}° F f ý

M" H c ey gW

i$ESP

p :ý- n$ ''Y

m .i ta ý: $ - :;: ý ! `-: : ý[ F$a°:

$FY.. ýg ý"

mc e m ög ö_ ü8" a f4rY ¬;

mm 'o$ifý =g m "@s. äm-c F S:

. 3 ö. ¬�a_i

°ýL .

f- s ee48 d " i j " °F . B ¬.,

- ! }'iS'"ýe"3` Y ýSB" °gf8r tE. °L

^ mö F [ m., 8-F^ý ý s'2 a ým Ec" ý"}ý}E"':

p =p"F4 a 8 4

" °$eý: ° : ý}t ! Eß. d ec: e U. ýgS3:

eLL¬ ýBE. ¢+ý: _05_m_m3öaFä}}

_, _ Yg -m WýS: ae3ia b 8D4i8i

Y {E°°ý Ez B: EE. °" EE __e

, ýB°f H°y°mg'` 3;, LS`S o . iýp FF$EdBuTea Eao 38+SF: Sý34.

öö

im ö°

'» g

3 F 4c`g '

Fö Fä

a 9 g EUC

ß` m

p

£1 111_1 1

511. 1

Siý; 1@ 11 1

i $3 8i 5 :ö 3nm 3 : 3äm Sm 1 ö

L m ä ö n ö

m :F a

t N

mi°o @ meFö§

5 m

Ä

m . m

°gYä -Y

gq $3 Eý

EF $ g I} ! h

si 8 f Bain

ö 5$ä

40g°m p °G"8 i p

m p :i

' F`e°S

ý8¬ qg p

ýg$ 8, 'n

sl H

$ä f ö 5ýe ý 3 ý rýýp y `

opi » c y0}" - ii. ä4 E`3E 4=L3 E $ý

N

$ ýY "le=S F

OS>ýý 1 S;

E"ýýäý. $t;

mg

3 $

T. ß Hun,

hil__ 0 111

83ýý'`Y ýiEe$$ S ̂ ; Semi

L

'g

4 E

s

iE 4 3

4a^

U o b2 '%

o 7ý :

z "

p 8 ?% -- -

ti : ýf

öi E! = L is iý 5 "

"m E Öý E F` F$ E iüc 0 s F

._ 1_

144-N

N c-t N

cl

'5 U

Cd

2 w

s.. 0

U 'C

ao

cz

121 00 0 0 N 4-1

CQ

C. "

N

Ö O Ö

l7 t7 t7

ÖN

rn 6

öömý ýý

Cý

a'a ° Q"a "++; ° ý

c ýY w ý S°

tÖ

o ti '

ý'"ö . L'v V ý

N m N

C

6

.4cC

i" p°m- 6 r

N C')

G NT

G

N

if ö. E C ý y . .

o, m L'ýv p Oýäq

°ý cýýN"' Ný

Cdd C_ ý` aög`Ö

ý "c E d

. -C

ýc mýr

° LL ý ýd °EA�

t öC

r _ ýuA gerýý

"ý icy

y C Qý Y

-dd=a ä `

^N'ö Öiý O O1'ä Öc E

mS . 00-

ö ?t TN �

ý, oc CNv 22

E-m ýyö ý m: -'

3 s -. ä E °ö' °

rg. öü

6qCq

9 12

iy ci3 ö m. g 15 M "g,

ayr-

`- aU 5Et w¬ý ý' ý0-1mc2. 3

Ö3 rn öü "S nÄö3öm'.

cäý3wm `°oý

cý ` Fi ö&ö 9°ýC '°' ° £ . '°, 8 c

"m E

.S 'c Yý`a

ýäöasý, QQýýrnýE ý ý V

dcc ý =mEmsEa

. (7 °aR ö°. 'Ev' öý8ä3ýi °öý in E=n E. E iE

äÖ N

F3 y

ö N

C6

cm 9C

O .5 mp CO

d dL ä$ rn O1 yýý

Örn

m m'

a n

a m öm " ?: v . 3öm ä

, F+ co öC ý S¢

,2 m

n ui ýR

ýn h h

ý- Y

of

LL". ý.. cýSiv ý'N LL

o O1 c ! m v cE

O 3 °c rp ' II 'g

mö Ln r.

w J-eý C

?

ýý'E -

G ýJ . s1 "ý .ä... E ,ýo

Ec

O it M5

5 ö

c E =

rý E

ö

ý" J

d tl S

ýG

a= d

a d

co d- N

cd d

v u

w s., GL

s..

O u N

'C

b0

cu

Yr 40. OD C) C)

N

u .. O

i. i rq

b

O q

n: a ¢8 i. i

: 'Si :! -

°" re6, ¬:

c ý° Cep " t. e; F : öä, G : rö ät

v 2S' t'

° C2ý:

='

!F`. " °

ý o-: sö - ý¬üý ý{äe

"2° g Y. E : `: "PöSEöS

3fi": 8 3 @: s ý v{ý

ä ä:. ;L $S " tý '"2"' °ýýE: "

:sý ^ ° Si 5: 3 ° " {

ý G i n ýö ev {ý: °

ßäY' " P

; fi : 5f . t:

° fi¬öf

. $'E !5 ý, v ý`f! f ýEa: F E£

"c 3=. c

S p -e°t: ei"S.

='ý"ý" ý B G[.

a j

. $°. (p"`{ 3 £ j

d 3°¬E s 5 ; ¬ ä : ö°

°G °

ä_ ;.

"fP `: f"Ft 9

. G" Sýý+Sýe: '°tji' sc°cF °. o$Eos 'ZC, F""F-""cdP. °c'

y° ýýE-`DCb' fig.: e b"e t° ý

_ t - 8[: E i g

f BEF ;" $ÜS: E

f i"¬ä°c,

" , 2 E 2 S"; ¬aL S-`°. ° ö °: :F

S _G: °gr" ", $i: i

: `" 8 , Fýa:: n X: Eevc"

e ,ö 6 of

v °: S: ý

E' ' ýEý; tý: a, vý S. ýF"°!

a-ötssY `- ýg"F 1 $2ý° 3 iEýg i

$= , an¬ `E

H a ESF: ýes ri

°^sj°=äS :s3

E " <auE

'ý ý°; y f, a c c" 5n $S, PYf P; ý E°- S"

E3 S s S Ei E p£! S°t

ý ° Sö ä5

E: c°2'- öf "c

: °iý ý ": . ;;. ^5äc

. EIA cg.

", $ý"E, D

rý" oý ° eý

e

. . ý v¬ýrE;

GiS ¬" [FY: ` ö

g = SF F

:" S{äFýt"a, c

P ÖE r

e"" ^ ýo ` t'p ýO: °d t' _" n; +

3 : ¬i, öýäýiývwi " aS5 ý5°t ýF` 55 ý-

S! f 9 ý " ° ä LF i+`!

, {Pg"c: `. i' ' "" is o S

ýý ý_F: ii { : iý. "

aü E .

03 ýý=^'` $ : a3> P.,. ° E

F E:

§. " nö SSi$'s2°'

ýg: Sö2 $

: öS ,¬ ý-`. -`0=ntýc C° ; 4.

ct: eý8e7"F : " « Y

5: "! _ E " S , .

s¬aSý.. j`'z:: ät `}

c PSL`, Fý"{¬Dýý bä

=m; -c ; vä""" iýýi! St cýcq fc 1

; g "[.. _; y - "ýe n. Ec oe$. :g c ý

" ö°g- ¬ c5,

°fý""Tto. " { 5! ".. s` "e

c t"e F . _f 5¬e"E¬i ä, c"ý$ý

$°8+"' [° @B1 12 a _ S

Yg G5 tö4 eiG öüi Fc

c $ ;. : EEF H

° :R '

° w: E ". of °Eý°F°ý o g c :

°'«c"D' : ý:: 7S ö *e $ ",,

t 'F°°F P tö;

-°"`ea"pf

ö l "°ý +°ýÖ

E g c

o r ý ä T ca". '_«'iEE

c = Y rim;

c g

+ . "iFGicCýý: fS E

M, ý£'ý 0, 27 % II"ýý : ¬6 :ý Lei : ;. yý am:

SES{¬'"ý , 2I8tg ýý`9 . 2E: b äýrg'ö¬ý iig

iD : 1 ¬ ý

:ý öi 6_.

; s. E; x ¬ý, C-` -s vF Yö si E ý

^ö pCbýELc - "$ E- S. o$ "

E"

,° isýc : ": g:

bý Ea"E

výsz2ý'c`RÜ "c 8 ý' " g`l0 " ° "

. p Y E"6ý"ý"FFaS

°5g " ýF

P " "y Z ! P: " <$t2; E

!3 -, E

«" 9M ¬""

¬ £ ý s

: _

¬ ." � y Yal, _° =a

e{ ":

'"jg . "a. "

+ ý : ff Y ý[""

E3C i1: Y" " " ![ . 20 Sä38 fi33 : ":; ý ý` E. . [ { E t ý ' ý @ ä f t 8 E? f 6 ü lSöCn w: cE6E3: ° : E<

mßää i: iwaüüi36 d: ä G£ i di g En Si 5 EEö! .

Sö: it

Emö fö Ei fi gs9ö g

C 3 3Öid 4514 I'Mi 6S

MUN mä 9m 3m ä X

o i X 0 6 ö n b r

e :y

ä ä SP'`' ý

e 0

N S: iiS3! ¬°ng

3 1

F,

0 ; HM

g 1 4 ` ýF " ý

ý ý B5

Ö

8ý

. o"

6ý i¬ýi z ä c" E t SäE

Fi Gg °3 ¬ " y 6

fl! _ FFä F_n SEFS F{wES N, GFvuC

i 2

E o

g

3 ý 3 Y "

p p y 1

L ý 4

f i ä f ü

m m m

rn N

C) Ln N

cii

w aý

aý cn N 0

v N 'b

x N

a

C°

Co 0 0 N 1J

. ýi

u v

PZ 25 10 r. Q)

V °

ö ö g c

c . o

d

.ý D_ N G

M ^ d G 6

°pýxý Eö rEoýC

.6 LL rAýä ý-'

ý°c Öq=ü

aß'9 Aý ý ä

°>Üo

cm ö=°= E

ö ýý

Eqa QS

ýý

o°° a ti' °6 u

ü L^ c

- aý z

4 ým E 2 Z

Ern °

f: i

ö cý ö cog ö°cg

d

cýý

u °'

°ö

SU .QU v" ^m

B öl o, 1 yN OI . yg yY Öl

y m a fp ä 1

c Ö ÖY

ý ý N N

U N

ý ý

W-- 0

N

In U'

^

w ; G" @

N

w

0

C ý . 12

C =A d

F+cä' E ßL$ ý'3

_t E= cý. L ý, ý, pP `E'ýz a

cL ýýö

AU h

WNW dR

d -- n$ m m ' rE '" öýÄ

cw

ý E rý wr

xE n ,ý2m

ö 2 dö

'

° T

d=i =ycCC 'O "ü 'C..

"

"tZ U 6

E. i E QýNd EE E

CEE. - s o o ý;

E E ý, 5 S

z LL

U2

Ü Cl

r-I LI) N

v

w

Z N 0

b0

Sa Z y

20 00 O O

. Sý

"ö

b 9.1 (L)

n ö

n ö

ö 6

R a

ö

,p 'flS i

qF

fi

faf9 ö ii 0 b4

I

f9 F E

OF

c°° 3 =x, ýýö n

gn o n d o

.R ieFý: n n°nx

, iR do

Ö¢ 6. St f: °

`i{is "t E °( r$ s b' .tS i' Yq '8 miE

ýt8S iE '`; f öý2 r8v ým`°S! ° ' 'g

1 : g° '"F P

F ; BEB Fyg Ell eC "' F'°

LP: Bg 8 r

:i ift iq ig

q6 gSy

g`( ý'E P Cg

Y, ig- lY {{ ftP{3" Fy rI 3'8° {F IF, G'

{g {' ° s. F` , 5.. °_r°° [_. t {"" PFE)8' Ye 1 fý" 2 °l

ýq ?=

FF iý tF i

2 Yi yyy t 3 ý, p F{i ý$g

° sy cF i l Yf g 7 ý

e ö ", tt" P°?; (eý `t ä3 !F g° { ! : i { . °{F Öý : ýi s =F p ¬: a g 7 r

8 (e' 8 'i S 'ý ''

öý g 5R:

y c, ý { gY EF ? °!

t° Ei

ý ! -{ q ? ° Ss$lý 8 {6

' ? = g {{ 8i'! 8

"tggý °' Hi

!ý- ý

ýi. 6 'i{ ý '{F °ý°.

` 'gg ; ý! 0 `1 q1°''5°

i 2 ! : I

c4!

P E tTi

'j %'ý

f°1 y i8 ° Y, g! m;

E i}

n i ;: , a : - l: i ; t ý q ' ( i ý l E

i t o ; w t`

. Ell¢q 3

s: S 3ý-8q gre'i )°E:!: mE "ýE g i i { {

" " F: P °:

i y g a = g e: E

iÖ : t! 'ý `d

, > r - t { t S " ! a g m22 t . ggg tenv: crg iq!!! {8,

btv. 8 °ý' .Y ti6 '". qE ° E )_ )'g`' Ee p ' B """

j E m .

{ e, ap ! 31 r. x Y/ EO" fp E :q E3{ {w i $" e! EE. P? gtR : : rF; °' ?. g E' ;

- gq ý° b '`g SF°$f ýFq. fl8iT E98Elg Ft ý ü ý -°!! ý ýý3

iýgy 1 L aý4>ü: F6. l. 1 'f-° "

ý ý

° 11 , F g xg o 8q8.6; ß{i{. ý. ýf °8gjE n3.3ýnS {iq°i iý F

?_;. san S' i`a ° EFrýla $ E_E fr {{` % ý¬ý { ýd ý l44, e0

"

_ i. 7 i6 >i ({Fe n7 gs3= F¬F ¬Fý

. Y.

sf E"E ý. ' a r$ , Err:

m .i ;°

. ýe i{ý Yýýfs3ý ps 8'f t5"EEC E' !" ")itiýw8ý'igf ä°ý q?; 3 { + '

git E,

ya q g E'1 r "¬36:

Eýiieý E> s °L :

. i P°, , i. 8 _n g

: gR: {, E °'jFgg S ýiý 3': Era°. °cý; g8ýe

{ii8 t Si

Q{ 2' E{ ý ig {{' 6 3E ' g ; l =

, g y ý {f ý' m1iE3 7

D" . a. iE nýßi qý

< : gE tä° g ° ü e

r1 {i :t : . wf°e{gSýl ; sER

.8 'e7tf °p

"8 j °u:.. i'fy ": Tft iF?

I i qg9? u°

t° 1 ^J !d ýý q"

ý ý c; PF ! f°$ F, ý

r m { a q '=ilýa!. 3 SF: ): F° E°t

s

! t R { ý; ¢ 8t{{Eý'fs

4 3 gq 2 i p; BýsP; 'T'°! ""{

meat [Ft+i 8i i . Cl Ef! F{ [g 'n "Et6°g tY{038. tEx is rP {{° t iý"°u.

{ ýFi. ' ga, !g dý" ä' t -

¢ e p !g m 5ý{ ý Y"Ei7g; 'PF ist ;

pEg :' 3ä !5 {6ýý ! i= yn Em ' S

c t 3ý "'{7°E: 'b'`r{ý; ¬ iý{g f; 4e ¬ %ý i su Yb{y C- r. gp P °

gý ý yg ýF ýýgi;;

t6; ` -F " Rýl " q

ý {t < ° ýtqö a°i" t4iY

bfr gli3aýýýt %

8g;. 8'sjit ii p ý ?gF ! i) ýTi "

, 1

fF iSF(ý4ýS qa

ýl Eg ( it ! ý. t f.; ýý

t g {E: {8! '_{"5: 3 gq{i; EE : 'i a i. `(: ýt¬a vi 1! Y t.

u#S { ý{

:{ T

¬!

zä:! i i{ß'81e6

F6S: =8 ¬ qY ° '

i 1 b g6 °'?! f? r"¬1 a ¬s.:

v P5: 88Y iEitb? iýi{°

i'aR(8t. ýg'i

S3', q"=: =!: g'

) {ýegge

I atg { ' =

_ = ? Gfii 3tx° Ft4t

g£s LS'C "° °FS 6P " ý ö

ýý !F{2 ° .

ar{ . '81{°!. g'"4 Y SeaE{t g $t g; t, a s`i "! EEt '

t 4s "E

E . Ell- '°3 o, t' ' .

Fg g ta

. i Rr i ýSvi i- ! 8'

ý° '' C{f

tigIfR Y{{{.

If

i"' 8 Y'n°" ý7T,

"°".. lBiR EE('41; 1

° t¬ý° "!. -t ä g°{ : ° - R

ox m gýF.; reu-i7 Yw{ 4

m

eb i °[a { s.

fF `

` It 8) F ö°{ °iq6 guy .Is gu q ßE je

"P, , p; e) [ifa

". - of (

i

_ g; D öfF-"g ýg

n

nn f E 6gr r, 1 S `: rgrgq { i'? 5: ° 5

g 3{: =5. ', elr"F1ý; "= g""ý; °. $j'R °` '

1 0 p. 3i -! p#"ýA nt IE

4p Iää, ° m

; iS ¢8ý 'S

3 g tt'rö8?! i et E ift3ää! t( . ! {° g '

! S <; m 4 : qgxf.. °. g e q{ 3{ { ! t° 8t 11

I1 1!

1 Fe :ý ' ° [S! ý! I °{ ifwwww= i l:: ii a %8P q`3% :n : i? f : F ttfi9rýb8ý 1 ? m{{ I ä:

° { 1 8{ _ iii i¬: {F 7i.

m i =R S' giEs i? ýn Sil°ýg 4tf 7ý

! C°j pF

p tG u

5 {

7F H

i"5" 3ý g

25" 3, -1

f' °,! 1

0'pýF°

E{a i ° :ö y

:3m : 3gE : 3äE 3Eä e :

o $ t

la ý 1

5 ö

R ; f{

S ' 4f ýY

°"E! H!

0 0 0 m

°' S Fý

gs 1 ýFý

E öý oý ä ? ?I ä

{ ý' F} 7Y3

i y

E. ýöx ß 9ä

F! E gmZ

s 'ii

ü `_°,

4 8gß{ýq cri.

i 3!

.f

[ g : flag

ý

p Fn q ý8 rC ¬:

ýEYýýfqýB

! p I JIM F , ý' E (j 1

;; gEýF

SME ä. <ý3 Y öiä3Bi +g i d6 ¬{: ý:

Rýýdi

t g o PE.

Ü F " f

° '

1$ 2 F

! d3 mý a

0

i I ?E

Eti S H

iß

`E `ý a[ ¬ 'OE

0

R, r L M1 A

«ý m: J

U ýe

Ü°l

b if

v as

Q'io t

Ci{Q ¬{

0. °fi f i Q 2](

N LO N

M

N

a) U

w

t1,

0

u C. 1 N

bO

CO V

f-+ w

OD O O N a. + N

U 'c7

'b 0 04

d d C -L d d

"N

C. N N

dn öe w°A

Eööö c, a°N ým ää. N9

e6ýý E .E 'O e '

ý

0.

ä

.N

. ̂_'ýý»$¢ä °

g 4 N c

°ý'a ö5 ýn Nq$" 9ýn . i'

°wSý, F , ý8

.ä ýö5_ý° Eö `°" .. EC in .2 L"°yeq

ä ýEö Äý O gy

uE SÄwE

- °ü

g ¢ F

ý "ý' s. d_5- öäfu Yw2W gö 6äm'i"ý'rn E nm3"ýn

ý-ýj mi"rn

ä_

NE m c P

m

. ýEq oüc

cZ'ý.

f5

°co°E t S4

7

Q;

L Tö `° ý i

. CO °CÜ E$ °m W. C' ° ö.

_. ýt=m'ýa3c Rmq

ME öu°:

-

S Ö

m .,

E E- j S -°. E - ßä d ° 'Sle 2 :n , ý ° =; 12 29 c ýs° ýä "-ýQý'`ý

. ý.

u äE

g, ýý3°ý ßm5 Z-e l>

'ö

ýýu$"ý iaý. _ . cý rv2. ýý z°EÖ xi4ö_

"° ` ä@° Ä ä q ý

r9ýsq ý Ev ü

8E 8 ýrS qý °L °c C

iro ° äö ö ö 2}. 14 E

äyteýEý ýý,. cdc`s1Ä Sýä ö °' . ý,

3rvä ýÖ c'" ` äsc°ýüýö - üaý5 ä ß: -ö59 °5 _¬` ää v

ü E ý ýy

ad ýcyS

m

Hf Y C. Q "2

ýD) ý

Iii m ýýý Qý LL Ä lei.

C E oCö -ý x ü ýU Lull ö

11101

ö c

ö' Ein Sý 42 tE" fi`

12 ö

o JA

E 2 Er öb os

E 12 ö C

J

E

T O

C =

Y CI

ýý

O'L

QSý

k22 g

Q

vv Cy

N uQ

d-C d tV

a

ýd-

N

a) U it

s, S-r U

a v a 0 ci U N

'Si

.. r

b0

h

4J

W

Co O O

r. G) a 524

a d

t{ 7 FX A

F

Ei $E

° Ö Ft {i

q Y{ a a E -{

1 }y

!ý 5 £ .

,a ,¢

;!;; i ¬ g is'F

"_ý ¬j 1v i

{td ý

:¬ i ,$

psi n¬ `} 4E !' `gis .:! ES

"3 v E" ät $ý E} iGý_ ö ý{`-Sä' `

iEpaE;: g« . ¬ýE{¬I,

}s. s

«ig _Ffv

a, i Fr1 S ; °i2 ts i '° C n

P ¬t

«iEt%¬. t ý. F '} i f , , , ''$`" ie{s{a; °ü i' ; 4? g 'H, EI{ $4 °:! EE ý F I ! I j , Y¢

j; iýa! { Y(ilü. g SFF$I{! ¬! t P; ai. fiE EfG , stE"i !6 f{ 'E ' { r ,ý¬y, ¬r r1ýf r, e¬E {FfiE

Pc£' ¬'t t} P"S 3 ý ¬E j c2ý ¬! °

ýii iýi 5i :

f ` ! Ei , - $. fý t°'

, {" t' i; °2f$j°HiyE

F'

; : `'2t 'aF. a{i a . + :

] g

'n { {E

g .t , .Wý .ü g}° {, YLP+ iý ic'i E: ° 1'c$F-äS iy ý. 't"i E, S t '`ýýii. (!

i1

1 iFS3 y° : E¬¬: ýF

g{° ¢ ii'"1 cc "iý ' F ib f F'

e i !

° j . E

"7 {{ E} E Sd{se 3° it q$$'ý1 {E{. 1

24 iý1 a¢t E¬¬ {" Tv¬ ýEE} EP £i. F E

ý [ " °

i. °d E"ýv.

re j{ "+ . 8

,; ? ¬ <iF iä :

_'[ }" }c

. t!! ö"E ¢! gi ° Sitiý7 i! F n' ¬ {ýi

( - GE j!

i iý j S S U , i ¬E«>sýý (>

sfýE° ä: « ý!

$iE1Y "t `ý ¬F

. b=ý F: vi{tCtF4 l, 'tg {i

N{" t ° {

E E{3 ¬ f

ý"{ ` ii °"ý°

9:, £ «' FtS tg

v ° 1.. f'ý xb

±=i $F Fv°fa

ý

3°$ {¬ B¬° ¬-°¬ ° Y ' f ° t

, 3, {sE { 3ý

, _ ^$ý; i$EEtt

¬iýY` dF{6 ,E ̀S. t! s¬ $. {

St. {!. F ! E"t ` ° ¬

°E1{ E !° p f ýj{ý it

ýi°ý : SY{}S. ISIS

¬voYý{ý{bfgý$ SI ý"7` ýEF'{14° °r E 8j I- ef$ i {iý

y q c tt

} o i °

3° 2ic{ icý

=2%". i'¬ daý°'F

ýrýdEý"ý¬t ýi "'S d" {F.

EY $a$a

t

XF{" dF "Pi Ef °; ý ý'i '. p° °, ä:

. 4ýt°`E!, ýf ii! ¬° E{`FL' 1; C

7 $i °`E¬1; t°p'LE ic $3 "

Ec .;

ýE p , ` °i d2 `iE=i{ý

E¬{ü ýäE E: EjSSiu` d! °ý. £iiE'4bF

. fF

E2. f ýlý °: E{ ti E E14°A E vY J ý E tý 3 4 i ! ]i r{. E : i, Epf i{7 ß i'ä iE : {eýF9

F'S'°' d" i ¬` ýd

mýý°: j t . F' E ¬ S6 {`Fý! E " '"

{y ý"

iF! y ýo

r c; S{: yý t , Et £s

ie , ¬"S.,

H! C # ýtE yiý iEE-: °{Ca "ý ti re Yt " r. {t

F i . z d °i ci . i° E: Yti9ýFf spa Sd,

Eiib F: ý ( f °'Fký C F$ic f4 Y¬°{...

_ EF °" FE, 'S °; Ei i' i ' a ý $ ' ` ,

j . (I }( i9 F E ¢ !r °t°c5{ý{S¬E F{ {{ i ý

5ýý '"ýU {ý ýý£2'

"" 3 ' t ' i63

H ! Y: SE

S¢ý {p {ý. j {; E {cGiý"ie: iF° f¬ ¬ES

npE E'¬ !! F} ei j Y S{¢ i1 I y i E edE. ü¬ Q }S! S C; t3¬S! °0 $S; E {'! t p1-ImýýUEe

}2¬ýý Ef

gy f f Iix $ ciS $ Se°`E 7 eýYbi tt {! `¬ ýyi

W :;

F

i E

:: {d 2FF E GE { ¬d i ¬cý Y tHS ý ý

. EF ý°3, i« ý4n °2 týv s ' i ¢ ö

°tF {. EFE`

E' a"n; a isc F E 8¬ ýF F] } }ý ¬ P

I ii iEFý ý ,a

i"ýtr E, ý ; fa

ý° t. }' °F. B e7F 6" E EE¬°H °

ýd F i¬y

ý` ° i c' ,

ý giE

�{F Ö

{'$ Eý¬ {$i} eý¬e`"E E: Er¬ýr`S °äý

{

cES-` " { ( !f tE, S{ "f' t .

i ° ¬¬q

N{ ' 3; F ti¬E°¬ di {pf li ýC !E ciý °t"ý lic

; :J

_! t t {{ E F 9 E

y ¬ 3iýi'ýdt¬°, s : ¬t

E`4 ýg

i{_

ý" ý{p 5

¬ý°6 F _ F

, ; _$ ; IE! $ ,

; l i g ! S ¬¬p, re

;

SYt ¬E$ t¬ i ig Pd`ý{

! 7 t¢jig!

EFEU' liEi! "P°: t: ¬$S { tE' bt. E., g6 Xv. ttn

ýg Stäý: träg: t¬ýS 'ig Sib JI'l ! F1ief !

p'i i

$iW di öid d3eid d3 väd d6i u

0 L 0

t

, °

I

3! e¢

{ö

r <E (iq ý t ¬ ýý"

° E{ES¬} ý''g'

; 6j t EE : °¬sIlFý; E °

3gß {"« Fi Sý; } « SSSESF. ; Eý F

E . d, EE", äý ßt3 =qi F{E

I}E F{ ¬¬ . FEc t

c°, ¢

3

3 S

0 ý

g p

° c a {

P F ö

F G

F Y

1e ¢Ly i {

N

Q

L N

a) v cs

w

a)

a)

0 v

bO r. s..

a)

s., w OD O O N '. 2 ti

v a)

Ü

C) tý.

A d

W d

'

e ö ö

ý

ä w

n

a

ý

: ro aý

NW

ä

N

m

;ý - öE

s; -_ - or 3 °

-° » m»» ; S; EgF La ý

°^ i

Z. ýa °E

ýö: "ý? 3mý3-ý ° aýýý"-@ üý'm-ögü?

p m" ü8°ý_

g°omgö CE töa". °E {ý¬ ` '

öce° c°5 ' 5c g'" ý^ .G qümEdm Sä: F s Qý g. gm gu

aýý ; T{°ýcd

,£Y, P°g 5'p '

ce "3 gmý$o I°8 B. os °3 g .

se It ý I& - ° ý

ý ß m om $ ¬¬R i5 w g9 .1E

[y "a t. rýz ö: mg comýwmümc ö b'3o ýo `E 1

i24 E S¬3 öý'. gc ý ö'

ä°g0 4 4P m-89: e ^3c

°a Sn ääý gg 2f E -' vý ý: ^8gýmý E°

ým ¬

s _ ¬c: ¬ e_ Er qn aýa°»'�¬4¬ rý Y: B ¬ýrö°.

8 äq' ý: e c

mCB$ ": ý

cö °; ým>g! ý_ ooEpa pEaý

Cm ö °bö -c dgmSa 5E5 °rýýF¬: E

a Ä' t n ö _ ecäc$`oýc öESý eQ ýIýýg"p°c pý 'oSF

c

IjsS Pi g£f gf: E^ä SE uB

' Ci

V q ä: ry ý ''=ö=ýöömýc° cäSä

cýgm"=öeäý 2ý' E z -=aBäT 'tiä ýý"_ ;"4

ýýF9ý`ä F=`e_ýeýi

g . öG°m 5ýS . Ea ae3T. -

mm°eý 'ý° " ýý ý 9s m°8°

-O d °ý m .

., gjýü-ý

°"ýuýEr Eeý öLö3VÖFä2ýeVS 'ý - ' ý

ä öcE °F^'ý

ý ý p p j g

`ýmý _°n 4§ rtpg ý5 p "

F8ý ° ö Eý G8a immg SyLýi.

ý -"c_o°b«. _

¬ic$ 3c omm 8e°. m

E eo

ön 6 L ` p ( o ýä-s c n. G ý$ L

a^ýS68F ä ¬ ý

üö8 . " E i° °°E° 3 a°ae=` ý, °mmcY 1 gg ä ° "

m_ýmLU_'mý

' ' "

4ý ýN °$FWC söF

E

r ý° ä2 3 8° » g -EmC rö o-E

Ps INMIN 1 m , cý ý mm yýe5'

ý °. E -ö" 5

F q Lý ° 2

ö j "'. m E d! °, üL g=`o -1 ýg: 6: F i ey°'E m ¬ °gOgýSE v=F

ö$zgEg g'°°

i L ; gfl $ $o yc- cis

ýsg m _;

ýrm° _ä°rE0; g

a ¬s°ýý ý¬oY¬ýý Eäz8 dm`s` 8ý

E gs ! %¬ Bgmr . ̀ie: :3 8ro

-mw " ö ° ýg .2 m ým F' E_c°E°cc°z öý öE q rys @ g_ge=ý 8_°ý äE

E- ý'c g »o °! °

E7F` °ogff"ý°g°' =aýgg¬°: s ý5 z t 3. sa ýö°

;, -M "6 "ý¬82

ü^°8°m_g ä 8m 8o >ös. -_

S nsm ýmÖi-

ö_° FýÄ3g q O, y ýQg., II

<m mFo3 1H 1-1 ILn -

Eä$ ö$58 $ ö ö$ .. so. F e ý

O Nýmý y

2ýIV ýN gýfV g! N v

ýdl°V g

1.9 L

3? m o HE ° ä° �,

°° äý Zpä

ä@m äl asa

m $ ßc

ö ä ä7W 0nüm

0 H. m

pp]p, p ym

e äpm u äpm

nö 0 ö

1L +

N ý ry 6 $ q

1

n g

K CW ä Q Kn N

U U a LLe a

S Sig - 8s r "

° ý

E @p e 2 a m 8

; °y Em°ö $¬ý : Eý °ý

° g

£seg'c

Eoý; f aa

gmF 11 2i 'o

Eg

s . ig

ý ° öc P E '2 E

rg it

üöE " n mm SE's'

E c? $

E $ u. °-°c rcö

ö m c s g o s

m ö

Z ° ° 8

! g c w ! ä

Ü

N. L N

c- a) a) C7 u U ice.. cff

2 2

S-ý C)

C) it

0 U)

'b x

to

a) E

10- Co 0 0

u u .C U äi

10

a) tý.

ýr ö

7 °

E B ö ä s ä " E_i3 ° 4° 4 p NSf

iä

o. ° ZF

oe¬ 00ÖF

13i3 S n y g -I $E¬ýE 5: °

ý¬ ä äc

s5 ý°ýöö

ö ö o ri ^ JO P_Lýe. G" G G G 6 C. ý

p1 tE. " . >" 3i ýFiF3

5¬4p8Ejaý '$ef E°"E "

°o ogfE

3 IIg

`o 6c" ö: g2 3c e`eg

tü äyE

=ä "°

=F °

"'"ý, °

=t SfäE

ýýg uß

: ÜS'i E{, °mi

ii Em '

gqE ýý ýee¬ s4

3

Fg. ¬ ö n :

ýE'= hic:; °. ö; iyý ý3; iv i: 8i "° g°E -ýo S-Eio_>>Yr

Oig-c ý o o= ý°E 2 :s $

I n 1 ; _;; 3- W" ýý'Em=' t 3

j ýg 8eE ^a°o_ t g1

S ß ý _ LüÖ. "WÖfB

s ° ogý°^= SiSý p

ö88 il8 g°8g S89r °

ä

11 ° Ei to- E°az_

! IE . i = 9 a$ _° si aE"ý:. H

E 8ý% ''3ýz8öc ý2 : E ; E

g _c P° pgEý ý °tqQ m ''Et_o a[gg's_° ° °p

p 1Ei $°E° gdýý. = a ä5 y"

12d

"F ö¬ý 8>ýt°

o $°. 3ýi8ý `ogý: dm"ssýO

'° °:

e3 Fiat" °: ýE i `; M1

'J" 1

m n$: . aYqýS°p y

eF ° o

° -iE o

; E:

cE"-i' ä E"°°! e: ¬

i°ESS gg°'D ý°

$ EESa. ýn°

2 °. tä _ $y. tFg'

ý cft Fe¬ giI' F 'v

. ä f Pm H

° c

to 8

wE ä eý ; S¬F ¬> E; ý°E

''

o. r c:: ý -°

.{p:

H1

2�=ý : 2 E"

c' ""

E h` S" p f . ciE_

PI_ : i8. "t"'

t A1 1 I g

E- of sn um

ý¬ °E =0 C

E. t trE@ýF. a

3ýLý{ö. ä L{' a. F E=4- c r-: ": E-v

n 11HI äs$ cdEFj "r" ¬. F

°p8 nR iS. s , ö-ý

öE- F

:" ý"2 eä E"t. EE:

WHO

f>m> . 25¬BT :

3

.

E Eigm4('0'v

a3 iPýC°S öN t n "

aFp" ä äB -: gi#.

_ c MINOR e:: g3ý

: v 5 eý3 ° m'Fai gE. `9II° 8i

E S sE g Fsc . P-s t^ä H=

0° 8Elb g °

c { :., =Eöe . ä" "°p : s' Er. i ý2E">, n ýe n°° übli' m e 'EP= m

£&i., °m2 .; " ss S 0 IE>

: °- `F

Eefä ý ý

Ec t£ B c

°

s

pE ii{ý FESE

ýi ýEiý

. mf Eb EÖ o3

Ei Hi

8 e. - ä yc= f "a°ý'S-

9 . °EL ö-

IM :p

=$°``

8

iu"m f¬ý=i: öE

E¬3 $

e . ý c. °¬p 5EFýg. E $E

$°C ýg

. , m

ie C° X18=i ON 1 Yc i m l2

o "°, i ö g ms E

- 4 .

a 6 i. '

m{53ýi p{ " E3{ , MOO

' g i

sý ö & : EýYiB = E =ý Fmý8Fv

°f Sm E

e"h io3

pü r . öE. 18.8825: g- 5ý. 4aäää.: "ý8. g i :3

i lo E

eý m o

se ö

i cö

g= on -

t_¬ ° L°

f=fig ¬Hf ¬ ýUa

6 3°O

HU E sý {b0 $p0=

Il

. tfP E Fp E

Rfy s na 3

o eSi eSim° , i5ä mm l , ! S; G m m ö ö ne

o=

g Ef °

n

g e sý

oq

HE

°cF°C ý.

On y

On

N

t P= 3 e! E_ä

e

4ýq-£ y n@ e[S_v uS

" p 3äö

c£ mC

Esp'

d -C : g^ y "

i A

Lg

äFoi s e

ýg Oil r

S ¬ P3

ami E p q g4 Ei4 ö Iý. e 4; of f, E

g ruf z p sä ¢äI ü O 1112

. ̂ p I , lull 4 D

° Priö ýn ° oä Q rg: 5 ö MEN

° "F

q :

9aa

Bzýäo ýgt ý °Pg°

ýýF ýrEaäý

P Eýg°oý°Er äm

Bý_Fö ý="i4 ýfl Bý

seg öý ý

ýp E k:

E $3 mCýýýý ýRiýg g " -fy

E Fseý

üg__e öE____ Fý ' öE¬$8ý g löFt: I äiü1i Ö6 q

d'P mR

t °

Y E

Q c

q ý¬ ý -Q.

E^

X14 it Ü

ý° I ;2 P ; ce 9

y

ö i J

öm f äE f= i8 t . öf w? 3

s m _

a1<< r ö

_I

e 3 ä

a 0 2

5

P

2

J8

m3 LE U

y! ö

UÖLL6

o cpä

0 . 1,11 m goo

cF°`

a.! QM

00 L N

rn N

a) a) C7 a) U

W

a)

a)

0 U N

bo

Z N CC a)

E i-ý

W

00 O O

N

v a)

U v

cC

le aý

NV ON

ýÖON

°ý «5°cýE°c5ýn vi

uý W ä« 3 cý

c9 Sý° E. ý

al! :1 Qr¬äcöcä o 12

% Yýrýý'° öo°a

YözC 9ý C `m "ý äý 'ý Týö

ýýý. ýaýýý"E üýýc E« E

5 =ä

ä ý

Q, ä 3äm

rý ý_ ýs ýý oý

.ý

d

5! z

rd

f3 ý-B

w .. °C. ý°c

2i m, 9 n `c°°E 3'! ö °wE vZ. `ýý 500 Nih Jilt'

pm

`2c öu; ý! Lý

t$ä

e JÖ`me

° q$"E ÄS öLw m äý

C cD

CaN

.ýý

Ö1

+ý vm

ýö

' ýN öis ä

výY äm

IN. . sý

Sü ý6

výýv

as '"

saöý

_«E

öEo

-r ýý C) N

C) C) C7 C) U

110. i. a CL)

00 0 0 N

v a) V v

"v

a Q4

ö ö

a ä ° ý 1 ¬vx4äi

d

S! iJI

eý l :: ý w ýEEýfi'E: V `wEs ýsýc°

t

:ä p

ý ýg ýFEgo_ "", ' --, Z: ý co q

_5" ! °T öAS EB£y Y fi °w : äSS °Eý " "

SýýE Ez E_ YoLn° "r ö ý gäSE"`p;? gýý%ä ý

° F ° fi ýe ýE. ý ' E Fe Ssý -. ". B° üc " va cs ! äö fI

L, v _°` ! "ö

¬ x: EnE °'ý Fösr -: '. Aa: ýEý

- iýsSýP Eä°

c"E° S. aýä gs°£g-=! a a_ a ! ýý0n ý: ý<< .oö > Fc°E. °eE ý" ;ý ;

. . :P"ý

; äß ý$: F= E E a .}! ES Ee., SY! E_

,t 3 Pýt a<<; r g ; _

j ig va Yý[. j, 63 a

Fgý! nS " Sý'

°: äg ß`; c°_ýSg4E

E ý ° "o S

ä 5if "".

"5 c ; 5 o'^ý"öO

L

8. _ y° ööi" ¬ : SpT3 ß "ý gE "

F 9.1°-[E°. 8 c"ý! " : c; ° "53 2 c°SSSEy. . TEi . SE : E: ýzmöý: £ý8 °o$ ¬ E"ý °;

ý `Ei; px} 3 ý ¬ ýn[i£

ýG "Y ý'ý

ýs ä . °"üfi w ý°ýi'S°ýFöiý ü

:E ýýýB. fic"-tcý'i ýS

«`E pjcýyE 3. ý'

_. ý~`F'O F

y : 7ý E

ä

e Sý wö

xg '

<- o E ýyP ö y; c ýS5i " ý ä

E" cE `§"ý _ -E

Q °oý"oEýO

ý9 : --EE : ý1 !..

o"O a .f QI 1 °- " ä1

i Iu MI i F5FW%

eý'Yý"i .i

=_fi nýjäýö;

wSqö ö °=fi S'+ö: Sý sa{EiS: öPc°E T' l3

böýpy rc=ý°p ý96 F

.. .5 .a ExýE S3 ý! E< <lö!: 1-9 f§,

[ rýriö SE ör EEZ

Emp .E aö em .E a^ Em °c ý. Eö E

Ul 2G m 2ýLNý

gCä N p ýLN gýN

4

u

m

äi °sE md q

S an d ý "- 9° g m E

N 0 0 N N ýp

- H M

n ä `Eýt'E

0, 15'. 2 mvl

,E

Sm 3m °3 ° °n

PwkN

E O

öý° öý S E -0. cO 2,5ü°cý ýv 'Ö wE

g 8i ` B ý8m°nmý F 344 t 8

E E°° E EE$w wt`ä EE °ü. `4-rn Y ýeý

E aý» mgr '

sg -8ý 88$ ° °

n ý ýE ý 8 m? ý

SEE mzý

: SE EAg

v ya 5? Em °q ^eßä¬ -E °

- ý5

n° E f E_"c öE53tEcwEi ýg m° e LL rc

E ý 8 _ p C

L

I

q

__ _

q y Qý

m m ö e

N

Q

N ý1O N

C)

C) U cd

w 3.4 aD

C)

O

'L7

x

O

a) E

ý° Co 0 0 N

.. r

UU C)

U

"C C) ai a

p ý o

ö d

ö ö

ä

a

äý cc c 'a°Ri to'o Ký°- m

o

v

a

Y

=

a a a

y ýQ

äe-z ve° c '°_ Sb

vä v° EC

42

E °Za rn E Sý ° Eý

ai° E! ý_

cQý ai

E z Z Z

Eý$ ö° .6ö 6m ö°cö ý .6ä

qcm

CJ Z . ýr' yN 21

LN rn ý'

9ý NN

2 rn 8 rn

N

y4Ö

N

a rn

O7ý

-$N Ö^mNEm

a iy

NHm

jj v Öm GN Üm

ý ýÜm

y

awn Üm

NÖ Ö

ÖN

Y O

m m

a

l0 N

ý

6

O N

fA

0

y j

N9

oN

d

w ai

. -. N It LL

p

to w CD

. ýý

Fý ä y yý cq tE EES Si

.Q $ L

12 16

'O-c ö ._ 1O $ää 'ä

.E' ö. Sr

a °ct mý

cE °iö c5'ß NE

.ý

ýcý ý'.

13 Ä EE

= ýý'a

c N3 `"d

ý' EEE gEE r ÜvE inmm5E3° Pc gc

n o! ö

t D Ö

I

to .6 ü

ri N C

ý

W

O LL

LL

Ü-

0

M

CV

aý a aý 0

cu x

wä

00 J F'

O° tf

N 4.1

d ýýl p

Laý

U ,.! E

GQ ºC4i:

.mf. . 10

m-. 1

R4 ü, ii I i24 dföi s

< QZýý [

«

Y °"F EY ° o

d a a

g{ 6

IF

pfldd! H

äipt 0 rf Sf

id

-

ý: u ä ° S i : .: a ¢ ni

icl! Y¬ tr 2{pt! ' ý'!

i ei eF °! ýýý : ?ýý! d! ý 4th. j! ýt`" ̀ t ¬ ¬R dd ° ! ý

°j ni°

Y t" iid

1 !!;. i¬bc

f! !i "F i° j, Y jl 'j F i ! ýý Et", { [ !; £i°F! g3;,: ýý: EY3" C tdi elF ý $! a ¬ ýý Y ý pf t3sCýiEi

., " ° ý4; $! ! .Y rý:. Y: E ° Y ! ! '

u 0

i o. ý "E! eE Gl t Yý! 'lTEE¬E ý °: iýFY ý$B yYe.: }G¬ ! '1 6j ¬ t ý ýý c : jcYYfl

E! ä

E 3-"}=ý t c Ei dd i. = ! 'FdT 'y' Et! iýi°. d Ylid tlj e ftf9! }¬lf :[ v` Y°Y i ä ý«

"Jt ý g

ý . j¬ f Yý ) =f ¬ aý[flý #ýcý ! dfn4ý °F'S °ýi dý

Yi ýuFa[

d ° ¬ ¬ ° [E ä c i! leb Y d3ýc ' iF d: iý ýi$ i j}[ ýý E

3bj6Y tFS 8. 2 ! -

ý Y P E! . Y° Eil giit

"E F Ez° 4_; vi

c c .6 i S j i ýtý }1. l8 F 4: "ýe ýý Y Edel $Yý Eý: E$i° t" RI j $ flY"sif! ' ' 'Y

ýý '

_ [ }q fl}y e ý8ý° } fl i:. ° ý

E-iý "

tü" ; le Y(b2 ! ý°Yg' ýýý ° " °

e 0t c uec ¬E'ýýE Fl3Yý i !

! ; $'t. E. _. Ys;: 8

ýPiEb , ý6 t ýu 3¬ !j rfl - fl ¬

. ýi d '. iEýS: =

st9 $ Eti E! °

! ýt ! !} týýY Sj ti p d !d Y$, ". T s$'. d[

3-i SY<2 biddi 6`iu =. tclii$'}I) S

ý' is E ° °° co ", i " df ! e Gi Yd$[E ° ý'

¬= p'e ! Gv= E °E! i !C d' " E E 1

jT !Yý 3° °r iY Y =EY dPd; o£3vEEtE Pici > Iflv. e! .. " F } ö Y '

:ed ! c! $£ý. Y E° .j Y° °qY! !d ; i( del t£ "tE ti $ FldlY6i!! dd u £i

i7 } EY'¬tY Y c! 'i}("ic d eýeE d $ ýFý . °[ {ii tE ti FE u Ef£!

F Hill )}°. i E ö : i. v ifl t! flE7ýý

:! E y °b ! -'iT E ;t 13

ýd Ett $ fl

°v E }E P 'Y } Sstd : °i c'Y' i u°

Eic°! a 'i : e£P!

ä ý $E7 YE 6' F

: -ct'it lÖ ý ýý"1}iýý St $. a i'; ý e

e. " e{jS ¬Ec YF

Tý ý S'"i$ E $i . "f

!. '"

! Pý 0 iait,. -e. F E!! Ye

e Evr F .E . C$t}"ý°"jv3f! ° Erii! cfl e4te ° '! G i$i. 'ev lif

cd i [ u

j tYg, YI 8 .Id

EY! " !de i" t- -Y ts'8 x- !} fl. l. ý" týd o.. $ . t dý piý$! ý tdt

v ! iE° Y

" ": =". ue"co ' Eoi a fl¬

i ''}it

_ t . _ . 'Ye a uS Yý"°Y¬ Eeo t t'! d' E-ý rv E-ý E It j't ntýY ='! ý! 4 ii cFE} c8' d

ý '

d 1[ ° fal e° d Ya!! 'E $. E. °8.

Y Y. t fl-i°) je )¬ olý u ¬t'ii"

o F ýj j ý} [ "

r _ ý}! E,

öýQo°° 'f °!: _. E -i. d Y ti; : fl(

iE° °i i' .i !ý E'

fl° cr °Y I$>$°[ iec e": I a{6 =jY'. i a in r. ei L! tfo e

g. Yd F'ý dý}e §I ý! {: i a8" i

$$: ¬ ta8 d ¬Y licY ¬t c Is}. e¬qcd t ,

ß =. i fl 5°". }Yv$$. i ý$ý dl ¬Y

{ YE

.°e . Ed v, F' ' 1

pq jý flýýE ¬E týfl 'S. {j: ¬ ý% .

ý, öö

iäi du7ý $. ; °e E fi Y{ -F d Fj fl

:. la . qj vSn .

¬tj "° fl

'; 'c. ý . e}} E av

E FDs -i

3'}} tDs uYi: =Biä: Ettflti # cca,

ii !{

}EnF eFý

6'{ýi ieE'e! pE !E S'` {

je 7 lE, t4 j¬e $ E! =

!p ]i==i. ! 3 E

p_ E"s Fi!? i! $a! !° YilýýýeE icl°

$ Y[ t j"E"" tt ýY

a. edu(fltj)Yjý $Y(d }IY{`. e!

ýi ý}! d ý t ' flý t E E u6 YE E t> ý iý i : ¬ýR

Efl, ý_ $e $!. fl }0 .1 fl 'E81)£'`d 'i E d Y ý

t E

Y ' ý! c 9° äY d;, ¬ ö '

$ a: di §

F! fld d, E tctld ! E!! 3 tý E i'': °in 7

(f! lýEtBE T EI¬fl cs ýi-; °G 8 E$ .ai. " +} tt d" ° ifl E cE!

'q ý c" }' 1 FYý ctSd ' ý "

= YI "°u , ! "a }, dFc ' tiii {n`"$Cidtldic! 'c YlEfl°F. 1°. )! ýd

. Y@! - S. e. ° °iE i7 }Y t! fY ýý

° ` [ Y 0

j c }. uEi y !7 !t c'E fl e d= y¬ [ E)

v. "tll utFi g F !6 fm 4 ° ° eFE i$ ei Y" °

Yeti ic °: is ":

iE ý ° .Y r . 1 gc !! ö ,c°.!

t F_ dac. i !i ucc°°«! $c }'

ýF°eb, ,(

1S£$e rt 4b $Ep y! }'rd$

Et - C "

} '

c $ '° E °° e . i .c d°. j Eo'attt. t$ "{, "'e ¬F ! up Clu i

cf°Ef CC "

{ $

[ , "F; EY¬ E°rE EPi 0;

ý stc lE °1-E ¬`Ei i>i b ¢ :., ° ! ,} i_Y"Si{E $ {} ¬ý. t

°' viY CYi lca. , Sc¬ c

_ '`' Y} bgj ý_

c

" g F! } i f[ii ! Y6 4f $=j! EE=FE 't:: ýtii

$a. tpc 6 ii_Eli F"Y}y ý

i'=! yi bý Y $pr i'¬ild:

ý: i (c 8Y ':. i. ý. l; i Yp Y! ý' -ý

ý ý$ ý ý

i ýFP { fl: a °

t°d}d :ý s}i e! a}u[, `ýt , }", 5 a"Y¬E'! pý ° /E" . -4(j . {t t°

ýý E-}0[

, F} }

ý

ý o 0 p

=c FE YF jeinp ýitýýyis ¬ci¢ -tile: 8 IEEdF¬F°v6E)lY. til !. flbiiE}d. E.. B"i

{; J., }

dn.: k iatjc

. Eu$ y iFS { 'i ý

HE t °-' ! dT "GE dd}= cr e'. ,F1 .c= ffl Y6 = 8! ¬iý d! £Zr.!. n°} e$0 Eiaci! iE id! $E. dlY..: SY

a . , ! b [}, 7 ýc

ý, dt; iý) tü°

°Y E0 y £Y})IY ci"E: o°l; u! i. i 3.0iii: 8eEýe¬E! Eiýg

o fä"

t= E

ý

!t F°

ilia F°

s! }n F°

S¬}g S"

SF}ý dFE °d

SC 'Ot U"

i :F di

a

:., f!! }

u

'F'e 10}i

üF. d }! dj , 1d

'4 Y ý e Me Me 13's

$ ý o °

"f}EYE

N F=i

ý

. tFdi

u u Y

=C ° : Ff $[. " d ej ý'E di

Bull ti_

,ý ,r

=8 G

t ;40

t fl

= ! Y dj aeii

°It. Fe a

r' Fi: i

öl, G "i $E o a ! E: }!

Y ý

;; Y! Pj % d! ý Eo: ¬SE Ei ýE? ý! fl[! 4

° if cý Ei=ýfE Y, p o!! E !$ EF$ý! öli; ýý

[ , di1 Y'°. Y

: Efyý} Gfc" Y

Ya :{j o i°1= eEiF"t

i " iY!! ="771 'ýiS, E°SC31

aYY¬ i

¬$jCet °

E, 3 Y'fl"(! fl i t Fdý . 9s ýýýfl Y

ý dc °¢ En ii- ý{

, dY a !E F°F

d t4 oEcBi!

Y d ý ýýl

rS l: niI

[ E

o

d E

u uF YE

i ä

_

!

YIF !

r ö c o: C oý; ý 7Y ¢ t .

tF

"

t1 13

' [

0 " c E ý is

` 11 e ,

ä

d-

N

N

a)

u 0

x v C cu

w

'C x

bo

4° 00 0 0

w rci 21 b

ss,

ý

N

p

N

o Ö

Ö

p

v lri

G. '

Q° rn

"Yv t`o C"C

3'y m ý-n ö

vm Cý

ý

°

m

v

GV 8j ý"

to rü ? j" s&-t ýü d 6ý E ý E

G

c° m

ý + äEa, e'n« ý e ý 1; E° gm $ ö

12

d °: °tWäö

.. y 3

-TL övcn t 'ü

cT '° y p- 13 d '. $p c$ , ill E

ä.

° $2wd 'ý

°`

m < c

E t'1 "°E ý6

E= E°

ý' c

w$° 'öES", 5. ¬a

ý «ý- E

"W ü

ö ý58, y

ut `o YE

ö E "d -R E

c° z S' ä'i

ýr r` c° n "- US

ö`

cýümýr°c ý" c ,ý v'

ý$ovu c"

Eýv - i

.nm N$ ýc `off %'

E ý

ä° ?ÄmWw

° ý"E $Eütn if- cvPES

' ' '

E -ii

.y ä

.ýE

_°rmr5r

Lr ,b=e Mor t' öc E°

ce3N5ý Egy ýi

Em r' ; 30 °'sömÄg

;cme

E60 31 öN ö

o"

o ö°m. Lý''E

' gö _. mn

N ýcýA _a ý. ý. ßý qE

35 mEEm E°$ S EE

9.6 _

d c 'c_ rn v ýa n 'ý rE

m ai °

g , öE fi ü

äöE? _ c° 'ý" " m¬° P °4 rP

= . '3

2 $L c °

t= ü f° m ö°

ü °' möEE5 3nEö 'n öQ

xi0 ä8S .., ¢mcn

ö o ý y Z ° Z

g

ýN

c ö

ZsäN ,6m Cä

d ü ý,,,

d Ö1 Ü

ÖI jü

Ö1

m ' öm m a°v 'ýv m

o? w m m ° o m

N

Y M

ý C`

.Cý v f0

LL

c" n ý'v lei U.

.T C

9C

q

E $

°ý ö rn 3 tiýý

'ý 5 "

CC E mäy"w

mac, f- ý ý+ ä

c Ný yaöEc

Q üTý ý 1O

öu .ný '

" °a ac °°'JQ

m ü ä> cöL Ö° c

ö 3o E° cý o cý Qt 5

.E to aEt? ä; ü

ö

c d

rs

E

ö

° A

.. y ä Y

K c °m v wvý .5

d= d d= a

10 I'D N

cý v 0

cý x ai u cz 0 w a. v 04 aU

0 c'7

b x

to

cz

t°

00 0 0

41

CQ

L3,

ö o

ö m

ö ö

v w äN d«

$ «EqO 8°Sox `«

'S 'ý`ý Sö

a w ä d

BE mo Zo E

w d

ßv$ä

°

ä wn rný drn

ES o¬ g Scgý` w`

s ýý r ö zz"

w d

I ýnÄg$ kýrE 8 ° v E"-cö: $nÄsxýücö°. -d ý. ý: ýý_ »ý

a® ` ý

¬'i, ýEýEö ý r ° +$, +pE$ýýw. ýS. c`°ýý. gwýpö

21 $ ° ý$p r - Ewý

ýo cg4uýeýýg'-ýýy E ý; e

Bý2rö E 75 ýýS:

C m Eq

2pg °5°ö " a

cr' g- ög

°

Ecä ü

e ä , E

r m : ¬vzö¬tm

°. äB

ä e+

5»'ß- ME 2_gSäýE ýö

an Ö °wä -rc k'ä ý

E Wc 6 tt"c ao

mý ý ?e

" x^c iö°, S=Eä° L-'ýý° Eö VZ $« " °S

ýo PI °ý: `°S G-

¬aý äýä r. ` t ....

. ýc `c ° ö" sö rc_ w

cuýmý ýdw cm5gm3

Ex ýBö ýrn<<

tows ° "a 3ý 1 ° >. ¬, E Ti r' i ¬

l 6F4 ß. F 5

»w '$° ö I-1.8. iä

-6 n3 ;

Z5

- >=oýýdc

cEäUI t ESt. -

xY S Sö-L°9i'. 2. -. b ýf S

töö°J i: $ä ZS°. c'o - ? rv c§Em

2 üc "

^e =°ö«4"söEö°

" E " : ÄÖ 8 ¬°ä:

n yE

° n ý ý

TEö!. 4

f t öccL ° ff

q$ ''

Amy g m= i= 1

yoýü tz . °. k

Ec o' n g p4 y ýü M'

4 q9Ä. »äpöe ca

c $g EL

ö

xE9 ü. ýý. 3

c< n

e nöÄx

e

m .5 C$

Z

I :

» .Gx$ä

g5

V O. 2« ýý

°eeö m«ÖÖ-°c o5 ýno_ 2

°möýqý oö öýc

c�ämý °mEy°"

öSSm`O g5 L°9 $ yä

" "p cöüce5` ýücý ErýE`" $ý^ E §c< ný ß ö2

» ääq

`g ü8

2E $ 2p i Öäß"`

°. , 'i. pE E e° °» P gY° -.

ýqw c ° °ýýä

ý z_ !1 9Ä Q$ O Sýns

äS

i « o5 ß o m 3y

x . oý«ö

e . " %

" ýEöE E _

ý ' ° : ýE Yö ý° n= : w mEä älä__ ___ w mEn ý i. ä o"

» ÜESISE' S ö8 ö3ý ýö.. "

ö

0 a

_

cö

. - ým

cý Ný N

coi

>CäN

cöö

ýý N 1; 8

4 _c z'ý üaqü yo' öü äo' däü äw deö y ä

- ý

Sü«md :: üd yy äud °

ü a ýý -°jW

äý9 mf/1 > äi9mN > ä> m0

C e8o xSo

0 0 ü 0

e o eý° m ems" n

JI N N fG fýl

ý 0

ýCF

n a ü + ýÖ a g

äý pw ýr

K, Fm 4 N Q

K LL

Na Qý p Clih S t

Q__

$ m $ E cm

s

ý E c d x. tQ `ä ca üm $üoye g ý+

ý jj ý`°x` öb ä

ö$ `h

üö a ýi'°' üvýyýäv

So w ü E E cxwOö gmý x x .S

ö 8öß ö . ý.

6 ¬ý

FEH üq

FEH `

ýý

d

E'a'ää°

öI E'E

E ö

mO "ß 'p Y

3-2 d E»$

q 2E

* OEIuý5- wES

C Nom E' bü.

K

m E F

S

ä o 9

° -

¬

Y

ü

°

ä x

C Y

E ý C E Ü

i

n N

Q

Co N

Q)

aý 0

x a) U

cý

112 i-4 U

U

N 0 U N_

x N

N

G)

,° 00 0 C N N_

41

U

cA

b v o.

Q

Ö Ö _ Ö Ö

2

ý

4

C

'. ý5$ E

'

C

Y

d

q

cL rnVi T"

Co

r

G

Bä6aöY Cý CvmijÄ :! m i0 cE ýv vý

ý - 2E 2 mä -$ u%ü

crý ö, 3 c°oE

3ton i iLCööE m m 3..

mö. n A ý

ýö$Wý. ý O1

ö-A

E vEi Ö.

E

C

LC-... Öý*OC

Y3m aý. _iý'O müt.,

ö

ö "ý °öu .H

ü r' r

°°, rn Ü«

an d ür

c; ö° cY ''w ° .y to ö

p o. "' bN ü{ ý C

ý 2cE! c om S q. Emýa

_ Sw

UILC

t E °r mm

öE ,

öÖ HE

3 o ý`o '^ v L' >c "-- c

m0

r3 °m öt ,°Eäd_ 2c93 iö '3

.

c i °ö -ý n- _oL ö g

m ýý ü ww

°

Y ü'ý ýý

ol

Y ¬. 'ma (7. mccm C C

5A EýW

1O

-N6tdC K}- m6

Ö

Eýö «

öEö dN

ý. 6ý

adN

w ý. E ao odN

ýc ao vN

cm N

Yý P

ýd S. 9 A

,m om

ö12 . ý3m DCa

ý8m m 1D ýCd

ömd. m3m

ýd

,

um aCa

w8.2 00 o öe

I -L o r ý

öe r

ei Ni vv v v

n r ý+

cv

w

K d'

n

K 6' m K

Cý ýtC

YNý ýn L C Ste. C ý$ ý q a

3a EcO ýE "

L

ö 9 _ mvm rE °' E YE

a c

.CÖE a

Y ý'Ö xcC

G-Q CEC

U °C C

GC

7`

'ý aö m Eo_ Ec4m

_` "=

H cc s

0

N

ru EE

Üvi

c

tom 3°

I

c

0

Üic Kö

° LL

p

Z

y

c

c

Y m

E

l°i

U Y

L.

m «i5

W

Ü Ü

rn N

0 z C) u 0

E 0

W

0 uu

10

N

CC

Sa

40, co C) C)

N

u

bÖ c0

cu au 04 a d

I

6j Y F.

Us"I

m ;ýg

m. F' y

dýýe

mill x! r Dfd r crrr

ark 1 alQl, 27<

6

y m c

1:, 4

ofE3¬

Ep

4Y f¬m n 00.

n .S., 7 , j{i. Elj ! 'i "j 0 1"( jf .

}°S! . F'E ,S ;jp FF Syj

ý !F }

tf }8 -"E9. { {(f O. S

jý i`i; e p j "

a! py ý Sc { .1

, ¢fýEý i° `

}{ )6J{°i (( §; ýy ýp(ji

ý 3()p! j: ý, {jtb t{d F °C s ý ý

ý; i ýzjSij Ej t i{ 'ý a` j} (ý ° °

d t ' F }" dji di j° r d.. y' d 1° dS(:. jýý }j iäd ý

C Ei

z ¢ Ice ý , +. i ( ciij 8 s(, mjt 3 'E pF° s4 F j'S: }'ý iYf `S4i q , I 3 E ! i [ j( 7! i¢ ýpj, ýij' 5ý,. {° , ij6j'PSI,

ýi , m4 rc'

" S: [Fa�! ýS 6 i'I ° }fý'mE8 8S f{, i /, i !3 ýEýý } ý. E t ' S !

Pk (E ' ('ýF°k { .1 ,l`; imö !! { ý

pS cEriýýjCr °� 3 ý3 j '°""; ; }2'c "j '"f , r . E. E'. °>[ e6pe j! {ug,

- Yicjici! 'i`tflý "S 3'''ý( E[ ' ( }

.S j F, tm} "(i j4{: f °ýi E3ztf° f+ dr ýýý':

F'yn"f¢il°iE yý {jS 'Li{'}d 1 {i'i a i; `

S {ýýt" jý

ýu(ý d '; ¬ "

ý , {r g

jI, S S i' IF ýi(i

,, [t{

Se": tt; ' t ýi , yi . Id. j"t}-e;

, ci, Ct}ý }°i u II : E". S 9y E j! ýy "' '

jj ;, j irp jt Cj: 3i jj, c" (9jj jj. !° tos ° QSzlý aE

ý j! j : ý}

$ ` j

fE 9i d pZm m II° F8 k'Si, f [, ýi! jr { E ý. } 7 1, ¢ i"°0°, °". ý

, ia J . djS jc. c

ý# g};

ii}i jS: ' j j{ j8

{j °"2°°: '. E,

ý iý }'

} j° 8i ('ý, e

ý`° { f u° ( ; (E Fi9 .' ; jt' i`ý Cj "F{`ýj '( S'ýlig i 1ý{öö ,;

} G (ý $¬ e

j, ý rr f(, ff j F ! `

'B . j t -. (

ý rpa ÖC

r{ i"F zt: i' S<jj"° aSjb 0 t}"¬c f'i4}

" 'i i!

} 1 1t m ji c. m tp e f., { 1 JFj d( e }1

°}i)8 1}`ýe. Ff {'!

3ý(`, { b¢ SSi J`tJ°r1ýC i"f, "

mpi, i jf , ji; j'r 3

d. lFutiJ 660 e( le C

(i j7yý (, {5 i Ciý+ C"+ { 6' ' l

! ! {( k 1: °! $! }ei pIiiý dE ý j ij tE,. ,

8} }Sf S } ? eS i! Fe"4' ; j °; F. jý itjýg ý. t6. c" it t ,. jS P ý

d , F °i ¢01 p" ! 0.! [ pj C'jj }c }E

ý; i ti ta j {}} j °(

ýtt`; i" }[

!! ý tt f it 'IS! ' Eju'i>i{" 4 {'Ej °j"Fajl l! °S:: ¢i "cl. i ji"yi 4'j't £ °u° ý

pk Y} E! S t '{, F, (i i{di (ý a' (j 9$F

!° j j ý { ) ° . j q'8I ýiy4ý"i}

p"ýýt}e(Fj ýd!

°rýýeo nýS, Teri ti i

ý 3 ý

. S: SS`jii-Y Epi öF

rE[{ {i j j, Q' S 3

C[ . c' F( !}° rE! fE. Eýip {# mF

} ii,! ;" ii d ' vt ,

ý Yi{ ' Ec 3S E{ 3, t. y p! a. , 6r'p1v4,. F3 '{ { °p ' sr

} si ,} 1, i1j[°! EE! "fr, jp yfiy, j1'i, " Jfý' i'''} ý7i (ýSýS {Ij! } Yki(mitý°

ýJ iy ri IqEo s{ "

ý { EE i r °t

"ks¬{r" p; {3ý ýlS, pfi S '

dm? " PS ,io. Vic( 'j, oe r , Si y 9FE. j! °£ý ¢,, °,! '° Y" fj Yt "

ýi . SBi, Fj ci j F(J! n ELs6 ce ° .S ý f '} E' pi 7S 1F o`6 P"s '° 4i3 E }°jý

;p fi; E}FýrrZic4 {°rvt }ji 7ý ý3"}dmj° {j

S3Öjpf` u.! dk ¬; 3p { -F { Et S

ýtif` 8 F 1° }ý

j iii ýý' F 6{! ', ; ýsiý i`" 2;: ýfýFi° ; c ," 'Yj: cä'ii}gYi°iij ;; ', ° Ei (8°{'4{j : '4E:

;! °" {}

i. i ! ý}j j} (t C {E

!; " , t j ! Fli: j? . 6. zji (y (;

[ }, Di qfa°°4, 'pS3 ;. E{+

, ýý(iýjýý} 3! ; Ijf'¢i ýiv, ý E;. "(°

ý ýf;

f ¢{( ým jý } 1 ; ! je " 1"

`d° °

j (, o6i ,ký

y ÖiE Eýpj "

, , F(a IS (j} j? Sejý. °t it!! di j" °j

}t{! {9{t °! ýl SS{8+ F dj

° t (, ý ý'

r' ° f' } ýt E}

zSf 'k{ ! i{dS,: e4} ji{i CS

6 ä'"! mj jkE }° E "' j

: Ö'j5 p !jPä}1 ü

iF S ir 9; (; .. ., t .";

" , °e

° 1 ,i i u jt " ! d: i td/' ¢ SÖ Eigq +li['[{, ( }"! ' } tjiii,

{° (S{" p ! cj jdP S"ý i Eý (ý' i t ýýij}'t §ý jýSs "

° (F q u. C( 4 Spi°f

- 4

- j C; S {{j °J ,

iF 3°. } ! ̂ eij {F p" L 6 r °EJY,. uj u 8

r[cý °i. iS j[isI,. , ý"'iýý `! t{ (¢ }C t i'i tt } }it

{ý dý g"n{ . J.

{ m! 3" , 8ä

cä !' 4} EEt! jl, i ý. j {} '° .

ý{ + ý(mii

Fýi ff 8 , ý° ý

j i i'. . {j iý

IcBij°8i i

l jISi !ý8 i}; ýt. S :ý

, y. 1 j jt plcm8d 'jd ' : ii

ei'ý}ct ; ({tt. ý:

1[ igit{: °}

S Fjöj 2j °''' tiL j " ý r , ! y t { , cl. y ýi

l i"ý : yu°)i" j: ' t°m' s t1 it kEi¬ýr t°. cS kj j}i, 7} 8"'}" ! ijj( °'FE'{. 4,.! j 6kß" Fiý ! t 6 i 1 ! i

( ý" °

"ý ý p gC a! ak. E .: 3 jE 0°E{6 s-h f

l s ° ije jFj ; F £r

46 . tty. Sý, ljt" }48 P ° {P .

L! 9 ''

. Ei j{ j fi"{ üjý i " °`d4 !tt! S¬s 8(" f ký}L

. E 8 'C` { : ! °. c iji tj ü "¢l n°¢"v°F 'e . Fi j¢j. { I! ii p ' }ýI o1(m! 4yifý j }ý{ I.. }i 'i4{y 'j_3S f t0}ý t { Jý ,

"S° sf z(sijE{: ý,

jE, ý° e.,..; Fo{ ' B a 'bjEmý, di?! SlýwmR '. .y j i}{ i 7 C { { =: }, °lat., f t?

Es rj. ii(t. tFä..

p F1 [ } ýý ¬ } {J ; ipii j6¢ )! 43. > . t° t r. { . j.. c: l: ¬(: ijýcdf"s7d, ýsjýý

6'. (s

ej ! fm c i° :

St0 °}

r t¢

'2

S

' i

t°

.Ö

+Fý{ 'ß';

e3e{

° 'P! °E d em

cES° lP! iq

d iý

3 'P! i ý äi

5 S S R t

ý, c

ý(ijs: r

; iF! Ffij F E °

y

ij }Y

14' iiF,

t Fm1( { Ssji j

. ý"F

"E iY tnr i F

?

diý{f{ y

c. z ,ý (4'

p; p : i: y

of ýF4ý ti

dy£jý v :(

`6'i°}f : 87' t

jEr ° F s£jsýýýýgýs

i :3 83 : II @ 38} B; }i: 83j.. il

' d u yF

E : e

ý }E

i ' F f

c

1 Sý ,q

. ýý }

IE

".

0 n N

N

0 z C)

1° C) C) Z Ln 0 u

'd x

bO

cii C) Ei

w 00 O O N I

x: 88 L{ v {ýt jfet{ ti ' f'

t: 3 tý { E; D { m "f t5 {t° d! : Fli if :° E{

! Eý : 3"g{ii ' i y

'{Dt a{"" äiD{ ` '

T: v {C

il{,

ý` {

=S if

{! iL ý t

ii Sý

¬(E '=lF

dI

Y Sf (! rr{i )

S(FS i¬S !! ES °{ a c Y

¬ {' i f'ý" !d e s1 "{ i ilE`E '

SF{: {[E %` 8t'E f--j a

.i iSY °i °"S., E p E.

c iý`sE{ý. E ! 2(, { f i

YE EiaS i m$vF 3ýFý`i EL-"ý

- °

S2 ýý1

s 4

.. i. ý

tf j' ý {i fii¬ E F3 { ¬yff; s

Sdf l' " C4e 3=

ýiýEd!

°¬%°" 1. E" fý,

(fE SýSESF {.

ý°

[ý t

ic! f %I _f} F Eý iEi .ý ° ,': ' ý "

. i" Sfi i°mi°. . 'E`ý r

-' ý

ý. %, S; EE6 ° t F°FEStFitýF; ° E f' =3

. '''` iC

ý° ̀ lE D! E°'%" i "cPd 3: '°oe i

j 3: l i ý" ° { T2- {

E¬ m=Evd °. g. E . t 'yiEiiY; Eýd F4 ` '6f'"'E{'

" 2{!! "YlllS"'i {CEý

s ý, E ýý °

° ýEE° i, e ýýi=p{: Fjd" i

k ! 3"r `{ S'S E T E [ "8 ýý

F , f

i d % E %. s3. . EeL'i aEE% ,i f

a S i t Efýro

. l: ES n d

c dý

6! ' ý'° z F'E 3E'° {{ i °

s..! { ý

F. S 3E i S f. ". a . f " L

and

_ 6 gm! ýT !F

'ý"%dDIE, t { Em i{ýc ,

ý"tEýS..

""=. "° " Y Eý"}! E¬ [j { E Ym. I! E, m.. l t

gt ýC'`°i :F

't -

Se{{ ed r 't' ßi2; ¬`% i; iý Y ( ý: 3

E jE% ,s °Tit$! f;, äý""E

S i °: im'82F

iýS ýi v. ä {e"' Sd 'e

"4ý!

. - " ,iý.. E E v 'ýý, "i. E F{ ýE ¬ % Bi: El it Sm di

mi, m emm 3 E'8 : SEsýi Sf" `j;;

.. S ! "a ý'

" m ` °i °SývE i'ýýif ý ýE { " "

¬3 {mv i, {lSfi j Eým;

! i'"ý (S i Sfm{E % ': ' T %: iýi E;

% °t % ["f , m{i. t [ip¬g

%f" ö' ; sFF to i: ft. i ý i}r e. s. 3iv { ii! 'EE. E E °! ° ` E E . ElFdiý:

fY E ý ti=E'iý

'

3Fýf: E: 'i Tmi. " 8a E4 ý

`d iýiEE' {. [. if3 qtý. ti" °

tE aEE° . mi 3": ý"m. fEEm

0ý if { f(, E% ýS' (

? Ei !% {t'm `

SSi' e t!

" {ýS4S ! ýt

ä, i: '{ _S Se scfs ,m

E { Li{ e SEE{!. vS

{¬ S r ý ý;

,t i%ao Er('E

°8E Sm E f` SiS -{°i {

Et iE%' _S

tý m F Li ", t 2{ ¢

i . 8. "

it a° t F'- { 'Fý"°v'aE {S° {%9 °

ETA i {i!! Fi öY"d ;` t'

, ý.

mi ! F t{

i S( SE SE9{ ZS{ I " i

EE "1S `

' %! %' gitý i#

dSy ii SdF , Eýii°, [ ' _ e. 'i "° . ;, ýýe"E4 ýoElNti

c l E ie eýýiFýýt'FF"{" d ß. _ ` Go D

° oý° iP d m

ýýý i"'{"t 1ý7 SE°. ý

g j iiE"SYýd, i ! L« 1 ° "

4 f' ý E ý2{%3° i '

f ýi id k %t°}.

v E tj{

I E ° .(! {'. r `E°, f. 8

d'

°f EF: f ! T. EE: E

Y = 1 dE i8: d E>iý - dEE E!

SP: ti 4°i : ' E`: I'%: S ! {... SE :1. { °E s" !

äB '! t e, 3m_ 4` iTv! {: '

"o E. F . E' .

f%,: E " , 'v= ifj (L

F ` E. s 1 E: {%E. 1''!

iii{ i ¬' " " tE{t'f. °i'

mf'F {"F fiLp f SttE " 8

" ¬;

d'E 't 9; if .

Fý. ° E ,E

i't! t{ t 2i! { ! a°(`E

; :! i : !d "ý3a SE. S o{°"!

'i lL m l ". .'EjE YF F

. E E DE. ' ! S" 1{' S! l iä l i¬E! f"e i" i, 8 E 73f Ee{, . Ef 6 j m3 4 5

¬E°"'iEi. ijYt E T is f" i týcS EcEa S=ym°%°t:; eE 3 ., % : E: E3 i'em! ý. 3 fm ý) f :S t4`: ib°ätS ý! [ S

"E {{liEE7"ý t"E ESS 'ý' ! 4i E ̀ { il S

''"! i "ý" ;. ¬teF: % 1 {e {!

, .. i Fi: ý 11

E° "ým SS iS!

5t p, S{i: E 'd°=v! 'StF' " [ {. 4f ýf

FeiE! ''6ý :° ° 4ýý

`afi p°m i f

lmf ii°. ' T{v ES` S it ý " " .i

SE. t°'{E'i 'Fi Ei "

° ¬ý 4ý Tým [ý FS 6

{ ý i

3 °{! 6{ E s; i ii m! f ?{t `f! S{: F

°fE! i! FE: d ": °

" ,j {! °m{S". fS{ t `i= `'Ei sF=6ý

Y{ ii tj'9 E. F Fid ü: im{: E f `ti SE. ESi !{

- :ý SEi

äfß

y! `t. ilvt E týEF " Fý >gE i{ä , ¬!. % %'"

F ý " " vc I" i fiE

.F aýL: E0 [

ýd E ! a`ic Tp 'Sj ii. [ S' i ý"Sý"{"E

Yýi"8 tf8

S Yt i` i Hbf , %SS ' `SS't"'! ä it i E E

! i'E tg@t ý' Ss'Eöý :

; 4E i al m {; Bf m $' .

SS 8q'i äL { . C F F' 8{E 8. = ß S %{Si" týi{ {F E. tj . (, fe 6, j t .. S .F

ii Eý{¬ i -ä Y: i { E: t m. iEYi! SP Sc'mF 46ýESY i% m-äg E EE Y{teY um-OH. iS lf. e{i3

i{{6 S2S {e "ý!

ry L° öö 'e Pý fi

S; Eö iS Eö äiW i; Emü E; E öm d; Emn

{

39 9

tl

Q

8 E M

se .ä

e

mm °E

ft

i$

iYisg§

4; üß ; ;ä d¬

ma SF. E,

X1p1yý4 I

9m S

iý ö8 E

"E! 6 II

S, F_$ý aii 3E° P1

tdE

a 0 3

It [ 4 a o 0

x

1

N n N

M n N

0 z U U

i. i 0

"2

U

O U

. cn

N'.

1° 00 0

41 W

U bÖ

ro

.

cs,

ö 0

m qmm

Ew ° . '`, a " ý° ý a'- 5 ý Cä äc ' ° ' o

r

ý "

°ýo ü5 uö tl

aý " L° w' ý°

- . ý. tl Eui ý «ý ý me C^ ä uw °^ý

, c. pE°: L' E ° mac ü ¬ßv2°cS °i üý wý ö

ýi ü_ i wüF`fi°, ä ý° ýEý _'r

is ý E2 °w Ecrti ý ý ý L^

Z. ä Div

än1! v_ $°, Ä mý

2EE! ý

Ncv3ýä i

2 CC i^a :

ý`Ceu1 r3 1 E `^

°m5 4sp ö ýE

.. ý -°ß.. Q --

äic ,ä

"-° ÜK ,ü vý䬰c°3ýý 42»

c°

E ý 6

.q ýýý ^ _c ýs$sgmýe ýý c

o£$, s'

ö5ä3°°cc _4ýc CSgög ü. 1n. a°ýc :

s öögs°ý u°Wäýý^ä ýz°°`3^=g A i äQg ägeöýWd^ö ü° y Ä: c»

c E ß° fi nnöE ä. 2:

üü 4` E°- i'[ Vc°m $_ e¬ Er h k

E c`

ü @

E V; JR

SW öV

i

° ~§2$ eLL

E`_ o9 S ii °V `- EEäÖ Q^ ö^ a? oä °Eööý. Y g" ¬ =¢ i e+

°^

Ö E c

E¬ °c 5 ý° ,, , , ý L ° S°ý

ä

S ý` C ' °ý Ec'E 2 2 YpC. ý "öSäcr

q°a°°ov°" cý'ýeýq

r EubÖNk 2^ ^

vi

S°2£YCO P Eöäüp

Z'- ö q¬ Ä

c Ly't° dýS ý'E °cäö tbL g`

ßy

4ýpýäer r-_ ^» y"c ýe

CuPuCflE°O: E

g r _

°tQýE E'O a .

r a r2ýin°¬a E_. 1-

ü E ¬. äýC

g

ö

äc r%"F"c

. Es Cý c5c 9D E. '?; sLL6--"

°c "6 `°

rv S

ä ß ý4

'g »8m s2 ä

8

ri

5a ýy

üE=WE

gý84a äýý8

ý'r 4ö

Et5g ý_. E 'ä ...

i- EöE 36 sV EEýýsoWO°EE n4 cý8. _ 72

a. g cüHBi°c2 q t

E r w cc $ 58 Ei

e ng `ma c°bf`i'ý

E

ü ¬

z ^

9

I, a Ü

ýt n N

n N

0 z v U cd E s..

ýO v

v 0 0 U _y

x

bo 0 v3 m v

1° 00 0 0 N

"ü v U bÖ

k 10 v a a Q

O' I -L Ö Ö tý

{h {) l7

N

6

S5

am0

C t0 4

N NN

N

' ý m

ý, m a . 3

'n E2� öU ra cC'

ä=gis ý, ssý 'äzgý

ö iý öE E° c$. E s E

12 Öeý

gEö ` fr ö äcds

Zý= 6ý t7 c '' ýýö 70

5 E° `m m E °ým' ¢üügm 2

Eo o°mýö

.CN

6ý N

°c50 üN

ö cm

NZmN

ß. 6ö dN

°. ýmo

d {Oy

acd S tp a m -°'i

ömn8m w'$m öm dwöm w8a ýn üm

.n$ ö SE ve

o v e v

W (n N

u ý v N

6 6 6

K da K Km N

19 ýE .. 'c SäN.

.L 'gym . c7a$

L ` aý u öy y

rn m L^ ä

Z .

5 ct

2N öEt ýý öüd ä

t'

EEE ÜäE A m°

c 3° Ü ö

ü O1

ý

öE ý'

ä

ýE

ö

ü

`w

ü

au

Z ý' K K

Äp

W U. {i

CJ - Ü

ýo n N

a) r"

Z

a) U rZ

"O 2 y

b0

ti tu

10. OD O O N

b r. iv a,

70

a S E

M d

D O Y

° . E4ý ý8s

o ö d d

¬a zi

° i=¬ a

m öf

n

Nmdýi`6 cý öq _

äm

m

°18 öö

° ýö öSc: " d n c a a°.: n i. nn dnn d Fg

! =b '0

fi9a

öý.!: S - -°fi" eC°'Cc 8u

Eo4

5 Fa oä ;

"g ý-' E -308

=" Et i., $E PC

'" LS

saeaS io° `iýoý a

ca3-czf! r- `ý` 4ý .ä ac" " ` F° o

{#, ge tt ö is 'gip 0i YF

Fi a. gýS" 4ý4$ }ý m °

N . i 3EI2

'ý C ßb'? '¬ c' 48 a' '¬E3SiEccuSE

°BiZ a' Fi =2 =

E o ' 7 $i`=o ". ° "` " a# ° p

N 5 >i Tb ö

S` 2 ! S$ 8'z:

are $ `3 a°eoaýcnE off'? "$. ES Bt. ý a

4 2E 9 rE E° ýiý E _F oodý

'go g

i8ý -"V-9 ý 48

o 4 SF 8" Y=-- $ö >,. r6. öf'i" $c

£ Sr"$EEd : VF tä S'

° ý ' ! c! '" ,

aý

-IN

!SP Li !c

1 2pq g i ýýt

E °c R" 8 `gm°n

y F `

ý BoF SEý s$cc"8ýii°-IT ( uý o

50 = $peýa e° ! $8_igoi. '-

E E fi° an

3 . l$S S" i": -"eci I-, " goi"FF, oý. 8"ý abccs ö Fo

ý E

iti fý n ý^

[ ' ý3 °ý- EF

ý8e ö ý" Fý

ý oý 8 F

=n. ''! «F V

! c we c o°c, .'n. _, p °c; { LSi a

t ö$ ,

=iiP,! ý ý EE9. e'c5 i8 .o -"Z idoi-: ýaQý re i -

ý o 12a o8 öae$ý£r-d'ýFý`

te f-bEöy. ` 8 ý, g8. üioE S°acýý, ý ýf:

S y", Kn

ý fi i ar. °: ýýýo

; b

<gý": >ag 3

C " ', n- 'It. 92ý`! t"E7

"-a: "cu$ .: -°

rL2 ü E {

Yeýl. E"E

a ýl. = f.. c_, sb ' e

ý " it, "Y

; $ß; EE. ý °aß.;

48CdE. ý " n g

y a

= C. S$ [ýy "fi ý

° , "" t. m

0 1S""ý°'E S"E« S

$$E ecr"""=-FS EýE. "Eö:

i^c q

E ' °^rE'

PY x='0'S: df ý1 " q E2! 8o p°i`m (

Sc°nt ii y

"

a ca . fi Efiý " o a>. E.

S3; c cco" a a. 23i +'-ioES sQc oe¬'fi-Y, {ö S $r t

-

iE SIT ftlrii 3!

ýtp Y,

ýfiýý F aGý

' E ý ý c G "

, 2FcI¬cS¬8 ,

E. '° "

ý c E o E.. m_ Fn-¬c eöýtjöööFme>8

. r= '

p ý p$!; 3-Ea u°. F°gli E"öa !e

s ý"g- '. ': eF E`F3toi"f

_ OC"P ý«_ ; a` h

`Et3 $ .ýc8: f S- a§iF.! '"rý` L

Si !F

4 : i$" }ý' 3 i

'LS`elri i °Lý. y

PrFILg ýýi F"1" t ' : gl-( "cýýe°s g

a 9 a"gtizC¬ý°$is�gýý dis i $ýt $

' C SS ^ýý3g='0f. =4[p; ýýl "ý

ý'Ö FaG, e

ý° EEC "

' glcu E

gals; g .ý

iP c 5- cF

,; °gs'-aa ý, l " "ý ¢ 's gC

.. 2"u s

y

.: $.. t.? iä g' ar " r b E eF a a! d =i'

a¬ ;,; "ýs=ýr: $` s fý fi -¬ 38 sY

aýý tSE i !

z ° ` mß bco-Ea ý: E- " ` ° ö

g sE .:; ýV r dý - $ý. P6ei' ý'

"i S<P - q e =tý, g M sia" asb ,-r ö°. a; `

ýýs b" ° E! b oR ý .. 3, ß

I'M Si S ° N

3 ä Yfi g E" E E§sY

'p@ wc. EiYÖ ý°

i e Z

ý , g , ý S

>liäß_ý EEYr taaäEßE: imFc{d

$ : ily :5

goi$dYa_, tg ý8ý - 5äcb$ ýy" c. vPBYety: F

ö" £¬[

=üru8ar "E". Sr"E 029

ß ýa aS "f'O" jEE ö, aE' ". 14la

$¬",: ýýfa: ý, 0M . BcoS, a:., äbi a

b }$ Oe$

" !! Fgg Yý6£gýyässg

8vmßi

°

°o m c

3 BN cm

gä n i@äg ä

=ä !n M .

N .; eo ; $S Y

- ALS U $a E; °m

B; :o 1121

: 3$m a E. 3am : 3am 3: $ H4. a

; s m m °!

,ö n $ $

E ý m m

m ui

m

e E: ý s `: `: rc@ ý ýZls ; o« uÄ LL @l8ýa ctBPa LLý ¬ 8ýý

,Y ý : äS

b

Cc

Sg

ýF rö "

3 e; E

d: y :c ýa ° 8 %ý m

s 0

ýc" : vd

ö¬ e gB., S_a - xr°`

P l Lý ; ý'iC¬ y . ¬ý

4S8: Ü" lmi2 8 PSS ýS". i "Lý: r

" F: a ö ci

ýo f $=

toýEöv cýEfi"i: _E1{' ¬zjF41E°8pfä E L ¬ ý

(

5=r e a' L ¬ö! E5

ý c{8F

.4 t B cEýü ý"E6ýE. E ö. ("aý8ýg bf

E£ F

8 -.

z: E $as gý

. a

ýr 3: i a ýR

e" . 8==, efi

o"Eý a öäc EL: ". a . 8Y; ýr$c

ý"üý"b ý8� ýý c =ý.

c yx. a v"

g S ¬6ý7 8dE g Y

Ua $` IS Ö: ifiaF y E eFý E 88! a absl: : ai°mi

ö °

N

7i ¢

p

8 c °"

ö1 t x

ö.? E

° m

¬ m -

öz E zE

ea Fý

e? i8

iýý ? iä¬ '

i

ö: gY

Eý 0 S F

ö 11 d` L J 4r E

Al

Z( - d

e 3 D

t 9

ä L

T 7

P

f

ýE I: _. I E Y

o8

L 43

U `fie

Nay=

ýýmö

mý oä

Q=ýf

N- n N

00 N N

pC

a)

cd

ti 10

x

cn cu

1-° OD 0 0 N

O

V

m

1o aý

d

. ems N

aýp

Nj týV N N_

2r N

t^ .

2N N

yr

.. rn i

m

l f '51L- mä rn'"ö«ýmc°, ER Tfn

.. -° a E2'Stmý

.- -- v=gm cö6

püE eö E3

rnoö- 8 3'Aöe¬ 9_

ü RN mS ° u $ý

c$ECcvE c ýöc . _o .

'd NfibC'°cm gL, °c.,

"c

om`' j $mc üö 2ýn, a

ö_3 °" v$

Ä S$ u

Ixv öq öýäa

° ` öE Y ý N ööýýc; °o L "

ärd ö ö'c a

, .. nm=c °"

` Z,

öd

. ý v

ý m o ý = mög d

oýo`O g ö°N

' °

A+ P! öQ cý nýöb ýýý6m ýýý'

°' ý ýL

'

'L =h Ö$tj c §g

cr'°v5ü°

ö c 'yýa3

6 d z ýgýt - ý5°

wm; Lc'EY'_``n _3 _ öüb S gc mc. ' 6 tiNä S

u C um. . _üýc 2 o"6 C`w aoE mYmE9 cp ý

E �. 'c öE cc ý

ßät L.

`ý` B c 2Eü 4m

,t ýý s

um ý° 5c ý°rn eaEýö ý wE o a ý'm

°c m ý° cqý ýo

rt °5 '6 .6°

mm uc ä 'ýnm ýäi. v cd. $31 me Emä m S'S ý

m W'3E p 6a-6 _b ä °u 61 Jm

b n mm Er, °$_n=

_üýPE ýE fr S:

21 ` `

-; S -b Lsým'9 cu am o. rn¬

g- r `

° ¢r m$ pa u`ýi

mo ö' ° ü ý

°m`$

p° 6E 5 Öcöý REýr ýöýý°ü °~ý

"'° > ri

r' S E

r1 'Lý 'v

"5 Y°ým °aýiw m$c. °id ->> ö

c öz

` " ý

yý n a e.

. °, ü«ü ýýö "°"0, ý"

E ö° ßc ß.

'o. °3ä °ýa'ýiü ü.. ° D ° ü °c ýn

ýEün, k w... «c Va Ec.. E C- Qý

° B

(7üüNý dcý N mý. = ýy u i

ä

cöýcýöi _cNÖÖ

v ý 3nc a

ye n $y äývc:. ý¬ý

` �ö' i .. . v öuc

E ö2 E 'c ýv

n m °L

6 ý e2. c mo_ýýý'CNg ýäw: r °c

E ý öR ýýö"`m1 p.

ýy Evb'S Z . E$ v -ýrnýcuEcý o

pu' °ý 3 E= ý. EoSÄ`ýE

- u. = f: ý'

c m m23° . x° öe

=c935ý öü cEc'`ýýä°n ý m?. EV U"ýC ý.. aN

1`üi°EüýýEÖSý. i. §ýc'v ý''iiqoüEo�r

ZLEüäýo m`$

g c "- ö ymo

cm ý" mö ýý äN ýwL

ma Z

-°c dN

Üm rnLým ý m öm

g ;! ü

ý3tm e

a, d m y3ým e

N9 LL W

e

N LL

Ö C /ý c cri 6

% LýLqü

_ý9Cýýä

m E m

cEe rn ä Eý°ý ü

= Y 1

a Nm

a_ Ü

d týj

ON L-1 N

C) co N

C)

u

u U

CSY

aj f3.

0 U h

x h r.

3.4 W

CC) O O N

-d

ö p

ö q P

m

Ö

'"

ýrv

mbmÄ $_ a ýa

m

Q ýý" g ö° W

Ö

°

C

C

C

rv lV

ý'1 4N

a

4 Y$ n CnÖ

Pef

p

Sý üm

öÖLCa.. °> Cx

CC

Eý

Äý°

C5

O ii wm vO Q

n ns

o '`woöý onücEýM .ý

ä« g9 : ö° yo cciý= '°mW nýö ö'ýý öä ü"- p om

p V vO gW@ä .

°, . c. ý äc

°'üLnEýä2= w5° mgm ý- ü

m ä$cc°ýv t ý'ýö'3c aý oEýr°Äirifc° c �v°ia cagý `

ö `Söý-¬ý "=EýNUý"`ýý-°°ce7-°-$ ý ýsöýý'ýý: üý8ýýý t

ü nvS'ýýz°" '°ýqvp° ' _ m«L$ ýmý, °e '- möd3f 'nýäWäK° n

E° 6 ö v ö =müm, m. $nýn-ý'cc ü ig umc ' -, r ' -v

` 7ý_a. -$ýmsäö cm ý' vm pc

cP E g u. ö b ä°

oy vý ccv`v ý ` ß -° c Yco ä äc ýE §ä-" i ý

n o ' L,

nSö- o tiý « 5 °m EcS ö rä e'; U g6t -. g.

o "A y ' n v o

v; üvüöFP c-.

- -§ ä. P

O

w

CQLIgmÖ

gig ýo^pä ü. ý uQ cyCye

Eý OEysänSt

ä ' , $ ö E ä ü

mr öýýa

r- ö o:

£ üäe dmr ü S ' äý mý oiq m

öä 9£ n v

P .. 2 ýügrvoü$cý 9

gw $Güü ° ö .. JWäö `ý$ýööäp 3püýda'3.

e3o-�ý ý ° äd ° c c

e 4 g l) w

Nq m

v müE E

mÖ 3 cc $ ücý Ka_mä ä iI -ää .izI

- P° `im`ý°miEv_°uEö E

ü5 n9. - ro

üöca c, nF ° x °

C$ü, t_ Mc EöX 2

.., 4 p

.6 m ! '°> ncmc3 .- $mES; E c °

fi5 " "d=häa mco m ý g SýuýSo

q ö ng $m =ü 9fý°äý? Fc LE öv-üýö pý-m? $

> r L D Öm

"

qS öE ýE qE

SGYm nYY °'E

q$ýg nn Oý °cm H

'E C

Kä $ m9 .

y Ö p"ä ö» Cö w N$ 71 LL

söR cos n5- 8mý

ü". _ m$ ý�'ö _gý

¬^ ° c$ý ° " ý

pN ü =ým äßa Eýý S: ö Äzm

" E $ g" °ä

Y Y$ $

.E . fi ALL- ° gsy" EE ° $mö g $fa eý^ä ya

b

-ýýT'i °Ö_ äv[ýE-? °=rýä$ýü=äöma. m =ýý >n $ Q$ ýý

-v °c2{äNý 0 E= : ojä ý$

, cb S'$t°%mEL'$ '^t

<a c'Eý_ _ °. Säm E, °. memgöE'ýE °c °c . Öc$r 3^_iaýnö6a ý Ö5w$E°$$ä198H85

Emä ýö co c cýö °c ýö

pýrv c0

üö°or a>cnmý

"m$r ýnývý möo$ cSrv

mmn ao }'E'ý+

sib gPW

a-smv : 30

m-

ý"ý, ým " n3vmv

Mm..

Dým ä3v"mN

Yö

sn- m8m

1mJ s`$ anü : 8m

ö ö e o

a eý ý(1 m

m ü ýN

$ ä 'E t4ä

N e g g

a y LL

Ne ar

p p_p B aU 5

4LLPO1)N

O

co cö `

ý

.4 's'

En äV tüm

C

rcm Sö ä

. mc-. ý

ý ý° Eý

'

p

n4ýnöCýoý w ä

ým E ö Lö ,ö 58mýQ L

ý= Q

°Y g0° c Ef tm ö ii 5'ä cc ' E Pý: Pcn

cn °n o°c $PCC°Ev Ö °

r p gmýýä. G$ R S EEq

wý 19 äý=

- pE C ý . Emc n"a ä- vg mý ¬nEýý

. ýy

sä`agü qa U2 EEn' fEý öm äE5oý5ccwE> ü

Nam N°eöýz

% Y o

u o

m p

« $

ä t

Y ä C

c n

Cm e

p .

p _

Ö U

p

4

Y

fc ' ' n

c IS n

E M

mm

m °ý

ý

ü

p

ö

00 N

N 00 N

aý a U

U U

tý

V x

w OD 0 0 N

co

b

ö Cwl ö

Ö p r r

m "+ n n

m

riý

¬g EI

8 "Süi c

Nw

ä- "c-

' oEcc , fr¬aiäg Fc

" c

w

6ýw 4 Yco c

n °q

p

S cW

d:

gS S

+g -> öF S

z` 2 ý> gö äg `ý ° EV Sý C S

t o°LE ° E

>Y ý-°°- " N 9 Y ý y tz.

05 ä mo g= C

mY wW Ä s" c C ý, _= 2E2 L}ý

ý ýQ Y p JB E Öä ü0 [Än Ö

33 ýý

ýc `gY` E° E üi 3s fa"

W _G K ý y _E

LCäL C9 ýM nib W.. C3ö a 3

O+O 7 ýE ýäONýcLýy cn gCÖ y Ti3 a

cä L

¬ �a gyög E Wo L c

+ äc t5 zC; en-89_8^ �°< ,ü : ýc

äW

'

'C 08«C =q c YSý5

4n

YereEC +" q r!

+

E B`s- ý "CY. + f°GC

q J/ Y

'! Et-H äýF öc ' =.

EE ý 'c öE: ö ö- f IL-awe=. `. S? E tý33W+L CYý ä 'nýP*, ý=°rv cýcPýö SgCsö

ä, o drc ü `z

>iEý'no 5"iEY"E öýtýi -1 V ` Ee ''" o°. ü

oc m`9 n_ý ýgö yy° %-y ü°ßäe

,ö cffý y CCýny S ý E°i öEf `üf s38

Y2ýJ §, möpSC öeöo° öw Gi g C3ý

ýw"icQýpEsc m a °

°z ý'E°°^ °iýE IIýs öTa° ^ °°, S ý£ýow s. e a: s ý

` U np f

o3W Wc a`5_ cy: Lýö

c_n3iYOCä 3 c Ct

B-". öýöý$^^YI-ü E T

N; q3 f$"

-TÖ ^rýcc F. ite_EdS äoLg `E .v°

: <0 Eg °u- $ 5t Y -°gý g 2 _

öS4 so ASäcß? ^ aP ýöpedüýs Cc

SE c tý

r 4 ýý �ý` ý Y5> n

ö ä ° m u

. L 3

ýo Si; - aE; g- ^ý ` Y°3°>0 ö

c5 " 11C241 C ' °

pý ` ux-E ýp ä'r ý

='` c B

ý- ^E E- Eß"ä" _ý Fa ýY

äw . Aoc 'i e"T,

' :: ` ýj8- ¬ö- ö 5> a

oro z^OCioöä cP ý

°> -a"ö5a5ü 2o `t` Eý"c Ls ä-OC

5 cY a eyöi §LE W'hý SoaSc o1-=E:

8ýüEý 6 a a

" E LY bE

ý E

i fröýmn oEc

: W' ýi n 8:. °. PE6 {

.$.. c- oyý

n fi Z 63ö>o rJ Ya ý« E ý ",

ooß E o_ öý$ö ý °cä b'Cý 'b

cöýnu oýE ßqä ` 3eiEýc

'g£E; üýi YtüEö°^ üýäý. 53 ü -"öd iýö gF'CoýE"E3 UEö °° N

f> " °a sE

_ W SSYTpi Z'Y $; ýýe ýäYCýE Hi

ý3WW ä "

ööÖ ýý : EY Y

ýÜYÜ5z_ a [p

E : 22 E.

91t1

" " 0'`ý`

b LWY YO_f rJ` WC Eý

c 'c 5

'E`e . YC äL P00°OýYyP

d i6c 3ö^$W: ýýw

o °

=öüy '2 ° Wof 3

s jy

t c c. e

ýw; °ý. S öýO"ýý öüc3 :. ý, °'

u¬ s° _ a ` n N N "y ° . .. w sý x ýr aý-

. 3W cEnp9ý_"o o 80 of äe C

ö ý9 CTS NÖgLYY

C a1O P ýi ýý ' 5ocýý_

I ü

^f 5 E; r gE HE HE

rc S o4oi L. 3Eoö s E U.

L ý s$ a 3rýc in -23 .ö qp s

s n üE: u ö° ü! SýP 5n -s: nc:

nwcEüäF_S6äi ,

fýI QönESeiC$iSie: s p p

ÜöSa 3

£mn o ao öýyfö

ýNL ýH ao °cyfo

ýN

ö `cö ". Ypry

`aö týiN

qcm eS

iitUQ 'iT

2 $a-na ü°

2YV« 2pSü sP

LL ö°ciw Qoac

om: 3m' aö 9

°üm 3'c° eO6 8OW ücö : 3m : 8m

ry m 0 Ö"

ý r Y M r

°1 m ö

m

n

u , ýi

N

1W

a m 0

ý LL

m

ä

N 13

U USW _ V

a: U

¬ý ý ^ yL

X06 aý

1 1 E 8

$Y ý3L

cr i 'ß .6 . ° _°

. 556

-F

xr,

oc

ccä ýSEmaä b'Eý f $tö c5

.2 °f3 s°

ý`= -Kü -""y=° I öý

ý ýýý ýCWf äks

Lf q

öoü ,

m" ö yEE C

Yc EEC

Ua°E ýn ° - EI; 9 1 QW ä a° r 1 . u

tý P

I

E ° cE $ S

. LL LL

L

0- 0

V Ü

M co N

d- 00 N

a) u

R+ z v C) u A

w

aý P4

00 C N 4-1 N

U U

.L U

tY7

v

Q

I 3 P 9

f 8 i t P

5 8

2

Jng YC

o mý E

Ue N .,

t.

WQö

C

Ö Se 8zi

x=`" CýQ- m"-e

E , <Z5(

= Y ä d 5 " Ecp

6ES d o

ne e

°c Nme_Sf

e¬ ä

Öf

ESýe :. °m :o ým ry

ý°ö5ý-°. n: c n .

nö nö

ti ý, ¬ d:

ýH °s: ¬ 's FtmsS inn inn

"t ¬i Fs<f ar Fec'° -$ "ýcc! $ °E ° ¢"

ým egB! e ! dý1, E IEeFmS! SSAFl rB 14 qiB v ` ` °

p EfH iE E

°dý iý0 $; m aF°n

pm^=f° ä "ü`. F¬' ý"ýr. "c Fs: E"iýýS -: °S riä

1011011 ý S fd 3 £°vc "tj L

IN FSzB" m st "l

Ye OR, n__ "tg"t!

'-r.,, °EuF 'z d ' ýS: u1ig ¬.

LS£ý E

i pä rv F ýlýv

m v B mý ý ; F. m°i ` .

iY°LEr d :d

c m ä

c F. cFe {F! ä. : '3. P Sf LS.

B ¬> c

iB X98 : =_i! m ýE

"

g ° "° f_d$$: ¬: "ätdd ad. 9g

.Q $ý: . 'ý7c't SH Ea $

°L ° EL ; LH :' FE

r f i . U `4ogE". E f°n Fa E gý l "5 E

ýiEm z E4c "s ý`

S'Etv`edm^f f 1 "n

p ..: S

=' `9du g : Eä

98

äs ýep

6E9Cm= : .. f FZE :

dýd£¬Ö=, i;; $ýi L ; °dd ¬ "

1 EE .- 4`5¬ gv°4 ; ¬naý$"y" ryF ýf: s:

' " s` <°

s d . . e.

t. E; " '°!

BFi : 3B°8; . "ju t B H ° "¬ý"e: ýEY 5"`: x 8. u ý9: SF ifr uR RI 11 is Mu R E` "

t Ig c:

Le° m; in'm

ö; $ "`ý

11101 E r

S; EdlF¬. e Ee tc"SErt gdrs°: s z" ý"

°m=d '"¬'sý¬"d¬ö m: ¬F F"'

i"tt I'v e_3° . B: IR F'c$

d" 11 "äs

v S4u 10 ý 4 1 C t

: cEdi: "Hýd s3'd,

moe"" u`v s` cc I ý". °

ää

mcd E mSII mE E: °mi°

" i

b " {_, . oFa "2ýat . L`"

"g¬HnF¬Em. 'vi . °ý°d`di [Dý{ t dý. i'O ý

°dä i ! : ý: i S ' ýý a"ý

".: 1

1 1, S<o: m-B ! =2

ý

LnZm- mH°p 0a ýýda . sF'+ ý i _ ö"¬ ;

a, 1 urýgEý ' t_: ¬{

"g g' 3g

a. e . :, ý: SE >

S a pd . d tv 8 dý d ¬dm. 3° L ( F°E' is if`i°t": `ö ¬b r{ý e. cý,.

¬ ° " '' ý

P dýEramem ° .u i$- ý i° mtrsnS äl:

Fdt Ii sdd

e i ° N

p e

d 'F ý" E'""i. :: L° °Iflf"

- £ 2

SF uZt

-a"öBD °d` ` 4 ¬g

ý ° E"D°: "° .L ". c.. b P =ts ¬ e'ý1 ö: ¬. 5ljc -2! -° I "F ' {

. Fm°, YL öm U D

E II g9 Ha c5 so ÖÖ- d": °$ "` ý E " c 0 o¬ -¬ Ffý

uS ! d'E. ' `: ý °

EE : E B` 9E 9- ü ° °

s"z_ ; ýL: L. -@gm - m e y F;.. _t j ec¬s'ýa°6. E fiýieS¬E gFtE°. ¬.

_ tt t".

FE " i eý cE8s H. ý. ! -B 4B 8

` orj yam "ý'f'. °-Ee ýWý ýý=`lB ýFg°ýiý

" ö. m F'i ° F( E ¬- "'=m

°d ý Eý "y W :: if "j$'ir" t! "

°ý mEý '

!m m ý Eý? e gi ° f° y: 3 di EdE ea

Yöiö°3t T'ýýFd -FdoE'f "mj 8' ° $ o n.. 2i ! c_ .3 ! o; -ýý g v9ýýf 6ý9 °'ýiý0ý ýFE yg ý ý S 4 " ieCi_: d4: c'E7ýn

_ dä c? w°ä"e iý ff F=<: F d dö¬ ' ' äi::. 9¬i. 2. _

E¬= _{

¬rf ; ý; `: ` ý B ýcFEý88, dýiE: 3ýý: ýap l F. ý. 0 Fv.

ö D

q _

c

Rý 4 4cäm 3Y; " t

c$" 3 b°

n äÄbß m1 1H-1 p=g

cÄfs S ýn

C(W i

O äD ä39m 5599 St 9m SY9m

2 ! s i O s s

r 0Z

mt=°E 8

°

m

E_ E EFt ä

°3m0' N N N -

ý

äýaF y Y°"c m U

fg °; iy -C

cm

A l

ýý Y -0!

e Ei

¬eýE 3""°

ý

y l

=

HUMAN! f SSc "

Zý° mm ý m fý9P . °: ö° fE

u y et nm ý

m deä! y4 3ý ¬E; $ l dýaýF°{g8 s

E " SYzä E £

Y_ R Emiý: ýa 5äE: i26 p ESü° i Hil. R4; öS b: r¬ ¬__ ¬ ö": l. ßdd. LL1

Ü ý = S 2 ý S2 II

pm E Öi ýE is i8 lüc

ý<

1

00 N

00 N

v cC

CZ.

cd

w

CL)

O v

. r.

bo

ce

4° OD 0 0

C1ý

N N

°

'yW Gr

<. -' S

° 'W

`GnZ` £ýs" 2O ý

r

bm

"

° ý° s ° °

mW

n

° y=9"

°w ° q«

c 9 8

8$$r8o& i> ýy ä

sac LL°g"Sn ý ýýs_Ea

ü ä c. SF ö'ö

°ýiEYC es :_ `

i: 5 tp c$ .

of¬£e £gýöFnýsL"`dý `gig Cfr tg S°ý S E = ý .m cý-s zz ö "-ýöö°ý! E ý4E _" ý

+_ - 0. q. p

:äs ° E' °c

q :t 6t °f". S. F "ý°ä° s

ý3ýö ° °

ý g g" '. öý

n8 $ýäme ;ý m'ý`ýE6 Ä°Y°- aSýY°°

°o° ocS<n`-_ef =EP ä £Jä i. ý°öý` T

°ýc. ys 'r

ý: aä ö °i " c öt ýL g7 B. dc°a-°rg s ýc=SEý ý

sg -co° f ýs

; öýý«$ 3°n-WEýy3°'O ý -o ý öO ýPI: ýF«ýFc9

ý _ ä

ccýre c ý_ExýQ SeE° °ü

1 °" 4i

¬s U&

!:! a F°, ýY

pSi

äR 2sý $^_. - .; ° K

s E ý e

ä¬ ^ s9 is y Se¬ 3 --a

° ° p v _

°-=¬äcm ýý$ -! @ma: 5: `o e R3 ;; e= _

si'=Yý °e ýEý°fý ene$"? ýf ä= °F W¬äßg< ýý

°

ý° : °mý

öE=- "ý =ý } Eg ýc 3£° ", rý eý 22

f 'y

E; --"-

; °ö ý: s . ýF . ý ! _ °

`S 2S °ý « afan öEg G

yzfEý ü:

C c ýEa y «d 'Fi b c4 °«'Ö

' 3' S ý $ ° cec 8 Lw

p o rý En

agd «ý Sxi tsJ ýerýiä gýö°ý E Q 4o E>ü _. -ý

c .

° ý nu ÖC {

ö zt . .° c` SZö `

s L . : i ä=ä p 'Ö5= !

gyýÄýom °. E NÖPEt eßä ü °" ye ý 3 ° . °cý dý d : to q ý9e FY : E°iF e s L" üu<psä Yý c$W; ý+ f" Y

c 6

ýg £: ý° S m °¬ °ä öüv öc2ýýv °gE3 °. s ý ý5ý ýý°$Y m c°c O° SC z °

`fir .:

` : '. a,

g 'e ýöö3 : ýEe Sö` :::: öxg¬ Eý ý58 s Q ým v .. ° .. ¬ 3° F -2 °c'r rvdm'` s %Bö. yý s

atz° °eLL: ýý . .. " ýQS°

°° q. ý

c YLý de S ýsZ pä:

"d äs"3S 'c ýý ö

ýEn°ögý t 9" e[ ýE`"`oE_$ý'ýý g¢ Fe«ý °

ý ON: Fo=ý'ý9beý `E"ýý8'n 5e .y¬

F sý ; a «P e E"

9e ä

° `ý ä3 : änSE - ° _ e fý_ ° c s2" $ c`

ö o5S° E ö°

"e

« il« ü. c ä mt`E Q -= ' "

ý ý

°

_ E u3 , , ým c ;, :2 "5 7ý =e8

=ü v9: nFE [X ý`F kB rF: E

Nq g ü ° <ää Fý& °ä" 3 ü SME ä11m

m=" F. A A° ä3o°1° m m ý m

X M

NP °ýn cý fý Y LL°. p3 CSGit LLF 0:

F

v E

g- rvg ° r«

u

°g °d e d y

g °

' ZEE S

r E. 1 E - u il

' °gaä, -

ý eo m$=ýý sei :Ep: 5

E a c ýc°ýL ¬ýE ßö4: E Öe ööEm`ýi

`5. ý pý

13E___ä

ö E J

E

e ö

rcE p°Y a

a cs

Cot U

W =

- G

NHS

n 00 N

a) U

C'. 1-. Z N h

z U a) u

CL)

b0

cC

42 00 O

ti

4J . L:

21

Ö

N Ö

m p

m

O p

m n ,°

ägß n. « d n

2. r S +ý ,Le

ihn` . b", i '£

Eý. "i .E . Ou 8 °° öYE°": ä

Lw "inn! " -; s

ý"ý+ ýý £ýSSýesc Ep

£ ý " H ß: 1r. =c¬

EL SS: wi«F

a E

1 g S

- ý

b fin: '

. " . "". ý ýwE

'"

ýýöca S" HS; Fý" 9 "

o !9 HIMUMcp PY

"c 1 } aä t"ý öeä_ e F

p° `pa ý£e!

ý . ý '8ý°ý>vieC. £$ ä" Sbýb

-' ý

j'" Lý :a

°ö" g öc ": b£F "

Iö ". c: i5 HIE°3lb

t fill

-` 'O o

t ýý°ý . wE: vi c Sýi

ý S " t. c"r2" E cý y°ao"a

ý"F S"S ui ug

° °" ins "C£ £Lgx °- EoH. H. f.

£i5

ý s ; ýS ä

£ Ff : i9 Lb b$ uk °

Sa'ca jýý v Lwuýr +c$a°t iý`L

fEoE L°-ý bs'ö "F't 5"£v 'ýat°sw

+H: H

ýS°ß5z S 'w£"Sýý

°i °F 4 ö

ý b

N° it 3_L cYS

ý i$ 3

i -ýEL °sý=: ¬ö $s°SýF_t6 ý, Y öý` ý '

i ý"$E'o°vý: b°ne:;: ý ý

° ýEfi : 'c ü: "0`ai Yta ý° ý

:

LýFt° 'ia ¬ tg

ýL srt ° } ¬F

- 22; öä_býýöä "ý: ° ýsaeý F

ect ý2 - it ýe o : aESý: °

T ýb' cý£i°°"öi rt. ýc a- ` "

ýä Löaö "°ný: ýý ",

[ ' ä

a ý. e 4'ö Z`a" o s Y £

ý, g F'°iEc ö-° eý ý

-Lo£_ iü. ý- epý

cw°e ý a .c °

ýý" " S°_" ' ý°

s "ae gs: "'

p t B pe ý

"a£cS °=¬f; °ws

°ýü °ýcoý

mL; °u

£öý ý: ü. ýiöö°'ý - ý .

£$w' .E ýaoY

t üai9 ý2 ` Y

fg 6 aýýY i

3 iý

z. Ems oý äY=ESLaE :2 :a`. " f £'Eet`. "i3" 6 L"

. F" ýB°Y E`- ý" e

öiSý"L ýo£ýe c

gf E -

C :. igJOýgý E o. ýF. ýý 3ýýi

. Fg _ Eý": ýý" Y E: 9g. ° 8g ý, ý S: sý _ ý «

.. Y äö T "F; F

YE s ý -°`b°c°ib ý"ýF.

, ý ":

SS_"E=`-LYE'S ýL

° °ý E""ý t p ý

yL% t t"ý' FE . "< - ° $c y "

¬E" i, rit ' ýÜ E

. ; fly t +ýýgEä io?

c Sýt y. e7°§. w -''

°Löf LS ýS" ý. ýý EA t

tE: '0ü«c: oä; Ös su: F `{ tl g- L

: ýe

z p ýGFb e:

ö £: S °yýg °Eý, "

f° ný: d` i a ag t`° z : ý lý

S ," c¬°: 8: $ Si ? _ _ S: g U. `ý

g w . zL S Y 6> g F. i° ý S i F`"" y gý ý

. f - "

£t_ lE: i e_Y- SSö O" £xe: iý S'LLýöSo59"ý: own3

vm E8

.

Föw quo«

Foi gýaýý

cro gý; $

°cF°o

flo Y FFö ggý 4

£` p 8'40"d z Z" E'40Si pö°°

g4 ? c 1 ä; ä03N e3ö0N

No "

ý 1N"

0 X

0 X

0 n n

X

m " aN

ý° ä p ä" °Fesý ý a

ýag F" F

ý ýý ý$ FF p9Rýp d : e4 y

s9 8 3ý

$ ý m t2 =

cl. °

#

w? 3 5E

5 ¬E ß ä%ä i

fsä : 'Fä FE. s . ýý5', 8r

ie`EýS gtý'ýd

E E ü 4

" ä '

Fp

f

c

S

E p

p p 3

f i C S

Ü

OS 1- mN Qj

p

ý S L

Co cc N

Q

0 00 N

a) U

N

ihr

z v

eo

zi u2

fýl

ýMM

CD CD

N

C. l

fA 25

10

Qi

4

ý ü ö ö 4 9 9

7 7

d d

e °; 9ö

d d d

9qß ý YgQ3 E

susýöa ý

NH-2 U

, ýý ääm

Z c'ä. ß $ z z z

äm° rN

ý

öss, ý ýädN

ýýýo dN

ý d

ýýý QýÖ

ýÖ

d

v }ý

Ö

d

S

">Y C

ýý

dpm 1

ý b' UmQ dý y4 ý Q OI

Z -di LIJ ö m ui $m . 39 md äi $m $m w$m mö ö ön n

Y Y

N ý N

off

fn N °D

D

ýn

cc C7

6

n

w LL"

6

N

cc CD

gý ö?

n .

d 6O u Y .

°p Ä $_ q

»$ dC

-ä3

'ö C? SöC 'q "E iS c' ö

.ü T ý

.... ug =Lý

ý"ý2 =, C$ Em

EEE

22 öEE U-n Kö

.0 21

n «

Cy Q

ö öE

I i

4 d 12

Ö

LL

LL LL

(J - d

0 rn cV

cu a) a)

0ö cC bo a)

a) u ce

a)

0 u

b s4

bo

ca a) E

w 00 0 0 N i+- y

u v

V

.d r. a a

i n C

E 7 ý 3 ' a 7 ry

EIIY 'EI

o m °

8

°S ti SE

7F

4s4s` Ilii E

9nFS bf

;E 7F

ä1 jý °

E iI b1F Fi ý '"' = °E öff :: d. ý68ý

" .E d:: d

° i{ S! 7 iý

!; °E" =e ýEIF

E 3 °"0 ' ° 1

4. i P St !t rSý: t' : ýýEjE6g iý ý" 8 ti! ' ' r ýi ° t

SY ,t9

(i '

E

ý ., , r;. ý

.; irvi

II. t, IE (iii r E S !, ý

cg If` ?; EE ttE, { '1e °°.;

ýY °ý ýýn

"y ý

rr Su. "! a =c y ! i ýý`iiygiir"{ý

Eiý !' E An , c, !b ýi:

3f YE" gi Ii: ° !ý

EE4fFit Ali p. 3ý3E cF-, E 'I ? 3i iýE °! "fY

"° i'

Y' f

ý° Y= ý Y "

ý Iu "

i iiý 1S

{. jlý! ýý iF'6 E gE gý EF Ep f p

ý v i

e m ý p

t"d"' "ii! ' FIEF== S "gý s

i EBg: ý3F iý u{"ýE :

ý! EiS : ýEi` rid{

Q ¬I S1 iý; ý SE ý c ým

i iCE 3ý EE i. .? Ft3.8 ý ý . g " FE_S'`°ý°EF9 ý

Ei EC3 is ýItr

" F'Fý f ý3°F.

_i ;! 5¢7

. 3'", Sii°E{ !n {Fý ý°iä'easý "I.: n"t m° Eýýi° 1° ö sa3 4le4 Ei ° en° ý ý ý 9F I u ' F ' -2 (1E F : ýý ( E

ý .i ti ( E:

ý', aE it

_ýIý 6'E f

ýý

{ sý '' {i' i ý F

't F ý

1 itliS ;: t3

"iy e

E ! Eiý E EI ýö i: ¬ ' l

`ii. 'i 3: i Ii. °m,

3E.. ". ̀ ýý 8ý i " i ý i ä

3 . . 2g ,

E ý a t Ep t ' ý,. i _ e"I¬ f ! dt! =ý°E ! °a Fý2335! ;3 Y3 f 9" sli

1"E"'i` t{tt. 4 E Y Ima ' II B i E ý i` ý ý °

ý

° E

;ý i rEpp "

!i Y{ °` Yi alý' ii i i. Y ý E E

/ 3" i E;: ý

: 3E ̀ "EFFt i>Eo

F . °i % ¢s e

¢ ;;! i! : =4<EEfcs. i; j

: iFFi iý !i Eiz°r ý' ° , °, mE= $. °r F EEý E jýý SiY E :6

> ý ; EE=: Setu ti ;l ý3' miF2i:

¬Y {ýE S E. "ýeiESFErýi ýY=! §ifiYgý°r3 Qý42-`. 4$ýýE i{ý °

ý: 5ý : " Y `

' ' '

tY ;ýý i I ii Eý y°, ý Iý

9 i <i I Ei ý.

. ogb e l1 Eý F

j yE ý !r s : rý ¬_ °ý; f E E: {-''u3;

°' g ý . E 6 E° iE tam. . tý ° 32°'; YI[ t ý_ Es"ýYi( ý-fr ma I mE=`i ¢ Y"' g "_t ". ii" "g °. iS e!

"! ö'it0 ! bý3'" rEF SSE

ýYi men °1 >a. mi° ° rFFS iug I F fE

ýi °m E;

iE 4Eg ý

{SýY i' 4! to

{ FYri! iS Y i 'ý i

° !

"fi; ýý

4 °ä

!3 S' E 3{_f Er!

E E

: pö ; :: ý; i E3 5E { °: ýEES : ýýE%e e! i=Y Y S" Sc e

p .,. , °F °m ;. om i= S P°Etä'1. '! bE (E''1 !4E {"i°q E esttsnu (f a' Irv 1 °eiEFic Eý-°p ýSýýe3

iliiE 3, ý E° E°Ea i1Y2fý°I. i4111ý

"'"E ýrý7, it ! i7 ö i° c"° ý

i EE Y -=

° 4i .! i y e Lý4 I pfSI'i Sp x 5%i ý: ý6ý'r Yi !

f1 E I i ý p ¢: E

'i ` s 'psi; ý; '38E

' ö Eri

, Y

=tS. Elj e ýEE=iý E{° E7

i' F _ ° " . , aýYý

" Ec SE iný. E. 6 IFF nri ý 9 !I i' E cri i ! E. ° "E FI I ; ýfEn°y, ",;, ýF.. ¬r"ý{",; p$; °EEE.:, "ý-. 5 . j, ýä E., Y=ýE ` 4 ° 0'

EE Y ! ý n Y iFFF gý` # YYýg "f. ý E. ° r

fa ;, m

i E ýýý "ý `! i Y4r j

v In E1' i°Se Y S ý ý ..

! 3E ¢9 ° 'ý iýIb, i ' 3` i E`EýS'

{°i ie i, diE2 "" °F ý 1: '

i "a° F ='= ! SErF°- I FEY{ °.

ýýji I'E'4 e3 irgiir°eýS. Pr, i !Et

1mF sit ý'.. E `iltýý"}ý'IEY

' ¢EEýir I°Ed,. i' r°ýE.; !- EiýF='6Y ¬u tFi EF ["- .iY '-" i °t ýe ?ý ý

ý{ I ¬ Eý i""dEEb St ie ° ei, }p 0 2Y S t' ý ý ;E ;i t

3 s °iE ""3 Y c

ý E EE

' Fiýe° { .i EE. .E S F"! `i Y"

t . tq EpLE . ttlýEIIi °fb Y'

S 2j "

i i°. if F° iey; Y{ c Si. fa v°ifis rvE=". I F {{. i. 'r iEE mliý I 8i' {

iir3apii:: i fýý F; IEi: Y ScE- 3 IEm

c'! °l2ý E °YSI E. {�=rF E jg E f ° E ý ! ' °E` t t ý

S. cri Yr iFf! ri iEg ý

EFIr 4

r 3 ý ý , E

i t p :S

:ý :ý , ci !. = °

; t! {]! Eý

IE iy tr3 7: i9 T¬i' ýýis`r ý'ýý ý

°EF: =ý ; EE

3 3. 'il ii iE lc r lSY .r : iEEý E - I Pi Y 6 P°ti! 6 !

ýsY iýýiýi4i ýý ýlj. pri'c°! 'i i iEtII>ý F ° t3i

eýEE'E °ýEY ̀ ýg i!

ý ` p ý

f'° " i ý '

yp6 E ý t ' ýýý! Il j't

. p°r S.,, Eý ii j E 1 a

YlY2g ,E. 3F

! 4i3 E E6 ý ! ¬{"E

, E! °Fý O": SJý ° F f- ?ý3.7 ý8 F° :i 8' ýj Ei .Y Ffi °°c

¢" i t . { ° ' ý ýY ij S ilý ( '°" f i'`, S ipýä¬ýi F®EeS, i. "IE`. =EYE

ý' Yd}f !° E° ý iS, Ip

ý' Eeic° .

ý""ii'¬E iYý EmlýIE. i pE, i°i° iic ýE ýý 'ý` 93 f i 'E E 1 r

iD8 pi! t i pr i7e" 1(°$ ýý Y7 ý F Y °ý

0 eýii E t Sis

t i 3ý: °:: E I`31i °EiF{¢°F ý42rc t°Y°i 9v S°Eu. em. mE4

{ t I tpI F 1t ,{ 3tE Etein" iýE ti t ýýt wi$SOEEEü ei efEC SE3 E! '. ý: Ei: i3:! vII . PF7cb 3Se. IIt .e

FE iY.. iEi [S¢ 13 j {f i :Y DýrEýEiElE. F SE. 2

ii E" :: ý .E 3ý

fe 3iYg

f°$ SýYN

f° 3iYý

'q >iýF S S

SFi PR

1ä

.c { 4F

' 'p%g ir'

Fý' 'Y ý . E{=

SY° ý 3t.,

"' 3Y! i 3 d

4 Y

°it ii ° :ä 33eE 3 Em : 3v i3Ei i 3 ä ä ý e ý D

D ö `

,x i_ i

1`

i iýFF

" 6 . 6

d ý fE

t iiýýs -i}i'

E

i'

ýx

! t E ll

' m

i p i

:1 ý 7& ii8 ö YE; {

SF e E' I; i Y

"; i jf ' Y>i' ýý Ed: m/ F Fa

° ýý"E ýE{ 'ißt i'EEE E =i ö {''S: 'YfE

" ý `ý st i

E ! SF Söl ifi° FS Ei9Y ? m 4'

Fi a

lI SEYbYiiE` ;i r[EEi{E E '

; ¬°r I

° E f aý Y!

=EHE ý Y iý vEi{ý$ i

m".!!! aEFIýä 28II E.

6ýfýý

ý ii

°III!

ýE=ýýYý Eri=.

ejfý a: äýs

°`IF

92Q'Eeq;? EýF

3ýFY E I=3 7 iE

.ý

ýýEEý

l ' E :E ý

ý

s

ý Hi! I

, 4 3 e . ¢ ý r

ýý ( u ä: t II°E ! ý ö i

`fl: n i. ¢1 7 3 E F

! ° g

Y et. s F

". r iE F ib iU ýý i

ö z e

lc s` 1 i { i11 -

1

i

m_ d ilf U; tl0 ng tl ý' ö

U Sý

öýFd d ag,; magi

m: d' 3

Y aý63 Qi; a

rn N

N O\ N

0 a) a) C7

cu bo a) a) v cd E

w

Csý

ti

10 x

b0

cs

w OD O O N zi

v a)

U

fY1ý,

10 aý

d d ö ö ä ö

ö ýi rv ö

riý n dý n

. F. ° e$nc °E° c°: ý E$ 3ýc ýir°we: s Eý$ö t ÖF . fi g

° ýl¢S: ieýf ßö5 _ `E ° "=ö °

öö �D u ö:. . ö:: ý°'ý5 : =''ý Qp

'f, I "

`t, 'e° -'

»Sýp "¬E j 91'

s -~ ö

"

`ü : i°: .

=ý ^_'EEF ý 1.

: ýi .- -- . E»5 ý- : VE ._ .c

ý EE "ö ZE Ei ý°'9 wiäýýöü c" : °"c C _ öýC "+q`ýS

p ý E °o Pý zq L p : oý EE2iä ý»S °w Yeý°S7

YiF ý °3gE. n

Z ` " scEE i ES

E y

'R^ ýi ¬

SEýýo cS a'iýn"¬e o `dN iýý os FLFcý. ýi : Y ö

£ =

LE ý'OS"ýý ° c `c LF :, F E

iP 'ö ' S

£ "E

ý

ý c6 LL

_ý P Söý LL ööiöä ýý°ýc°ý" -° iL

Eý°: ýý: E"E `. 5"ö : ea a

uýý°

°S""='°T=° ' 0=ý"'-öýý"t_°

q _° ýttE e ö? c °?

'°'

»Yt3 ný°° ^-°n cE°E ýE {

ü S fö brig

g E °ýös ý

S _ -OSýc'$ IcI-

E tc n y ¬

- 21 :F

E. ýsýý

1E

cýc öF¬i + i S

^5= Ei5! P ö äöý '° rdrý: :"

` ° r cc

ýE If Eý. C : T: äý: i oü! öGEEEEE

H °äm 5. E Emö arm em , £d

n ca

ýEm

9ý

ý, Fm m

ý, °

4ýÜýR

S . °. b

2öW S

2b

1>ýQ -

yýii

ý°NW SuG m

mm

b1 Fb

ýj g bim Wm a. ' ü ý` ECd ö

ný Vm nTi äO , 9ýW Ii ä .6 nýY vm vi ö8m h8m

eýC m Ö

ö Ö n 6

aF , Y1 YI N V)

° nS

° rý

"

r

=

r ay' g

44 örp

ýg _

ý m s

ý H"� sý 'i`s ¬$8ýä ELL

a_m8 I S.

iQ 2R. F- c0K r

y Iä m

ää °3 4ý tu., g@ Egd

-'gg9$göi 2 -eo a ens

@rä cß°g ° öä ö8 S`e ÄZrý°8 8ý $ý gi"

4 o ` 3gä L8

Es p

F 85

c 8t .8

` 4 8Ä

L E äöEfi

8'iä

ögg E, ýan

3s E "ý

8gß cE Ü

t'o`E Ebc EE �bE i ?. ý $

m ¬""_S $p

E cW

Ec E3

, "o2Ä ES

"e y ` ESE

cb2 ..

Ea`ücýdB ä2e3 g . E'

- t5 "o p s "g ýý"

Ee E"° am ä iE E E $ ý i

i iEä

Eä ý G856Su1E S ýnöýg°IS rc

3 i c

ö a

b

-

P

E

6

n Y «

T _ _

_ n a 3

E

" w ý4

` O

- E' M

�

s J p

Q N

HS s

g .ý .

K a i Q V

2

d d

M O\ N

d- rn N

cd V r_ a)

0

oýS

do

a) v

"O x

bo

0 0 N LJ

. ý4

'C

ö

ö d ö ö ý ° 4 4

o. d d a vt

d

Sg n E¬

IE - -2 ý

EH

-a i IV;

OýÖ Jý

g°'ý ýeE

ö Eä5

Ä us vö

Eý$ N

O ý^ Cl

ö °c5ö ovo

ýNýdN a

g"Eö ýd ty

ö ý- E-° $

@

oc G. dö

ý

cG

0ö y` N 2

Cc u, vý LLI

ýi « d«

ömäüm

dC

äm i üm

ü Oý dD

, 'm ýoSC E öm

Ö

'ý`d

dNd

öo .ýo

v i vE 0

CL in ee on

wüm o n

üm

n M vd v ý

d G.

ý p fý

N

1f

U'

m 'm

f: i N 0

~

LL ý..

U)

(J

Ný ö

d> W

h

C N

C`

ýy

ßd6

LÖ

výY 9ý

. di. eä8

N

Cd

ý

ý U

ý"L d

iý

dý 3 d Ö, Ndebd iý

LW EW ö r

C

1 E 2

m

'öö ' ccO1c a!

.9Ä 21

N ý 1$

ä5 .E

ü$ 1r. m

E ö

p oý ý

SEE iv ¬ t rn Ä Ä'S 6 s° ci .ý

1O dö

ý 6 ý

ä

YC

ö

S Ö

W

zd ý LL U ý

0- U

L rn N

w ... a b co b

v u cd E

w t a P4

I

(L)

21 Q) o+

P

2

s

2

J YE UE

4io U) an

. rc U"LLB Tj

0cE $

o. 8gg m Gov 8`

x-E v E

CL E aEöý Q 2ý6 K

.d N p

NU

Mä

ýPEb

gYý4 q

U 3eCÜt

b y N QC

Hä

3 ? p3dý

JÖpýýär

bN

1ýN

UN

pý

N

pý

AM

6ý

a zýý °. it I

i1 g

d "q YÖrO l q leafs d äý a

¬ . gymp 1ße5 6P w 1 g

SýüEu ýý, S'c ö : "`

'Yý ,E m`°e üý 'zn- ü: S e l

of§ äE n£« =ý

. . c^Br Etc$ ^

. v

Ei2aK 'cam c 6=- öb ' ö 9-

« 1 1 1 _ =cam

nöý 1

. sI ° . 1. 1 gy_m t a-

9 `,

¬ 10, 7'cy -II i 4d

5q p .

8¢ß, E lEýs

°

1; 11 1 1 1t ýýi'3

E

- , ý 1 ägü mýýý96mnÖC

öN °x D

j dHEýýstý

ý 'Fýi[Sö1 ýb-s$ il

6 "'Yi "«wi

. +. ¬ö

`¬CÄm +'. i iii. p

ý6düdd2`ýC^WG

ý ßw93 =ýÜ °ýd

F px$ b^ 3q m

g-^ y CL

" ""g^.

dz2

j

Q 5222 P . iý

N `c L: °cör . 'S C £ Eý°yS nSc°�. L'mäo^ll H y3 LR

°moüö=ö ý(ý6; If

ý `° 8a

^ ^Eý

ý c - ýwýw "w

E

d xýäý

ý, E

mö äýi; 4E Xö. _ p ` ß'3d r3

. °ý" 'c

° "sL

Pgu ;° ýn co; _ ým yýy t`qe' yý$a°ýmgý 9 ceý=ýF.: Yý X E

ssFý gö g¬ö"A-to'ýým`'ö ' g

S

o a<<zý; m'z°

5w+ - . äE g°r

, Eý == bý"'ý

aN ö Eöi8t `5 vfF

4mä°8'Sý"

EE¬L ý c WAIN. °E>ýýýa :

ö y@8ýa: ý ýE>

"

m c jj, dI-p=' öý " oÄ"cý3ýpmF

E°8 2 ý'ý«?

'°ä8s$

o

e$: F

O °fiFö c85 ° ucýn . ! '=. ý

p°ý3 pü ° ` `

ýýoiý`c s _ öi°

: u¬w ef; ýýoätýEsn mý W«_ ÄC ä°e äES ýsäaPý`ü$ý".

ömN

(7

ö

Eg

3

g . Fö i°cnN , Em

fill. fat

gý. Rö an E

ö

p

2ö 11 m aw om 1 a 131

r m m ö

"

N m

x

4G =

ý

rv

°

Z m E 7 '1

c _`e

Eýq- ý

ý' Jý o r pEý" ,

5n 8 pA^ý'

'ý T°ý C (fin ýý LL o g

$ai' a

.? äq Zd fd( C?

sp

. L.

L

ä` Ný n°ýý

d° YüÖi = .6mö

Sit E m. ööý °c ex d a ý'_r Wa

8°w ß ä

t 7ä E`nw U°"n Pcü3_ ýE'6EE E 3a!

p m E i p

5 ¬

o ü

E d ü °

ýö Ht ` 'E

QNr d

v QdQY

0 N

L-1 rn

w ... a 3.4 cc

b r. ci 41 a) u c3

w

a)

a)

O

N

'b x N

bO

in N

cu a) E w

00 O C N

. ýi

21 b

iý.

o p ý o Ö

4

N

Ö

h_N

hN

0�c

om 111pm8 H°

P2 Eäp== L l st

S

;

N

ms °ä r..

rw Iw°'cg°

"a

°°n 'c ° "; ö if

15.:

G

ä8 C9ä yuýi ý'- 58

o y, c, ° 41

a. ., 9w mgýR m"em $ i. ¬ 6

m° 3 is o

q E äeä - _: 5 EP

a? ýS°¬go

°o: pia-ý-° @ ööa 5 ý g cin

aie

.ýý;, äe

c � p E$ýýmd{ttEop Äuy5 0-'

5m n ö; ö-; e y °c

w' C öPÖm1m

a

E ceýoq Qc=Qöü c`ý °

`ES x üä nt E

`

p L

°

p Y

zý ü ö

ý° c

!

EE°ýý °eý Ecf"n ým

ö 10 Q ¢ OE

m . y c ` Ä

SUN nö ýý: x @" gö 6ýp2

m°

m ý 5 c_p2 ßi7}9 °2 E'oE8'o - üWß

ä ` ¬21 61E ý r3

M tEupE9

Fq "

ö . °ýc° w`-°5 j-2o°=`1 g

'

i'A WAS I $ ` Ä W:

" z° L

ä ý gEva i

m Efg - y

ý

.8

8N a.

m_ß m 110 aW AS

IA 1g` ý'` C&° äö a

Nz !_

. ýgS yt8

24 ; =_ý xýmsýW gä .

c=Eý° _Q . ý ¬ ynjg<$g Um e oi ; m"2o wsIä ö-a

öc ýN n'Qy

-o=` 'ý

" S". Ep: Rc

, 6.

EC 11IM . Eö°: °

-

ö aýyp gV

2 4 E66o cö »ALL»°ýý

m'S 2Tdc EcQf 5ý c "-: wwv

o c

d °ný ÜE

aä E mö

ß , R -"` c°ýO cmfä eo `I

° 2w: °

g"ý 3 Sai dm°Äm3 y¢

° °p5 'bb j luau sö_ OMMON LL

-Ä Ü`c °'§Sä R

ö ää°ß3m :

ö .1 ý

ö %i° 'O l C _ý ; T; g arvö Ep e :ý `m 2Yü "ý mýü

en _ USE"ä;

jH! jH 10; 4 j

N, so e

H

ins (E7n üE vf= ý°ýs"°nü £" H6nö ä' n

---ýwoýý ý'° ° EÄO

ai wýý ccgc5 _týö uV oß. 6 ü6

ýö ö

gc°m R1 1 8

0

is cö

t1 äR ill! öýäm y 3äm b. s -o ä.. ¢ý

" yy " 3

gýF m m m ° m m m3ym ý m m m m

m m

Eý2'ý E m n (7 7 m2ö? s

t 5 Ný

Rr "a

Sa; jcý Ný N °m6i Ö

ý= v e Kill 1 11 LL

at `Y%N no 72

°cE 93 amý° c

'. o* $g

gE i

8i

51.1 ' ̂ H IcI

-EyN2 f f 8 5

e ? E? C °

s 3öm S

1 ; `SýSöý ý C E2 pf y $ W Hm '8, `q ný

S ZZo. .2

I 2.

S ýü u= 24

ýä mo " o`pE Q

ý öä «8« oe 8ý =mN c¬ X ýE_ 1

°.

$g E E a 0

Z E i

vi S'fiý

n ý r ° mý

t"ý ýý ý q g u y

Ü -

ü d. -d

L- _d

z

Co rn N

Q

rn rn N

..., a b CO 10 zz

V

W

3.4 CL)

a

co 0 0 N

au . J.,

P4

tsý

IN q 0

ö

'ýN AWO

rv

h

d

MNN

a

n' i

e . :¬ ýzöa '; E<<. Edeýr¬ý ý =Eý äýý"$'n° ý ß w: % .

EiücrEä; ýEl ý =

ýe`'S°ÖC ä°Eýi e ° % ýF SY oä Sp

S 1 s: ý'

{` 6 !'i s. s aýý( j

" eLL : 5: qý

S ° . c_

e°9ý

6 ýp gn eý ýPe w

" ý $V °°n

°FQÖ naöFýä° £° °° " :L $

EA g nE.

e ü j Biý$ý

£x CSi

¬ýE ý EE ýý ýý8 L ýý ä

'c c

n g. Oe ` ÖOf iý t piE E ý ý Q wý L

S ö , t E c gßä Eäa; s`a; CýE@ °

: it s "

! E

c8ý; 4yý8sýs" ¬ P S ° &

Zs`4ä: äi{g°QýBý 8ý p s .' E¬x: d-$3 b

3 : "ýýsi

ö 3 3$: e °Q

4 :;

rü - 2 g ýä ° L p

2 vý

ý ý $ :oü $a R }.

nS Ö ps

Y` espý ZiS ° ; ti 3

ý gBüE .1 tfl

ý ä ä2:

= v

His- ° pC F .+ ,wo c

o °

ý ýý { o °a

. ý

4S{

Ez$ýs°E F3° ý[ et gl

°s E °a " -E Lä ß; ä! c¬rs

: p a yä¬

g .t ý °F ' ,e s

oý '

sý

'ý0 .

ýsfjlV-

s$

ü

o S Z,

' <Bög rÖPä gýt

tE

§

e ý

ý:. 'ýjE__ö to

o !

r ä a6_"3 'eö$

=E _`q $'ö n°"sý! ný¬ a -r eE`+öFo'>ä

ökni f

$ ie

öý sR ! gE ý

2 S eEEý'E ü: m gýolý: $ ö. °rvö äSC ý Ö; ýýs g$

$ýý. , ¬

C g e t Eo e°soö gýöV: SS a ý

ö n eý ýe

2a ýQEä± ýco gqY{e"P F

EýQý so6 iFf-Y

öEf w°$i'ö i 15 _

ES äoRE

Emö fö cd cd ýFö

m 2 ýýrv

GO ämbmri Fm_i

^= v 6 ücj°° Sý! ýp

ý'öm s

Sew äiSmN Igum Mu ; 8m

R ö ö c

m °ýs ' w c

äEä

s p 3ý °SE4'

po ° °ý "3fiýPýS

` S 58 3ý Ets°BS 8m 8, ýB ÖB: ög%miýB

ESv E°pm 5

E .

öE$ t EE g, jjE8° j M

m2ý 29 R1

2 U c

2E k'ý0 ° 35

°°

ä

5s$

Ä

fEF cE { ý°

0E . F$Si¬wE$

S E

ö 3 g s F

3 Ü 4

ä m " o i

m

K i

C) 0 M

C M

w

4a

C) U cC

a)

a)

0

U

.: i N

C)

E W

Co CD

O

cQ

b

cý,

4

p p C ä v ° 4

v ° °

d

N 6 6 S

E LL

N ä

232 Am

Cý

Q S# 2Nö - -

5' ^

mär -,

t F

9m

s°ýv. ýs s cgs"ý

¢Ö 2'w Cr Lam'

4,1 +j 7E

0

1= IüsäE z z° 2 Z

°c 6°o pN

d

E°m ö QrN

C

m . Sö

NdN cm

dN öö

pým `N

cm a

ö cö dN

gý " 1.2

m u' m tc wcm m v-m-, w _ ^m�8m «8m öm aN 8m 8m n 8m

ZR ý9 v v v

gym. ,N v N

U N

qý °

9 ný ý rý

d

h

d

1ý

d

qý ö K d'° d' R's K "+ C7 Ci A U' LL s I. "J

C 2 2 E. 9 U

2 ,w üaE -c -Z:

gZ ° c; G

°ýý'

%Z -- I

ýmc3ý cog .` e

o. ý He ýmvm

e E%

Lc3 $m

¬'ý. öESLE ` oSý °cýs'A ý"g ;,

c. S- aß. 2

'Ei ? '8 2 dö

pdý. ¬ý5ü ö 'Ec g

. ä$ s.. $ Ea

ýtý. ä« ta4 t cm 4 '32'ä

ßäc 9'-in

E °, !

.. 39-. A

19= EEE

5ýc

c `9 ° 5 NÄ Eý cv LL mNE° ý .

°t c U .kc uni K

.Q

N

CC r

C

y

A m

$ Eý+ o

cE Y

$' 8

S Ac

ü

S Z

S K w

U C

ILL

lL U.

Ü- Ü

N O M

. 125 O

Qý

U

cu 9.

U

U

O U

b0

ti c3 U

is

OD O O N

"C

" s ö ö ö 0 a"F P EE 0 0 0 ö

l

ZS

8 X34 fs3s

L N

i fl 18 Nmöf 54

ýY 8 os

c _c 4 'oiiö "n

nn t a

SS. io

JÖÖý¬i.

`nm

M1h C

ýp

1r

m. ýieo 6NNNN

ö.

pry

Sx . BFý°

YY Ey ° .

ýY Ft y$=F°t a ,, z tý8a, _, E3E °" e° t e.

°; = `

=t8 xi"2 ýa` 8 z° I `N

ä gýs g5 t

"Y ý Q

g ,; aj '

, 0

q3s`_=, S E2 e F: Sä ö iE

SE'ýe°'i- ` $° ' t " s 8; ¬ .s i. jaE; 6fla

s

= Yr ai 2 H 82F LEW °a'6f[ E _

ä :E E¬: p ý Q8ää-"t Y3 22 a> 2ä'f. £°sS M- -RU"

`" '

SE ýj 3F28

2ý3! d

iiL ý

FSýf: :: p s: cli iýýSiä

} 1 Y°:: E7 Eg7'°°°'! `rgf ýý-. ä'°: ý: Y i'ý22ýý ý°f °!: i`ý

' °

, " [: F" °

P. S ý "iýi "SÖ 3iß pi

ý"@.

ESEm °5S i r

: d'> s, ¬ýS _ :3 gp{t s`3ý; °= . c;:

ug: ýE .°8. B. "býaý . ý" "ý2"F bý 4 ý "@ °

; ' $$e'gýxy'S"" #° Y SSi '°ýf - `c8e2` "E cý

ýc . u ý

i° ý

4 {Ea'ze c` [a.

"t FeTxý" 8 , = S . ý" r äý

8! ä

¬ r-; E'-�"sr

iL F: ý2 " u eb e26;

S: °' g ýý°ES ý§"

pEE iF F¬g '

2 f ý $; °ýF Lga -°°ýEN

c c _ :6 Sä ný; °

C a sý. ýYE ; '¬gg . "i ` i

;. ý e " Sqc 5 s_ i "" %

f° " °c :. °". ¢Y8 ¬5 .,

° `°c 5 8 ` e, 8 ; tEý 25R Y

tý" N

, _. , S"Sa _ ?E 2mß°

c

, c_ = . c, L. c yE nSf .: . 83äi8

-

3 2 ý ä 8ý ý . _ tv fýý`Etg'

E: e:

ýF: ° 8°°¬E. iýF_ýl: j: i . r° yS ¬i ¬L ý = "

0 : ýx` g Ö=Eýgo

I Y iýe- Ec'S °2c4 c 1sp

R ° 'Si 5aýo

L tw 3 eöL ! E cE 6$ iPcYs¬Eä'; 'c °#°. ý. 5 t` I SS

°Y¬ 8uiS fýL¬: "° 8 "" rb S"" c o: H

PP;,; F B32 b ="°"}" E ä. E'S

ö aioöý: i iE ý.. ý : S° ed

: : ýi "= Ys 892$ ý2 E ý, m I

" Ftb ýx°ö 'i : ý; 5_mýb L! I SE t v3Y. ý2ý

x ° no fL

H 6ý +ým; S. Ftlr'ý ý &F¬_e 3

'`s äE'°`F ". by S2ý ý2. 'ý E

' b_`4 S t ä2 Itrýýý:

lýS "" _ Lß: '!, -eLtE ýý3°2 : °, " 1!

'Ü"L'- 3° StES L "3`°Z o

t8 °ý"i "° 2: j ý: ýS g . SFý cS 2 tý ` " ö

o Ew88'SEp"E

'üf ro

ýs iE-j pý"-. F- Et- "ý E £ ö`öct' i bp; tF

?Y ! t(£

FcE E e: csF +O cf

o 8 :F F ! S Fa gý g" ;; 'j;

4 ý ,8 .i LiEi? 8. e L 3ibf8

"8ä_öSS , ý o ý Ff oii. _..

ý.. öii `. c"! ý° _ E86_.. ES . M2 : '8E

ä p°mrv

L

yf

11 fý

':

sö Seen

5m E ng

cm Snag

jm ýýýg

SLFi

OR

fäi

ääw ä 1 ä_4'E

äFgiFS`gd

'P, 3

11.

npEö

3om

'ILS

: 1c fi g :a

m m ö n

;t m

ot"

"` iE

al : Ä r

w _

ýi'f g v°n s'" °EB Fgd

. - ý" «gis c Ö

ý' ,, g '

Ecs j Eö

ýa m. ý

¬, Fm( " C1ý S

°

=YES

BIZ:

rlg i ýS. ! "&

i_Ü:

i EE fe

C'-

Ený

FY s

f Vs4

SES p Eö! i EE , ä

i8 Ek g ci igS!

N S; ¬ _P ii i. =ý

bö ö PYs°

Yi 8x b$

ýýFSý°a[ý. Bi 3

;Y ääi : i + g E 'EF!

ý t ö

Ü 'S` ont, 3F HEMME 3EIS Lp y ööf!: iaCümi

E Q q ý gEF ý

4 E 5 z

L

" ä: i ý1 ;i E=

F c5s

ý wä t äf <F i¬ i8 id¬ ö s f

"did _

9

9

i Y

2

Jä

U. ö

ä.

sp

U E: g

s;.

c°ýs

g'. Qzzi

M O M

0 M

cC

V

0

a)

cz

7. r

v CZ4 a)

730 .N

44

W

"nd

co a) E

4M1

W C C N 4I

U CJ

U

Cp

a) a a

N

Ö H

0

lo

bNN N

lý ai 4N N

I- TNGqq°üu ni Ivö6 ü

ß

ö° CV =` ýa p

° `

ä

cD N A

2N

I? X! FÜoQý öy p p cj

E t' S 0

g '

4u ü rE

ü ö° ä$

° Vý t

. w Fi

GÖ!. °o$ ° fA $ COC

E ° 3'c ^ 3 ° Y ' aö$9CýeP

LL3ö

üp qý wýýo 'c ýýS .L

°ßc7 v ` y ý ý 1 .5

ö9 Sd c_

«m v ö � Q

w ° b örc

3 E

v 5 tim m

g_

ý 3ö O ei~` ä°'x T° '

mzo"°v u . r2 ýNßm

,°c 'ý °$ý üg9üm

Nt i° u id 'ý NaY3

' O ' °WcÖ i¬ Ü Eön ý° -GO

W=

1 2 ö G E `Ö °p ü q °

ö ü ý y ý i

ägým Möm gý' c"°'

. r 'ý qCc ö .ý 3

^. e9' 9 E6

.Y ,4m ý_ Q

ýüE Ec, ý. ýeýgEü Y $'}

y mEöGö°"Yöm5

ý ý uEä

m` rrüwr °Sum L' ü` :pCÄ

ýc d, °ý, ýb gcä'qým E

ý" " ggvý?

°x ö ö K rýäSä£mrgý°'E.. ý

6p ä-g I. `-v ög ý ¬ ý m 1IliI1U N

4ä o "E ,$ ät

Sp °C9 ýc_ m`-"sQýa E? ýý 9«3"' Eö°¬mg 5't'eý3 " öd5

g3 'Ü a'.

ä 8E5 'Ö ýs¬r3 T�ýcrÄ boo 'Ö sü

g ý

ö

N , .bö °är ä 10 1° 5Mg.

M ; öm �3öm m

e

tifý mE r="` äo

LL ý ° f° o

°ßä

E np p9c GN 8 G 8q

r3 JG E'ýuäE

Lü 5ü

ti 2 c9

3Gq cC3 p

GqÄ °

öh $

°Q OE fö fK '

_ ýE ý

.6ý

C E

° C J

E

d= d d =-

0 M

e 0 m

cd

0 U

cC

w

u P. aU s..

w

bO

cl U 5

W

co O O N

"U r.

04 GZ.

p °"

ý N

N N

n m

rr

söa°p: ä 'bý a

äq° gs

" x > °

-; _25 ¬¬ a 3'°-S ý

y:. E °"

iVry

si'P sE F; ý 5 " 5°

I P*ýYt` R s. k g s' ä

ý : s-f ` " ' ;

ä, !a osi :" 's, g ý

t öý ": @-E F ^

°s : e5 ®s°" 3

p Ei bs S

"c--ýý z_ :° `

. °ý^ E+s o ¬ ; i - °

¬Fo2 c$i:

" 2"5±"ý=';

S ^ '

E sq;

.. -. SSA "

¬ - -5Eý;: :- »ý'" äýn

a c.

E `u; . .c °5

e

is L =ors; ` ; -j

ba °

° Hi

- ' T s ý s y

"ý ""ü't ä

ý :: '"üc°f

n Y e

Owl f: ° öu f "v ^S:

ý_" i°ýa zý ' ýa Pic ` ý _°LL: `Pw t refi ¬"ý £

_ ^

"`" ö: !_ ' tit =¬Kc 'F °: ""

ý ýLE_göc « l ' fi It ö°

c ö: ; rö rs i ,. gö i ¬ 5F'#-z o£ ýS E

c _ý ga"xä ±

ýc ; äze°'

g¬ ý.:. "" ° ö¬öa° z °SS; ýý'ý c a E i.

; ßy2° F x `S

a , ä2 `s ýö. ý: s 'E a i; ° ; Sd»¬« Y°

I tit°

ý ¬ LEý": =

2 ¬ Ems

.i ý° " ¬5 Eýt" ý ý ýoý: '

°_sYý E c, S°E S5 ' ý X03 a

a E¬£3F nS ý "9 ý

"äY : öý° oÖ öý Ez 3ý üS Sy. °-¬°eYaý £ý

E -ý t>. ¬ äY . 8ýT"B F " ° 2 =ep ý Sý "

o

c ` ¬ F ¬'S `VC ° ä. ý° v. `'ý &i "

5ä iý i ä

" =u

¬ `

=° Li L$ s;:

5`

F 'G

`

$i 2 P

s'e

y S: ' c. ý25ý0 i

PýS! ý Ff

g¬

_ý at o; ö£. F! _"<< ° rxý<g_ ß < _ gý ; FF¬ös; ;ý ý¬ $. ý¬ a

yPö üL5°Ö:

c°w9Y --

! °ýY: "ýý . 'ý. c e¬ýe LeE¬£:: ý¬üSL F!

=�n: °öý3wSQ

5¬: äSa°ta¬P

-E. re

ni: EÜ

: s".. [[ý "! ýg

-ü5£n°ö: ýC: äSL

Fg4s; ýg STEe

k

ýCLfiStöK"'°ý ý3CýLa ÜGýý"Gc

W" PCäNý zýagý ýýý q

.ä S 9üri ý yew y ý3omh s 3omw ý

n; ämN

0 ö

f ä

5 s

° 3" 3 2

_ °

. ýG

ö L

ýw°, m Eai "E f Ri

R ý£ä

9 fEä GäC

iý. ý'

ý ý n

Lpn

t' ý

C s

tp try

g f

%

N C M

00 0 M

b 0 a) U

40. U

GJ

N 0

U) N

Ici

N

b0

N c C)

1-i w

00 O O N

N_

U . i; U

cý

'o aý a a Q

4

ö ö q ä ä ä 4 q 4

Ell 1; gý;; {S

J ý ýoY8ýýýo

,

tof

agS £ý =

aý c

a CL

° ¬a ! i9! ýBEj¬`i@' Fý ; °=. ý ay5? °Srý " ä 3; e£@ ii:: ý

: ge hr azae: $f ! ä: ý: F"

Egp'4, ̀s`: ý i=Y ¬ " " ; sa° z' : ä: ý ' ' : '

r.: FS

c o >[$1ä°ýü: e" g F':: c ääý; ' PiöS` ° Cü Yc '-y ýyK

c . 56: ý£ý9 E':: "Y! iE:

a

: : F, g . 4c i fi

:" 8ý; (n8 (ý=E £ý

S. $¬ýLf

f°`¬. ý£L : ö F"g{: c{gti . 5; 3 [8"f°sýEF 3°eý. ý" iý°=ý 35 g °

c ; ýgý; F ö "Sý S$ý

' £cSI

" F - i°ý 5 -e8 E%r¬ýt n -$ 2. eo : g ; äff. Y! :. týý: 3: ¬`4 e"t'8ýýnf "=a

¬i F scaa ýýýY° :

EP"°-., e y_ e '$`e ý" "s

" 8f o

ý e o c"Fä

säiii°i ";. ciSe-$, 5 °E er. $n ¢ 3£i cz F$ ,3 EiFýýa nY! nä

! : y

LL S =F I". 5. F 8gg 5äe£` ý4° ` ° ý2bgo $ ' " 8

c a- Sät ! fiýö;; 8p §f°i ec°ýö8E 8P i: '"ýt: SýýýEý¬'p $ö" "ý

, L: i°i

ß`c8 2: Qýö ýýý

c<8 o

iFt! gP Y, F ßE: :

£Q ä'^e 5°ý¬i8'y. s°gF

``"

$ ,ý i"P"ýp `ý£ {ügf ' f; e n. Fya-fP° f°ý6c ! n¬ ýý a

4P: " S¬S"

": :ff °S

ý x c°":; if uý > e3 f i Y °e öý Sö S

" : ýe `

Yg M. f F S E ' [. ;° ;; cq° sIö: &; i

iýýi ! ýý °kY. nF°

ýt; ei o" , ": c- P

ýöYý_. = a5 "

ý'ýýii 2`i vex- +" `$` r@S s 3 9=ý pia ýýc°cý_°¬¢:: [" "ý"ý

E¬ in ,E ý s

" cB"

bnc. "8 _

St - c c° f-°-s. ¬''rE". Öa

i ' t f e 75 Ö-ý ". bt°oý c5 E +i cY 3¬

! ö S 'F°n Fx 2. > -E x ä , i c _; ý! " "F :z ý> tc o3 F

tEcýyýQýfi t 8s $t `ffff"2

¬; a "fý¬a `S lS c 't

.

° , f" £y Ö

` "Cg E$ $ f £3 ýYif o°'-$ 5£ S.

¬ EgFýý8g' >, ýý3 SPd F e g y y : FyE SýýP

¬Y raf ., oö ää Saýenu, >>t0, üCn> {{

_, ý[ 5_sua"", c_fi"e

ý. 2ü } Fe"5ý£° 2 =

I

eo

Sig g°eg g' IN gg

Is- g

aým 2' a

ISO 3*? °13'

e i s 4i M

ý mö d

ö OS 3oi

gX f ýF

P Pq

ESQ n $ ä $ a

ýý84r g `ý ö ö: n d LLý ö S.

ýL F "L

p .. $"öät S ý't ec8! ce

a -ä5 S üPp ýa a SEgo3 Y3"ß 8 c5c "4

X55 c

X

_ ý` eöggäi"g 'ba f

gt_"ýiý cgs cý°! 8 ° i v@

° E

n

o` ° '

ý. tz tia ifi

3 £= %ý

", ̀48¬EY gY ý, 3 s" 2n

Öä¬ a e-"

io ä ý; cc ö

psi ý

§ ýg g F

s $° "E+ +P

ä E ö

c Eä ;ß ö58i E f¬wE°

$g'öPE`ý m"Ycau z°

¬° uo

E ý: ýf `sT'

ý

i 4

; j $

2- 1t

s" 8 c t

p a

"y F g S° ¬

° > - a Ü Z2

Zp 2CC

W

S uU-- tl

`- e

ÜÜ

rn CY)

C) r-i M

u

N O u

C14

aJ :a O C4 O

71 42

b0

Crj U

00 O O N 4-1

DJ

Q, tmý

ö . $ ö p a

E d

m8¬ d c

¬m

Öm5

gill "ý no Wm'FöF r

S 'as

n

i¬ 4

ÖL

_c cg q ým m m C: : EB

JÖP_ýi. G E: n iý" ýýn ýf

rfC Y

.- °Eemg3 -

bg e $eY

°

E [E i

t° _

ä° ý't FgS 'fS E : °. t'ö! pivPElF S ''!

"8 °

mt .i tI $ °S

y ý$e ° °; ý'n i2` LL °iim Yh Sý L ^°

ö 0

°F_ . .-

E#_"

ýö.

tan. Y i? Ygý`#'öp °e

Ei.. 6`i^e" °a " .: Y6%ýfi ; s;

lý : E2 ý8 P: t tt

0Y ." fti62E :: fi. 1' Y

Ei. tt¬S jfi: Si"' = EEi Ez

ES . 7 ä° : cýiFE°eEoý. °P

43$F Öiyilt'{ES08#°A a¬ li

a# smb YyE :i ° ý ' "

ý " F

"=

ý4 : aFgmýný

3 ý- tit-08i ä: 5! Yä°""i

Paeg {, P° zý °8 it s" j ,

$ö]$ g o°5; E= iF Yi :s ý

" $

( - ý! `@ "1 °£"

4q ace5¬m bO"cFPý"b-ä

SýtlmEL. `0$"

E SEe d'j {f

ýEFi Si°i fý; S"äEE '

Ši:: i: i"L

EýSSSoFS °, ' E i E

¢ geia°. ay°o: Fjtie jý SS{S#"ötn, °ý; ý

iF°p! c I

ý

" iý3 S

: 4't; =

bjýäEi"a F1Sö_E iveL=ýs Bim?

°Ea ! {a°°

°: i3¬° '' i -'

"¬

° y.. f. "I ý{i Ee . " :

>S'SOjS: °3 : ýýi i8it' °

t [-kfö`'cm °]°"5 n"

ý. gfVy"°F 33 S

=T", Y P¬ f'°n] ýsý tEý fF "j EF S ¢a mýn ý " lýý ° B

u

¬'"ö BT:

¬ :, :t °oýcF i. m F

, n, 0 $"o Fna e °ý°ýSýpý8"c

F ý E_ý : S'¬ yi t.

-LSS ,$

Eif sp0ýf: `ý°"' sf; i r "S iEf. P""S - "8ý9p0F° ýi # " ° i

_ ý

4 a $. ý«F Sm. E Q '7` E. Er

Y"Se 0 EhRtl 4 a. ýi E°ejE Sei ild e i g B E `LýS ý !

c 0

'pä". E jöE

i. i ° ° '1 '

eo' 3°"°c6 r <; Y Ogg g$

°`g°-E n°- ° Y'

ý "° - Da Ei-'"ýs !° 8t''c ¬

i. : ýö3ý iS4ýý :: =

CF_"ilgiiE ic

n. = it ým x{ cE{. ` E :3 # ° ;°

c i . ý n .E tSnE .; . tFYr"

t`°pc4 l- i$ct-f°iY4n sA di= ý ¬-Y °

¬i ijns s" ° ý '

ý

°. j et e - .

°Et ° °eýS 5 E$"i9 ýe Fp sor` H

_ ! fm !8

S°ýy° : ' '

$ e " - c Y: E ¬ ES£'e .tt Y i sön e£aý

f i i Es iemE° o =EtS"{ý¬¬EýfF' SY HEN

ö °$tOm ý Jý j° !! 0 S- w -E S

! 4_] S SSi .iE T`c qm HI ,¬ m 02 :

i -E x. #{jE 3_; av' öSii b O i

-. Emi. iý" { SY Ft.. Y_ äl_ ,a m£8s

g £'f Y " F e 8E°m z {i ý¢¬ = ö°- + EYE" E] !"t7

aäj$sr4. yn'eji Ete t ,,

I

to 0 s

"! `o£ýE ° g8 öi "="ý£ji t __

st ýS ̀ f ° °

ý ° taE; - "5 ° , i=fi tg°ý E:

#ti'"i" S" '

" t E

=P°pe' S ER{ )tE _0°3 0ý8'!

ý$ ýC °une"#S

fa E ýFS

m ý m ffi{Ea°

' mp"ö[°s '`a

$'_; Y°"jc gd 'Ei

; , iYý: "i86= i" ¬i; °

: r Ei" 7E (# !E °ö te".

°°{ ti. `

F "'`E -p

p LLý j7' is lii 888 C°i r E PIN, Os: m° üif 4 bi (

"ci m # S: 8 102, t ý j S, nHiE "

yS _ pees j_*ýY mfg. da<i

a¢8- " n SQ 5 '. °df° ý

e¬ b: il 0`£ : SEi e d° `ýti °{E ° '

( f e ! ram j ý Ed Ys{$c

ý°ý cE $O° #gi {S! Ep' YPi ' " ° E t i ' Vie` . fu :E °°°°_=8t_`a ic. s_ hn"e

$ üeöiüS__äSS ý $ <'L: fE_eii S a°e ö- m ö ý S" t"9ý

_ öFao njb. ]. Sni. iFe iF:.. °ýS. E7" 7t n$ m¬ýý

- a1¬

0e q t

ca qý$g

nn flm Fn

oifl yls 1,

oil

o 1

i0, D< is 141

= '.

0 3M9 3.

P_ äSÖm

Yä e$90

9E4ä °y Na äSDm

-2 Y"fý- eS00

L m 0 Ö n ü

3 {

pm = o EY ý'

E ¬=ýP

EYEEE :

b

ä °

ä m

e

m

_ E ̀cE" _ °EP ýýý I'F

o l 7°n t 7n pS#ä tE °$g t

''

aýý

ý°

ý

öe

ýL a m! . Y_P

Et C m<

F n y E °

$! g ý

E 'S PEä ý£ m __

°niD y

ö t .m° ýFEe i

a cFP: e S 78

aýEýa: ý

E° ' c

e ` cE

EEýY

$f g

ü5ý8c ü $i a_ :E öý¬a8ý d "

B ýý : B: S°nmx

E E d¬ 3ä iE

7Fi

ý = _>

a 2

28

-

j 2

i? bs

"m `- Ö f `< Ei sF 8i ý

!1i1" ý

P

9

0 S

e ä e

2

Jg

oö m

ýn L3 U

tlltýp

o

Egg E

x 0"g! a;!; ae°

i0<

M

N r-I M

u

N a) u cz P.

a)

a)

0 u

x 4

bo r-

css a) E

w 00 O O N L ti

u a)

V

E

a a

o ö d ö

ýý n ýn

w V`o

ä

o

ö

n

6d...

ý$ßd2 ý° ;2 d120T., d. E"ýc

ýýCCz m" ---ý cN `idö8

dýN °o C7 Eý v rn ý°+

6 c na

Eý E d ýdddd

ý'"p 3'

ei°? '` ? '° `w aY8B

E 8ý caý"Y. 6'm

ýnýr1 . YfrY dNýW

sd t 'ý'[ýS ýb'Cidd

u

ß ät

3 ü=.. dý c

ýÖm CýCý . ý ,

^ ý

.

o ý'd Eý§ ýh ý E ý n ö ý ° O . dr. 3 D° n . '0.5 U' O ö ° 1= 6 U' vNS° H' c

ö .Eö .ä° ..

Eö Y ßv

ä S -d 6 c a. Qy c

. aý. Hý23

m °e m e

Ný Vim= C äi

° v

a

v

wU 3

cd6Qp q

tL

N

lL

T

C B.

0E -c$ Ä

QdEER to I4 üö

O

cc t3

iü '9 -- - °' . Z3 - cc

, E, NYC E"2 dE

22-2 güv

E-- ec t°° -ö N

m` ig

g = .nE 8` a ?.

m g . 32 $F6c

ö

. --

f

o° ös m` üi

r.

.s°

A2 Z: E3 2w

in öE Sr

E ä E

r C° J

y. 5 CY

dg N

ir ý

? Y

ö"8 = ä v, ,:

ä= d a=

a

M

N u u co E

v

0 L)

bQ

C)

00 0 C N CA . r.

U V

. S; U

X

v G1.

'm m

ai ': ä. s '°: =ü

OF ° F. ¬. e°

g: FM t: Y

._ S 2° "g! Ei°ý. a

d vy ui' ci '6 SSi " ý

`

L ý"g 'e ä%k'{; ;

T; =- a.

Lýý# Q {: ¬E{Fý ýä F: ¬ : ° 0 E¬ Y

t Eý. Fý: ýi

ý ýt

Fi EC F g; Fgý

ý `^¬{? "ý e FF °

i

45 ý`ý

= i F Z" ° . ciý; g c ei-v. $

z$ äýS EEE ýä: 3ý: E F' dit ti"c:

¬ýSý "T. D$aäEäSLfý ý: s°

ýcv. S > ' "°

. v, ' ä U° tE Y i$ E°_

'r. a? F: F$c°a ii

ig S g s

ö:: _ °u '6 g : ý: es'. ""

3 Eý ?' ýYF'nII i

"z { j tF

ý f 3

S { ýa i<i- ý :: `i

£Eigi °! =t d: 3,

St °Da9 'ý r:

cOSL o6 : a[ aSE Et LcLd

Qd tE : $' S vc t ; i 'E {7g :p eýi¬

y"d "cetöýýr eg f°{ä" ý EC ° iýu"ii` ý

"

_ F ' i _ E

'E EE ý ý`° 'Y. "z y " 6 ý

. vc Z $ý c " { E dS "E a: 5¬£:

ýi Sr Sý:

a! "E t: S=ý fi¬ý; ': ýiý"E i ä '

o i$ä rs E. '3 ES" rSiO

y4: ß r"t°a° S te g

Ej af =aEe°F ¬

.y EF ý {, y t " F ýt !f

dh" ý ? , p£zm F "p vyý ;¬ Sei e f" D" gE, i$ä ö£'S E3z `Eä s$ 13

. " Fý¢t

r""dE"n- E: Y 8 E3'_ 1tf'6S, c ö . F n='S'{{ý

Fxi LA Y af L £ ý, Y{F°EES E

9i { cE l Y =

r ec " v. "ýEta aýSf ýýF S"a. vE8 2

FcSY. °Fi ""ý{i ý, 'Fi$${ý

eE

a YývI 13

c= .i `2E EYc¬.. "t:

g

Es i¬ '

. ýEi Yfi ¬¬

iä '`

: n""

tE° {t ¬ °

!3 S t

'ýSEÖ f ab$ E°ýýýö{ 2Yc 'EE¬ ro "

cEl ¬1: jE-`0F ' ` `

_ ¬

: sýY;

ö` ; °i ; "t t= '

: cT f

iiiý: E{, °E

4y 'b ýý: d" .g

E FFF

EE"°¬ :Fa nDUE ESa" g . ¬{d ä°, liýFt` yr aFi ät

°d v §.

i , ±¬ L-"E7; F; ý

$ gSu"y`°e °c! @F° e. SE t" z

S" aE iS3.5c ýe¬Eg

? =Eýi ä F%E Ei-'$FEE¬äi ¬ ^ ä t< t° i¬: $I `tYTä

' Ys! S se (i ¬ '1 ZC �ýj I i,

L F Eß ; ý

m { ; v uE _äc=YY

n i3 E- iöz{¬ijFS bE T-ti oy ga$ CEE E9 gs" °- {ý°Se{E ý4`ErýF£ßý'

" F1_ ." . - is : oc_ r eYFf; $'E-$

' 3: =s`. ' { {{e {

$ i £F F °

. t td` P¬ : {E-{oý

:` ä

i } °i 3 d{ý :t " °z ce- 9ýg a eý¬ ; {; i:: Eýlý=ä ¬irig `laýbý

c>i! Ep SEe rv ý : SY -: ý

ýEcýý3,5ýwe"

`E'z E" Pi y' ' ` $ '

E ä "aiE-S ash ' $ý,. ýti

cp ýiE i9 iii ö¬'`. Fe tE

f c¬ ý° Y i$E {: ¬SE° n"fY: EY ° ý

, { jY .Yy=! SE7 Y Fý°'3d it { { E

, t. "e 4: "°"{" im '

ý ° E9 t _ ,-

i eF - .5 EP_ä: " nä>. b: t' ý; ýýIE 'a

ýý SD c ¬n 4+p ä , ý. ýa°{E{" u

T "az" { Err F 83FE.

; nEc"$.. >fFy3 {ec' ýýO g iv ; 5ý¬E. _ý S Y " ° {F ý= Y g

$ '0F- i º-v

ý z{ýEý sý'y "ýL r cE=. °y ° ý{

e, {

;: ", :, E a¬ eE C¬ sE ýrý = +s 13 ° S

e 6_ a'ý°F::

E. EgF E{ E{`' t : Ft ß ES. EE..

{ ` iEFYýi } ¬tY F

� .p i .

: 0 yE''f9 °Et a8 ß 3v "ý

!

EI:;

!

s° ¢a [

ý 4 p Eiý ý y

E iYFe : ¬YSc°_F: SEý: X500: :ä Eä öE° ýd Y¬a S t

nEt"4i q EF {E

Eög ýý

F° icän:

Fö Scä ý:

F. tö Ei3n

EFpm

¬CN

Fm

¬8 Q Y( ELL

y4Ü äD S

-'° 3 oeý, p; 8°--"i

Eýý B.. 'gd

pý 6Eý gpi

oýw : $mvi :3 UN : ömri : Em : 8ý ö ö ö n

d ö

ö m

ä ZßF p

" y

a "s: Ei8ý3 [Oil F1

a° ý9 oq ýýi $tF ý

FxAlf

ý! 0

oý ýý fb

Ecy»sr pý S

2 i' Hv It E

! y"! F° EEF i I

3EE

H

H s ý2E EIE E? y:: ý i3

i i i i1

4i2 '¬f

"2ýII f __

i i

S

_

E .

3 E

E q

E ý i4

S ý " 5 °

s

L D q

E ý ! F

ý c e

¬5 ý

t L

" w ; i

E

ö- '- - mmm t s

Cl)

ds3.

U) co

v

U

ti

. 14

ö0

cs

00 O O N L

b

c O

O Ö Ö

N

Ö

r

C

1 G

°

6 p

« ö

t00 °ý

3 ý ý

Aa c- ä? A

3 Äm, '

Eon=° a t ý

'ý= Z '!

ä

ö- -g ý 13 öy d öö "»ý me K °Z

A ES

°

ts ö? ö _ cg o' -. R L. 2düS 22

ýST i S ö. m

.n ü BE a dd

ßxä S'°E3 ýEýö

w

-X- - E°ýö

vu 6ý

_ -

ý, ý. y yQ ¬Nh E'

A . °..

ýN 6p

N N ü Sý ösC

ü =ý ýý

Y d

N

__ ö ý öý"z, Ec

ýöý ýN Er$ si ýö8"gS d

ö Eý Et Is '9 .

Nc ° °

ad m am -lt E"ý A

. 3 .'

c

Y « C « ýy, ö-a'a lli

öm' üý w'

8U dý0 Sd

o d. ' ü0ö S d 'w 8E

Sc Q3

N -

6 d

N

6

N

ä

't T ým

öün d `m d .C

mU

C

a cE 13

EE aas .Q "ý '" W

.c5 aRi Tt

ýn ö 9? am gE YE" 76 d

c cös 'O$`fir`. _ m

"ý. 5' Ti

o2 E

5a.

... Ec

Cd .a . ý5 ý'c°"-

.Ö cc nIS

än tEmý

h. ` ¬ c.

va EEE

v 4 äEE

cs

Cý of Qý

rn m5 3' c Uc K

E g L

ö4 S

0 oE ü

öZ Y K

s

Y W Si

E O

LL LL 1

CJ - Ü

10 M

Appendix C: Scoring documents for measuring risk disclosure performance

Pre Crisis Check-List Summary: Criteria: A. 1.1. and A. I. 2.

Accounting requirements - exact specification: A. I. 1. Investment risk shall be measured on the basis of the recognised risk-monitoring methods used by the group. (GRS 5-20 par. 29)

Depending on the type of risk involved, one of the following methods is generally appropriate to measure investment risk: - determination of the anticipated loss using a pre-defined level of confidence and probability based models, such as stress tests to take account of crisis scenarios: the use of back testing techniques is recommended in this context, - asset/liability management techniques - scenario of sensitivity analysis (GRS 5-20 par. 30)

A. 1.2. The disclosures made on the market risk of investments shall, as a minimum, contain the following information, if this has not already been quantified by the risk monitoring methods used by the group. (GRS 5-20 par. 31)

- the effect of a 20% fall in the fair value of shares and other non-fixed interest securities - the effect of a1% shift in the interest curve (upwards or downwards) on the market

value of fixed interest securities and bonds

Results of the analysis: # Name Scenarios

interest and market value of equity variation

Effect on Embedded Value, Value of in- force business, Value at Risk or Earnings at risk quantified

Effect on equity quantified

Effect on earnings quantified

Result

1 Old Mutual Yes Yes No No 5 2 Standard Life Yes Yes Yes Yes 5 3 Legal &

General Yes Yes No No 5

4 CNP Assurances

Yes Yes Yes Yes 5

5 Munich Re Yes Yes No Yes 5 6 ING Yes Yes No No 5 7 Hannover Re Yes Yes Yes No 5 8 Generali Yes Yes No No 5 9 AXA Yes Yes No No 5 10 Aviva Yes Yes No No 5 11 Allianz Yes 12 AEGON Yes No Yes Yes 5 13 Zurich Yes Yes No No 5

Summary: All listed insurance companies quantify their investment risk using predominantly probability based models (European Embedded Value, Group Embedded Value, Value at Risk, Earnings at Risk or Value of in-force business) mentioned as accepted methods according to GRS 5-20 par. 29 and 30. Only one of the listed companies uses the alternative method named under GRS 5-20 par. 31 disclosing information on sensitivities that affect equity or earnings. A little less than half of the listed insurance companies disclose that information in addition to the predominantly used methods. All listed insurance companies fulfil the accounting requirements with a quantified investment risk disclosure in their financial statements as of December 31st, 2007.

317

Appendix C: Scoring documents for measuring risk disclosure performance

Pre Crisis Check-List Summary: Criteria: A. I. 3.

Accounting requirements - exact specification: A. I. 3.

An insurer shall disclose information that enables users of its financial statements to evaluate the nature and extent of risk arising from insurance contracts. To comply with this, an insurer shall disclose information about exposures to market risk arising from embedded derivatives contained in a host insurance contract if the insurer is not required to, and does not, measure the embedded derivatives at fair value. (IFRS 4 par. 38 and 39e)

A quantification of this criteria is not possible.

Results of the analysis:

# Name : Scenarios interest and market value of equity variation

Disclosure that entire combined contract is treated in the accounting category "financial assets at fair value through profit/loss"

Result

1 Old Mutual Yes Yes 3 2 Standard We Yes Yes 3 3 Legal & General Yes No 3 4 CNP

Assurances Yes Yes 3

5 Munich Re No No 3 6' ING Yes No 3 7 Hannover Re Yes Yes 3 8 Generali Yes Yes 3 9 AXA Yes Yes 3 10 Aviva Yes No 3 11 Allianz Yes No 3 12 AEGON Yes No 3 13 Zurich Yes No 3

Summary: All listed insurance companies disclose that embedded derivatives are separated and measured at fair value. The majority of listed insurance companies does not disclose that certain embedded derivatives are not separable so that - in order to measure them at fair value - the entire combined contract is measured at fair value by classifying it to the IAS 39 category "financial assets at fair value through profit/loss" All listed insurance companies fulfil the accounting requirements with a quantified investment risk disclosure in their financial statements as of December 31 St, 2007.

318

Appendix C: Scoring documents for measuring risk disclosure performance

Pre Crisis Check-List Summary: Criteria: A. 1.4.

Accounting requirements - exact specification: A. I. 4.

There are no accounting requirements about risk reducing policies. However this information is essential in a crisis situation for the investor. The knowledge how the company is preparing itself against the effects of the financial crisis allows investors to judge the company and how the management leads the company.

To name and describe risk reducing policies like use of derivatives, sale of certain financial assets or diversification of investments is required. A quantification of the effect of risk reducing policies is not possible, because not every policy con be accounted for as hedging under the strict hedging requirements of AS 39.

Results of the analysis: # Name : Use of

derivatives as a risk reducing policy

Name the kid of derivative and what kind of risk it is reducing

Sale of equity as a risk reducing policy

Diversification as a risk reducing policy

Result

1 Old Mutual Yes Yes No Yes 3 2 Standard Life Yes Yes No No 3 3 Legal &

General Yes Yes No No 3

4 CNP Assurances

Yes Yes Yes Yes 3

5 Munich Re Yes Yes No No 3 6 ING Yes Yes No Yes 3 7 Hannover Re Yes Yes No Yes 3 8 Generali Yes Yes No No 3 9 AXA Yes Yes No Yes 3 10 Aviva Yes Yes Yes Yes 3 11 Allianz Yes Yes No No 3 12 AEGON Yes Yes No No 3 13 Zurich Yes Yes No No 3

Summary: All listed insurance companies use derivatives as one policy to reduce market risk from interst bearing financial instruments and equity financial instruments. Additionally two companies sold equity financial instruments in order to reduce their market risk and nearly half of the listed insurance companies disclosed diversification as a policy to reduce their market risk. By naming and describing at least one risk reducing policy all listed insurance companies disclose specific qualitative information on this criteria in their financial statements as of December 31St, 2007.

319

Appendix C: Scoring documents for measuring risk disclosure performance

Pre Crisis Check-List Summary: Criteria: A. 11.1. credit risk

Accounting requirements - exact specification: A. I1.1. An insurer shall disclose information that enables users of its financial statements to evaluate the nature and extent of risk arising from insurance contracts. (IFRS 4 par. 38) An insurer shall disclose information about credit risk that IFRS 7 par. 31-42 would require if the insurance contracts were within the scope of IFRS 7. (IFRS 4 par. 39d) An entity shall disclose information that enables users of its financial statements evaluate the nature and extent of risk arising from financial instruments to which the entity is exposed at the reporting date. (IFRS 7 par. 31) The disclosures focus on the risk that arise from financial instruments and how they have been managed. (IFRS 7 par. 32)

Following information are required for each credit risk arising from financial instruments:

" Quantification of maximum exposure to credit risk at the reporting date. (IFRS 7 par. 34 and 36a)

" Information about the credit quality of financial assets (IFRS 7 par. 36c) " Analysis of age of financial assets that are past due as at the reporting date but not

impaired (IFRS 7 par. 37a)

Results of the analysis: # Name Quantification

of credit risk exposure by investments

Disclosure of rating information of investments

Disclosure of age of investments that are past due but not impaired

Disclosure of change in fair value attributable to change in credit risk

Result

1 Old Mutual Yes Yes Yes No 5 2 Standard Life Yes Yes No No 3 3 Legal &

General Yes Yes Yes No 5

4 CNP Assurances

Yes Yes No No 3

5 Munich Re Yes Yes No No 3 6 ING Yes Yes No No 3 7 Hannover Re Yes Yes No No 3 8 Generali Yes Yes No No 3 9 AXA Yes Yes No No 3 10 Aviva Yes Yes Yes Yes 5 11 Allianz Yes Yes No No 3 12 AEGON Yes Yes No No 3 13 Zurich Yes Yes No No 3

Summary: All listed insurance companies quantify their credit risk exposure mostly by disclosing it together with the ratings of their investments. The required disclosure of the aging of investments is disclosed by only three listed insurance companies and the non required disclosure of change in fair value attributable to changes in credit risk is provided by only one listed insurance company.

Quantification: Required information about the ageing of investments are not disclosed by all listed insurance

companies. Therefore quantitative information does not receive 5 point but only 3 points. If additionally the required information about ageing of investments is disclosed 5 points are granted.

320

Appendix C: Scoring documents for measuring risk disclosure performance

Pre Crisis Check-List Summary: Criteria: A. III. 1.

Accounting requirements - exact specification: A. III. 1.

An insurer shall disclose information that enables users of its financial statements to evaluate the nature and extent of risk arising from insurance contracts. To comply with this an insurer shall disclose information about exposures to liquidity risk that IFRS 7 par 31 - 42 would require if the insurance contracts were within the scope of IFRS 7 (IFRS 4 par. 39d). An entity shall disclose a description of how it manages the liquidity risk (IFRS 7 Par. 39b). An insurer need not to provide the maturity analysis required by IFRS 7 par. 39a if it discloses information about the estimated timing of the net cash outflows resulting from recognised insurance liabilities instead.

Results of the analysis:

# Name Disclosure about liquidity risk inherent and Asset liability matching performed

Quantification of liquidity risk Result

Timing of returns from assets

Timing of obligations from insurance contracts

I Old Mutual Yes No No 1 2 Standard Life Yes Yes No 5 3 Legal &

General Yes No No I

4 CNP Assurances

Yes Yes Yes 5

5 Munich Re Yes No Yes 3 6 ING Yes No No 1 7 Hannover Re Yes No No 1 8 Generali Yes No Yes 3 9 AXA Yes No No 1 10 Aviva Yes No Yes 3 11 Allianz Yes No No 1 12 1 AEGON Yes Yes Yes 5 13 Zurich Yes No No I

Summary: All listed insurance companies disclose that they actively manage their liquidity risk. As recommended in the IFRS 7 requirements only three listed insurance companies describe their liquidity management by quantifying the timing of return from their assets. Five listed insurance companies fulfil the requirements by disclosing the timing of projected payments from their insurance obligations.

Quantification: All information disclosed about the liquidity risk and how it is managed can be considered as general qualitative information (1 point). The legally required disclosure about the timing of obligations from insurance contracts can be considered a specific qualitative information (3 points) and the voluntary disclosure about the timing of returns from assets as a qualitative information (5 points).

321

Appendix C: Scoring documents for measuring risk disclosure performance

Pre Crisis Check-List Summary: Criteria: A. III. 2.

Accounting requirements - exact specification: A. I11.2.

There is no accounting regulation that requires a sensitivity analysis in regard to liquidity risks by listed insurance companies. However as liquidity is essential to insurance companies significant changes in the market value of investments, in the credit risk of investment issuers and changes in the projected cash flow due to the crisis should be commented by listed insurance companies. This can be done most convincingly for investors by quantifying effects and scenarios.

The cash outflow of insurance related factors (lapse or mortality) can change as a result that these factors differ from historic assumptions. The cash inflow of investments can differ from historic assumptions due to a significant loss in market value or the default of creditors. These effects of changing cash flows can be disclosed as sensitivities to earnings/equity or Embedded Values.

Results of the analysis:

# Name Cash outflow sensit ivity Cash inflow sensitivity Result Effect on earnings or equity disclosed

Effect on Embedded Value disclosed

Effect on earning or equity disclosed

Effect on Embedded Value disclosed

I Old Mutual Yes Yes No Yes 5 2 Standard Life No Yes No No 3 3 Legal &

General No Yes No No 3

4 CNP Assurances

No Yes No No 3

5 Munich Re No Yes No Yes 5 6 ING No No No No 1 7 Hannover Re No Yes No Yes 5 8 General! No Yes No No 3 9 AXA No No No No 1 10 Aviva Yes No No No 3 11 Allianz No No No No 1 12 AEGON Yes No Yes No 5 13 Zurich Yes No Yes No 5

Summary: Ten out of thirteen listed insurance companies disclose the effect that changes in historic insurance related assumptions have on either earnings/equity or the embedded value. Five out of thirteen listed insurance companies disclose the effect that changes in their investment related assumptions have on either earnings/equity or the embedded value.

Quantification: Since the disclosure is not required all listed insurance companies that do not provide information receive 1 point. All companies that disclose cash outflow sensitivities on either earning/equity or Embedded values receive about cash inflow or cash outflow

322

receive 3 Points. Listed insurance companies that disclose sensitivities on cash in- and cash outflows receive 5 points.

323

Appendix C: Scoring documents for measuring risk disclosure performance

Pre Crisis Check-List Summary: Criteria: B. I.

Accounting requirements - exact specification: B. I.

Following general accounting regulation require information about how financial instruments on the debit side of the balance sheet are measured. An entity shall disclose in the summary of significant accounting policies the measurement basis used in preparing the financial statements (IAS 1 par. 117a). It is important for an entity to inform users of the measurement basis used in the financial statements, because the basis on which an entity prepares the financial statements significantly affects users'analysis (IAS 1 par. 118).

The implementation guide of the general accounting standards on financial instruments lines the goal of measuring as follows: If a market for financial instruments is not active, an entity establishes fair value by using a valuation technique (IAS 39 AG 74). The objective of using a valuation technique is to establish what the transaction price would have been on the measurement date in an arm's length exchange motivated by normal business considerations. A valuation technique would be expected to arrive at a realistic estimate of the fair value if it reasonably reflects how the market could be expected to price the instrument and the inputs to the valuation technique reasonably represent market expectations and measures of the risk return factors inherent in the financial instrument (IAS 39 AG 75)

Results of the analysis:

# Name Categories of [AS Fair value determination Specification Quantification of 39 used and of these financial assets disclosure of their alternative measured at fair effect on the valuation value without balance sheet and technique active markets rofit/loss

Directly Indirectly measured by measured reference to by an active alternative market valuation

technique 1 Old Mutual Yes Yes Yes No No 3 2 Standard Life Yes Yes Yes No No 3 3 Legal & Yes Yes No No No 1

General 4 CNP Yes Yes Yes No No 3

Assurances Munich Re Yes Yes Yes Yes No 3

6 ING Yes Yes Yes No No 3 7 Hannover Re Yes Yes Yes No No 3 8 Generali Yes Yes Yes No No 3 9 AXA Yes Yes Yes Yes Yes 5

Aviva Yes Yes Yes No Yes 5 Allianz Yes Yes Yes Yes No 3

R

AEGON Yes Yes Yes Yes No 3 13 Zurich Yes Yes Yes No No 3

Summary: All listed insurance companies disclose the IAS 39 categories used and their effect on either earnings or equity. All companies disclose their directly measured references to an active

324

market but one listed company does not refer to alternative valuation techniques. Although using them only four listed insurance companies specify these valuation methods and only two quantify the financial instruments measured with these valuation techniques.

Quantification: Disclosure requirements about the IAS 39 categories and the fair value measurement on active markets can be considered as general information (1 point). Explanations about alternative valuation techniques or the specification of these techniques can be considered specific qualitative information (3 points). The information how many financial instruments are measured using these alternative valuation techniques is a quantitative information (5 points).

325

Appendix C: Scoring documents for measuring risk disclosure performance

Pre Crisis Check-List Summary: Criteria: B. II.

Accounting requirements - exact specification: B. II.

Following general accounting regulation require information about how financial instruments on the debit side of the balance sheet are measured. An entity shall disclose in the summary of significant accounting policies the measurement basis used in preparing the financial statements ([AS 1 par. 117a). It is important for an entity to inform users of the measurement basis used in the financial statements, because the basis on which an entity prepares the financial statements significantly affects users'analysis (IAS 1 par. 118).

Results of the analysis:

# Name Categories of IAS 39 used and disclosure of their effects on the balance sheet and profit/loss

Specific measuring techniques of fair values for financial liabilities

Disclosure about own credit risk considered in the valuation techniques of financial liabilities

Result

1 Old Mutual Yes Yes No 0 2 Standard Life No No No 0

-Y - Legal & General

No No No 0

4 CNP Assurances

No No No 0

5 Munich Re No No No 0 6 ING Yes Yes No 0 7 Hannover Re Yes Yes No 0 8 Generali Yes No No 0 9 AXA Yes No No 0 10 Aviva No No No 0

11 Allianz Yes Yes Yes 0 12 AEGON Yes Yes Yes 0 13 Zurich Yes Yes No 0

Summary: Eight out of thirteen listed insurance companies disclose the IAS 39 categories they use for financial liabilities and six of them specify the valuation techniques used to measure the fair value of their financial liabilities. Only two of the listed insurance companies disclose information about their own credit risk being considered in the valuation techniques of their financial liabilities.

Quantification: The disclosure on the valuation methods of financial liabilities is very limited. The reason might be that listed insurance companies do usually not depend on bonds as a financial instrument or that financial liabilities are connected only to certain insurance products (like unit linked products) being sold by the listed insurance companies. Due to these restrictions disclosure about financial liabilities differs strongly among the listed insurance companies and shall not be included in the analysis and therefore not quantified.

326

Appendix C: Scoring documents for measuring risk disclosure performance

Pre Crisis Check-List Summary: Criteria: B. III.

Accounting requirements - exact specification: B. III.

Following general accounting regulation require information about reasons for measuring insecurities. An entity shall disclose the judgements that management has made in the process of applying the entity's accounting policies and that have the most significant effect on the amounts recognized in the financial statements (IAS I par. 122). An entity shall disclose information about the assumptions it makes about the future, and other major sources of estimation uncertainty at the end of the reporting period, that have a significant risk of resulting in a material adjustment to the carrying amounts of assets and liabilities within the next financial year (IAS 1 par. 125).

Results of the analysis:

Name Statement that estimates and assumptions are used in the preparation of the financial statements

General reference that financial instruments that are not quoted on an active market

Other specific assumptions used for financial instruments in the financial statements

Specification of risk from change of assumptions of financial instruments

Quantification of risk from changes in assumptions (actual or pervious year)

1 Old Mutual Yes Yes No No No 1 2 Standard Life No No No No No -1 3 Legal &

General Yes Yes No No No 1

4 CNP Assurances

Yes Yes Yes No No 3

5 Munich Re No No No No No -1 6 ING Yes Yes No No No 1 7 Hannover Re Yes Yes No No No 1 8 Generali Yes Yes No No No 1 9 AXA Yes Yes Yes Yes No 3 10 Aviva Yes Yes No No No I 11 Allianz Yes No No No No 1 12 AEGON Yes Yes No No No 1 13 Zurich Yes Yes Yes Yes No 3

Summary: Eleven out of thirteen listed insurance companies disclose the statement that estimates and assumptions are used in the preparation of their financial statements and ten of them disclose general information about inactive market for financial instruments. Only three listed insurance

companies provide detailed and specific information about the assumptions used and the associated risk. No company quantifies the risk from changes in assumptions.

Quantification: No information provided leads to a negative judgement (-1 points). Statement of estimates and assumptions as well as reference to inactive markets can be considered as general information (1 point). Explanations about assumptions made and the resulting risk can be considered as specific qualitative information (3 points). The information about results from changes in assumptions in this calendar year or before can be considered a quantitative information (5 points).

327

Appendix C: Scoring documents for measuring risk disclosure performance

Pre Crisis Check-List Summary: Criteria: C. I. 1.

Accounting requirements - exact specification: C. I. I.

German accounting requirements on risk disclosure are stated in GAS 15. It is assumed that all listed insurance companies can comply with these requirements by voluntarily disclosing such information.

Management's expectations for the future development of the results of operations and financial position of the group shall be presented and described at a minimum, as either a positive or a negative trend. The effects of the significant influencing factors shall be discussed (GAS 15 par. 89).

Results of the analysis:

# Name General information about future trends

Discussion of significant influencing factors that have a negative impact on the company

Quantification of future negative impact of stated influencing factors

No information disclosed

Results

I Old Mutual No No No Yes -1 2 Standard Life Yes No No No 1 3 Legal &

General No Yes No No 3

4 CNP Assurances

No No No Yes -1

5 Munich Re No Yes No No 3 6 ING Yes No No No 1 7 Hannover Re No No No Yes -1 8 Generali No No No Yes -1 9 AXA No Yes No No 3 10 Aviva No Yes No No 3 11 Allianz Yes No No No 1 12 AEGON No Yes No No 3 13 Zurich No No No Yes -1

Summary: Five out of thirteen listed insurance companies disclose no information about future negative impact on their company. Three companies disclose general information about future trends. Five listed insurance companies discuss significant influencing factors and their negative impact on their company in the future. No company quantifies the negative impact.

Quantification: No information provided leads to a negative judgement (-1 points). General information about future trends leads to a positive judgement (1 point). The discussion of significant influencing factors that have a negative impact on the listed insurance company can be considered as specific qualitative information (3 points). The quantification of future negative impact by stated influencing factors can be considered a quantitative information (5 points).

328

Appendix C: Scoring documents for measuring risk disclosure performance

Pre Crisis Check-List Summary: Criteria: C. I. 2.

Accounting requirements - exact specification: C. I. 2.

German accounting requirements on risk disclosure are stated in GAS 15. It is assumed that all listed insurance companies can comply with these requirements by voluntarily disclosing such information.

The expected development of the group for the following two financial years shall be described, together with significant opportunities and risks associated with this development. The expected volume of future capital expenditures and the expected financial effects shall be discussed (GAS 15 par. 84).

Results of the analysis:

# Name General Specific information Quantified results No information Results information about risk that might or dividends for the disclosed about significantly affect the following year or development group in the following next two financial of the group two financial years OR years in the negative statement that following two such an information is financial not disclosed years

1 Old Mutual Yes No No No 1 2 Standard Life Yes No No No 1 3 Legal & Yes No No No 1

General 4 CNP Yes No No No 1

Assurances 5 Munich Re No No Yes No 5 6 ING No Yes (ne g. statement) No No 3 7 Hannover Re No No Yes No 5 8 Generali No Yes ne . statement) No No 3 9 AXA No Yes (ne q. statement) No No 3 10 Aviva No No Yes No 5 11 Allianz No No Yes No 5 12 AEGON No No No Yes -1 13 Zurich Yes No No No 1

Summary: One out of thirteen listed insurance companies, disclose no result forecast. Five companies disclose general information about developments in the next two years. Three companies disclose a negative statement when they state that a result forecast is not possible. Four listed insurance companies provide forecasts on results or dividends for the next one or two financial years.

Quantification: No information provided leads to a negative judgement (-1 points). General information about the development of the group within the future time period leads to a positive judgement (1 point). The disclosure of specific risk information or a negative statement that no disclosure is provided is measured with three points. The disclosure of future dividends or results can be considered a quantitative information (5 points).

329

Appendix C: Scoring documents for measuring risk disclosure performance

Pre Crisis Check-List Summary: Criteria: C. 11.1

Accounting requirements - exact specification: C. II. 1

German accounting requirements on risk disclosure are stated in GAS 5. It is assumed that all listed insurance companies can comply with these requirements by voluntarily disclosing such information.

It is recommended that inter-dependencies between individual risks are described: it is mandatory to do so, where an appropriate assessment of the risk is otherwise not possible (GAS 5 par. 25)

Results of the analysis:

# Name General Specific information Quantification of No information Results information about about risk about risk risk interdependencies interdependencies interdependencies interdependencies and how they are effects or

measured or diversification recognized in the effects company

1 Old Mutual No No no Yes -1 2 Standard No No No Yes -1

Life 3 Legal & Yes No No No 1

General 4 CNP No No No Yes -1

Assurances 5 Munich Re No Yes No No 3 6 ING No No No Yes -1 7 Hannover No Yes No No 3

Re 8 Generali No No No Yes -1 9 AXA No Yes No No 3 10 Aviva No No No Yes -1 11 Allianz No No No Yes -1 12 AEGON No Yes No No 3 13 Zurich No no No Yes -1

Summary: Eight out of thirteen listed insurance companies disclose no information about future risk interdependencies in their company. One company discloses general information about risk interdependencies. Four listed insurance companies specify information about risk interdependencies in the future. No company quantifies risk interdependency effects or diversification effects.

Quantification: No information provided leads to a negative judgement (-1 points). General information about risk interdependencies leads to a positive judgement (1 point). Specific information disclosed about interdependencies and how they are measured or recognized in the company can be considered as specific qualitative information (3 points). The quantification of risk interdependency effects or diversification effects can be considered quantitative information (5 points).

330

Appendix C: Scoring documents for measuring risk disclosure performance

Pre Crisis Check-List Summary: Criteria: C. II. 2.

Accounting requirements - exact specification: C. 11.2.

For each type of risk arising from financial instruments, an entity shall disclose concentration of risk (IFRS 7 par. 34c). If the quantitative data disclosed as at the reporting date are unrepresentative of entity's exposure to risk during the period, an entity shall provide further information that is representative (IFRS 7 par. 35).

Results of the analysis:

#- Name General Specific and Quantified data on No information Results information representative risk concentration disclosed about risk information about risk concentration concentration arising arising from from financial financial instruments AND instruments negative statement that

no risk concentration exists

1 Old Mutual No No No Yes -1 2 Standard Life No No No Yes -1 3 Legal & No No No Yes -1

General 4 CNP No No No Yes -1

Assurances 5 Munich Re No Yes No No 3 6. ING No Yes No No 3 7 Hannover Re No No Yes No 5 8 Generali No No No Yes -1 9 AXA No No No Yes -1 10 Aviva No No No Yes -1 11 Allianz No Yes No No 3 12 AEGON No No Yes No 5 13 Zurich No No No Yes -1

Summary: Eight out of thirteen listed insurance companies disclose no information about risk concentration. Three companies disclose specific information about risk concentrations (for example investments by economic sectors). Two listed insurance companies provide quantitative information about risk concentration arising from financial instruments.

Quantification: No information provided leads to a negative judgement (-1 points). General information about risk concentration from financial instruments leads to a positive judgement (1 point). The disclosure of specific and representative information about risk concentration or a negative statement that no risk concentration exists is measured with three points. The disclosure of quantitative data on risk concentration can be considered quantitative information (5 points).

331

Appendix C: Scoring documents for measuring risk disclosure performance

Pre Crisis Check-List Summary: Criteria: C. 11.3.

Accounting requirements - exact specification: C. 1I. 3.

German accounting requirements on risk disclosure are stated in GAS 5. It is assumed that all listed insurance companies can comply with these requirements by voluntarily disclosing such information.

In the case of risk threatening the existence of an enterprise, the forecast period is generally one year (GAS 5 par. 24).

Results of the analysis:

# Name General information about risk threatening the existence of the company

Specific information about risk threatening the existence of the company OR Negative statement

No information disclosed

Results

I Old Mutual No No Yes -1 2 Standard Life No No Yes -1 3 Legal &

General No Yes (negative statement) No 3

4 CNP Assurances

No No Yes -1

5 Munich Re No No Yes -1 6 ING No No Yes -1 7 Hannover Re No No Yes -1 8 Generali No No Yes -1 9 AXA No No Yes -1 10 Aviva No No Yes -1 11 Allianz Yes No No 1 12 AEGON No Yes No 3 13 Zurich No Yes No 3

Summary: Nine out of thirteen listed insurance companies disclose no information about risk threatening the existence of the company. One listed insurance company discloses general information. Two'companies disclose specific information and one company makes a negative statement.

Quantification: No information provided leads to a negative judgement (-1 points). General information about risk threatening the existence of the company leads to a positive judgement (1 point). The disclosure of specific information about risk threatening the existence of the company or a negative statement that no such risk exists is measured with three points. A quantification of the risk is not possible.

332

Appendix C: Scoring documents for measuring risk disclosure performance

Post Crisis Check-List Summary: Criteria: A. 1.1. and A. I. 2.

Accounting requirements - exact specification: A. 1.1. Investment risk shall be measured on the basis of the recognised risk-monitoring methods used by the group. (GRS 5-20 par. 29) Depending on the type of risk involved, one of the following methods is generally appropriate to measure investment risk: - determination of the anticipated loss using a pre-defined level of confidence and probability based models, such as stress tests to take account of crisis scenarios: the use of back testing techniques is recommended in this context, - asset/liability management techniques - scenario of sensitivity analysis (GRS 5-20 par. 30)

A. l. 2. The disclosures made on the market risk of investments shall, as a minimum, contain the following information, if this has not already been quantified by the risk monitoring methods used by the group. (GRS 5-20 par. 31)

- the effect of a 20% fall in the fair value of shares and other non-fixed interest securities - the effect of a 1% shift in the interest curve (upwards or downwards) on the market

value of fixed interest securities and bonds

Results of the analysis: W Name Scenarios

interest and market value of equity variation

Effect on Embedded Value, Value of in- force business, Value at Risk or Earnings at risk quantified

Effect on equity quantified

Effect on earnings quantified

Result

I Old Mutual Yes Yes No No 5 2 Standard Life Yes Yes Yes Yes 5 3 Legal &

General Yes Yes No No 5

4 CNP Assurances Yes Yes No No 5

5 Munich Re Yes Yes Yes Yes 5 6 ING Yes Yes Yes Yes 5

7 Hannover Re Yes Yes Yes No 5

8 Generali Yes Yes Yes No 5

' Yes Yes No No 5

10 Aviva Yes Yes Yes Yes 5

11 Allianz Yes Yes Yes No 5

12 AEGON Yes No Yes Yes 5

13 Zurich Yes Yes No No 5

Summary: All listed insurance companies quantify their investment risk using predominantly probability based models (European Embedded Value, Group Embedded Value, Value at Risk, Earnings at Risk or Value of in-force business) mentioned as accepted methods according to GRS 5-20 par. 29 and 30. Only eight of the listed companies uses the alternative method named under GRS 5-20 par. 31 disclosing information on sensitivities that affect equity or earnings. A little

333

less than half of the listed insurance companies disclose that information in addition to the predominantly used methods. All listed insurance companies fulfil the accounting requirements with a quantified investment risk disclosure in their financial statements as of December 31st, 2008.

Quantification: The accounting requirements can be fulfilled with scenarios showing the effect on equity, earnings and embedded value. This information can be considered quantitative and are granted 5 points.

334

Appendix C: Scoring documents for measuring risk disclosure performance

Post Crisis Check-List Summary: Criteria: A. I. 3.

Accounting requirements - exact specification: A. 1.3.

An insurer shall disclose information that enables users of its financial statements to evaluate the nature and extent of risk arising from insurance contracts. To comply with this, an insurer shall disclose information about exposures to market risk arising from embedded derivatives contained in a host insurance contract if the insurer is not required to, and does not, measure the embedded derivatives at fair value. (IFRS 4 par. 38 and 39e)

Results of the analysis:

# Name Scenarios interest and market value of equity variation

Disclosure that entire combined contract is treated in the accounting category "financial assets at fair value through profit/loss'

Result

I Old Mutual Yes Yes 3 2 Standard Life Yes Yes 3 3 Leal & General Yes No 3 4 CNP

Assurances Yes Yes 3

5 Munich Re No No 3 6 ING Yes No 3 7 Hannover Re Yes Yes 3 8 Generali Yes Yes 3 9 AXA Yes No 3 10 Aviva Yes No 3 11 - Allianz Yes No 3 12- AEGON Yes No 3 13 Zurich Yes No 3

Summary: All listed insurance companies disclose that embedded derivatives are separated and measured at fair value. The majority of listed insurance companies does not disclose that certain embedded derivatives are not separable so that - in order to measure them at fair value - the entire combined contract is measured at fair value by classifying it to the IAS 39 category "financial assets at fair value through profit/loss" All listed insurance companies fulfil the accounting requirements with a quantified investment risk disclosure in their financial statements as of December 31st, 2008.

Quantification: Information about embedded derivatives that are separated and measured at fair value can be considered a specific qualitative disclosure (3 points).

335

Appendix C: Scoring documents for measuring risk disclosure performance

Post Crisis Check-List Summary: Criteria: A. I. 1. crisis

Accounting requirements - exact specification: A. I. 1. crisis.

There are no accounting requirements about risk reducing policies. However this information is essential in a crisis situation for the investor. The knowledge how the company is preparing itself against the effects of the financial crisis allows investors to judge the company and how the management leads the company.

To name and describe risk reducing policies like use of derivatives, sale of certain financial assets or diversification of investments is required. A quantification of the effect of risk reducing policies is not possible, because not every policy con be accounted for as hedging under the strict hedging requirements of IAS 39.

Results of the analysis: # Name Use of

derivatives as a risk reducing policy

Name the kid of derivative and what kind of risk it is reducing

Sale of equity as a risk reducing policy

Diversification as a risk reducing policy

Result

1 Old Mutual Yes Yes No Yes 3 2 Standard Life Yes Yes No No 3 3 Legal &

General Yes Yes Yes Yes 3

4 CNP Assurances

Yes Yes Yes No 3

5 Munich Re Yes Yes Yes No 3 6 ING Yes Yes Yes No 3 7 Hannover Re Yes Yes Yes Yes 3 8 Generali Yes Yes No No 3 9 AXA Yes Yes No No 3 10 Aviva Yes Yes Yes No 3 11 Allianz Yes Yes Yes No 3 12 AEGON Yes Yes Yes No 3 13 Zurich Yes Yes No No 3

Summary: All listed insurance companies use derivatives as one policy to reduce market risk from interst bearing financial instruments and equity financial instruments. Eight out of thirteen companies sold equity financial instruments in order to reduce their market risk and three the listed insurance companies disclosed diversification as a policy to reduce their market risk. By naming and describing at least one risk reducing policy all listed insurance companies disclose specific qualitative information on this criteria in their financial statements as of December 31st, 2008.

Quantification: No information provided leads to a negative judgement (-1 points). General information about risk reducing policies lead to a positive judgement (1 point). The disclosure of specific information about risk reducing strategies is measured with three points. The disclosure of quantitative data policies is not possible, because not every policy can be accounted for as hedging under the strict hedging requirements of AS 39.

336

Appendix C: Scoring documents for measuring risk disclosure performance

Post Crisis Check-List Summary: Criteria: A. I. 3. crisis

Accounting requirements - exact specification: A. 1.3. crisis

There are no accounting requirements about the disclosure of how a listed insurance company is affected by changes in the market value of certain financial instruments. However this information is essential in the crisis situation for the investor. The investor gets to know through the media the names of defaulted banks and toxic assets. They want to understand to what degree the listed insurance company is affected by this.

This crisis specific risk disclosure can consist of three parts: " To name the categories of investments by listed insurance companies that are

predominantly affected by the financial crisis " To mention the reaction by the listed insurance companies to handle the market risk

arising from these specific investments To quantify the market risk of these investments by disclosing their book value, their current fair value, the unrealized losses and the realized losses (impairments) of these investments.

"' To disclose the expectation of the capital market (rating) towards these investments.

Results of the analysis:

Name Category of critical investments

Reaction towards risk

Valuation of critical investments Specification Result

Exposure to subprime

Exposure to default banks

Reducing mean

Book value

Fair value

Unrealized loss-

Impairment Investment grade of critical investments

1 Old Mutual Yes Yes No Yes Yes No No Yes 1 2 Standard

Life Yes Yes Yes Yes No No No Yes 3

3 Legal & General

Yes No Yes No Yes No No Yes 3

4 CNP Assurances

Yes Yes No Yes No No Yes No 1

5 Munich Re Yes Negativ Yes Yes No No Yes No 3 6 ING Yes Yes Yes Yes Yes Yes Yes Yes 5 7 Hannover

Re Yes Yes No No Yes No - Yes Yes I

8 Generali No No No No No No No No -1 9 AXA Yes No No Yes Yes Yes Yes Yes 5 10 Aviva Yes Yes Yes No No No No No 3 11 Allianz Yes No No Yes No No No No 1 12 AEGON Yes No Yes Yes Yes Yes Ne ativ Yes 3 13 Zurich No Yes No Yes Yes Yes Yes Yes 5

Summary: Listed insurance companies are affected differently by the fall in market value of critical investments. With the exception of Generali and Zurich all listed insurance companies state that they have subprime investments in the forms of asset backed securities, residual mortgage backed securities, commercial mortgage backed securities or commercial debt obligations. Generali and Zurich do not mention any involvement in subprime critical investments but do not state that they are not affected at all (negative statement). More than half of the listed insurance companies disclose their involvement in investments with critical banks such as Lehman, Washington Mutual and Monoliners. Only Munich Re gives a negative

337

statement about this involvement. The other listed insurance companies do not disclose information about their risk exposure.

Half of the listed insurance companies disclose activities to reduce the exposure to either subprime or default bank risk. They name swaps, derivatives and sale of bank shares to reduce their risk exposure.

All but three listed insurance companies that disclose their exposure to subprime or bank default investment risk explicitly mention the book value of their involvement. But only together with. the disclosure of fair value or losses (unrealized or impaired) can investors evaluate how far the risk has been realized. Only half of the listed insurance companies provide this kind of information.

The loss that can be further expected from the subprime or default bank investments of each listed, insurance company can be predicted on the basis of the opinion of rating agencies on the investments. Eight listed insurance companies disclose that information about their crisis specific risk investments.

Quantification: Unless explicitly stated by a negative information it can be assumed that all listed insurance companies as major investors are affected by subprime or default bank investments. When no involvement is disclosed it is a no information -1 Point. General information are provided when the investment with market risk specific to the crisis are disclosed and the book or fair value disclosed 1 Point. Specific information are disclosed when additionally risk reducing means to cope with the risks are mentioned 3 Points. Quantitative information is provided when the historic and outlook are quantified with the amount of impairment and the rating of crisis specific investments are disclosed 5 Points.

Special advice: Listed insurance companies can be affected by the crisis because they own crisis specific risky investments with a high market price erosion, or be not affected at all. These companies can either disclose their risk or they cannot. So four types of risk disclosure performance is possible: Type A: High market risk from crisis specific investments in their portfolio, extensive risk disclosure in their annual financial statements Type B: High market risk from crisis specific investments in their portfolio, little or no risk disclosure in their annual financial statements Type C: Low or no market risk from crisis specific investments in their portfolio, extensive risk disclosure in their annual financial statements - Type D: Low or no market risk from crisis specific investments in their portfolio, little or no risk disclosure in their annual financial statements The extent of risk from crisis specific investments cannot be detected unless disclosed in the annual financial statements. However the reduction of equity from the beginning of the crisis until 2008 can be an indication of financial difficulties that arose as a result of market value erosion of crisis specific investments.

338

Appendix C: Scoring documents for measuring risk disclosure performance

Post Crisis Check-List Summary: Criteria: A. II. 1.

Accounting requirements - exact specification: A. II. 1.

An insurer shall disclose information that enables users of its financial statements to evaluate the nature and extent of risk arising from insurance contracts. (IFRS 4 par. 38) An insurer shall disclose information about credit risk that IFRS 7 par. 31-42 would require if the insurance contracts were within the scope of IFRS 7. (IFRS 4 par. 39d) An entity shall disclose information that enables users of its financial statements evaluate the nature and extent of risk arising from financial instruments to which the entity is exposed at the reporting date. (IFRS 7 par. 31) The disclosures focus on the risk that arise from financial instruments and how they have been managed. (IFRS 7 par. 32)

Following information are required for each credit risk arising from financial instruments:

" Quantification of maximum exposure to credit risk at the reporting date. (IFRS 7 par. 34 and 36a)

" Information about the credit quality of financial assets (IFRS 7 par. 36c) " Analysis of age of financial assets that are past due as at the reporting date but not

impaired (IFRS 7 par. 37a)

Results of the analysis: #T Name : Quantification

of credit risk exposure by investments

Disclosure of rating information of investments

Disclosure of age of investments that are past due but not impaired

Disclosure of change in fair value attributable to change in credit risk

Result

1 Old Mutual Yes Yes Yes Yes 5 2 Standard Life Yes Yes No No 3 3 Legal &

General Yes Yes Yes Yes 5

4 CNP Assurances

Yes Yes No No 3

5 Munich Re Yes Yes No No 3 6 ING Yes Yes No No 3 7 Hannover Re Yes Yes No No 3 8 Generali Yes Yes No No 3 9 AXA Yes Yes No No 3 10 Aviva Yes Yes Yes Yes 5 11 Allianz Yes Yes No No 3 12 AEGON Yes Yes No No 3 13 Zurich Yes Yes No No 3

Summary: All listed insurance companies quantify their credit risk exposure mostly by disclosing it together with the ratings of their investments. The required disclosure of the aging of investments is disclosed by only two listed insurance companies and the non required disclosure of change in fair value attributable to changes in credit risk is provided by three listed insurance companies.

Quantification: Required information about the aging of investments are not disclosed by all listed insurance companies. Therefore quantitative information does not receive 5 points but only 3 points. If

339

additionally the required information about the aging of investments is disclosed 5 points are granted.

340

Appendix C: Scoring documents for measuring risk disclosure performance

Post Crisis Check-List Summary: Criteria: A. II. 1. crisis

Accounting requirements - exact specification: A. II. 1. crisis

There are no accounting requirements about the disclosure of how a listed insurance company is affected by defaults of creditors of certain financial instruments. However this information is essential in the crisis situation for the investor. The investor gets to know through the media the names of defaulted banks. They want to understand to what degree the listed insurance company is affected by these defaults.

This crisis specific risk disclosure can consist of three parts: " To describe the financial involvement of listed insurance companies by defaulted

creditors in the financial crisis that are common knowledge to the public (Lehman and Iceland banks)

" To quantify the financial involvement of the listed insurance company with these creditors

" To quantify the loss the listed insurance company took as a result of the default of these creditors

Results of the analysis:

# Name Qualitative description of financial involvement

Quantification of involvement

Quantification of loss

Result

1 Old Mutual Yes Yes Yes 5 2 Standard Life No No No -1 3 Legal & General Yes Yes Yes 5 4 CNP Assurances Yes Yes Yes 5 5 Munich Re Yes Yes Yes 5 6 ING Yes Yes Yes 5 7 Hannover Re Yes Yes Yes 5 8 Generali No No No -1 9 AXA Ne ative No No 3 10 Aviva Yes No No 3 11 Allianz No No No -1 12 AEGON Yes Yes Yes 5 13 Zurich No No No -1

Summary: Listed insurance companies are affected differently by the default of Lehman and Icelandic banks as creditors of their investments. Eight listed insurance companies state that they have investments by the above mentioned creditors. Four listed insurance companies do not mention any involvement with these creditors but only AXA states that it is not affected at all (negative statement). More than half of the listed insurance companies quantify their financial involvement in investments with critical banks such as Lehman, Washington Mutual or Icelandic banks which defaulted. The other listed insurance companies do not quantify their risk credit exposure.

Half of the listed insurance companies disclose activities to reduce the exposure to either subprime or default bank risk. They name swaps, derivatives and sale of bank shares to reduce their risk exposure.

Seven listed insurance companies disclose loss.

341

Quantification: Unless explicitly stated by a negative information it can be assumed that all listed insurance companies as major investors are affected by the default of the creditors Lehman, Washington Mutual or Icelandic banks. When no involvement is disclosed it is regarded as no information - 1 Point. To name involvement in the default of the above mentioned creditors is considered a specific general information 3 Points. Quantitative information on the involvement and the loss are disclosed 5 Points.

Special advice: Listed insurance companies can be affected by the crisis because they own crisis specific risky investments with a high market price erosion, or be not affected at all. These companies can either disclose their risk or they cannot. So four types of risk disclosure performance is possible: Type A: High market risk from crisis specific investments in their portfolio, extensive risk disclosure in their annual financial statements Type B: High market risk from crisis specific investments in their portfolio, little or no risk disclosure in their annual financial statements Type C: Low or no market risk from crisis specific investments in their portfolio, extensive risk disclosure in their annual financial statements Type D: Low or no market risk from crisis specific investments in their portfolio, little or no risk disclosure in their annual financial statements The extent of risk from crisis specific investments cannot be detected unless disclosed in the annual financial statements. However the reduction of equity from the beginning of the crisis until 2008 can be an indication of financial difficulties that arose as a result of market value erosion of crisis specific investments.

342

Appendix C: Scoring documents for measuring risk disclosure performance

Post Crisis Check-List Summary: Criteria: A. III. 1.

Accounting requirements - exact specification: A. I11.1.

An insurer shall disclose information that enables users of its financial statements to evaluate the nature and extent of risk arising from insurance contracts. To comply with this, an insurer shall disclose information about exposures to liquidity risk that IFRS 7 par. 31 - 42 would require if the insurance contracts were within the scope of IFRS 7 (IFRS 4 par. 39d). An entity shall disclose a description of how it manages the liquidity risk (IFRS 7 par. 39b). An insurer need not to provide the maturity analysis required by IFRS 7 par. 39a if it discloses information about the estimated timing of the net cash outflows resulting from recognised insurance liabilities instead.

Results of the analysis:

# Name Disclosure about liquidity risk inherent and Asset liability matching performed

Quantification of liquidity risk Result

Timing of returns from assets

Timing of obligations from insurance contracts

1 Old Mutual Yes No No 1 2 Standard Life Yes Yes No 5 3 Legal &

General Yes No No 1

4 CNP Assurances

Yes Yes No 5

5 Munich Re Yes No Yes 3 6 ING Yes No No I 7 Hannover Re Yes No No 1 8 Generali Yes Yes Yes 5 9 AXA Yes No No I 10 Aviva Yes No Yes 3 11 Allianz Yes No No 1 12 AEGON Yes No Yes 3 13 Zurich Yes No Yes 3

Summary: All listed insurance companies disclose that they actively manage their liquidity risk. As recommended in the IFRS 7 requirements only three listed insurance companies describe their liquidity management by quantifying the timing of return from their assets. Five listed insurance companies fulfil the requirements by disclosing the timing of projected payments from their insurance obligations.

Quantification: All information disclosed about the liquidity risk and how it is managed can be considered as general qualitative information (1 point). The legally required disclosure about the timing of obligations from insurance contracts can be considered a specific qualitative information (3 points) and the voluntary disclosure about the timing of returns from assets as a qualitative information (5 points).

343

Appendix C: Scoring documents for measuring risk disclosure performance

Post Crisis Check-List Summary: Criteria: A. III. 2. crisis

Accounting requirements - exact specification: A. 111.2. crisis

There is no accounting regulation that requires a sensitivity analysis in regard to liquidity risks by listed insurance companies. However as liquidity is essential to insurance companies significant changes in the market value of investments, in the credit risk of investment issuers and changes in the projected cash flow due to the crisis should be commented by listed insurance companies. This can be done most convincingly for investors by quantifying effects and scenarios.

The cash outflow of insurance related factors (lapse or mortality) can change as a result that these factors differ from historic assumptions. The cash inflow of investments can differ from historic assumptions due to a significant loss in market value or the default of creditors. These effects of changing cash flows can be disclosed as sensitivities to earnings/equity or Embedded Values.

Results of the analysis:

# Name Cash outflow sensitivity Cash inflow sensitivity (investment related)

Result

Effect on earnings or equity disclosed

Effect on Embedded Value disclosed

Effect on earning or equity disclosed

Effect on Embedded Value disclosed

1 Old Mutual Yes No No No 3 2 Standard Life Yes No No No 3 3 Legal &

General No Yes No No 3

4 CNP Assurances

No Yes No No 3

5 Munich Re No Yes No Yes 5 6 ING No No No No 1 7 Hannover Re No No No No 1 8 Generali No Yes No No 3 9 AXA No No No " No 1 10 Aviva Yes No No No 3 11 Allianz No No No No 1 12 AEGON Yes No No No 3- 13 Zurich Yes No No No 3

Assumption: Only the interest and market value sensitivity is recognized in this analysis and not changes in the foreign exchange rate.

Summary: Nine out of thirteen listed insurance companies disclose the' effect that changes in historic insurance related assumptions have on either earnings/equity or the embedded value. One out of thirteen listed insurance companies disclose the effect that changes in their investment related assumptions have on either earnings/equity or the embedded value.

Quantification: Since the disclosure is not required all listed insurance companies that do, not provide information receive 1 point. All companies that disclose cash outflow sensitivities on either earning/equity or Embedded values receive about cash inflow or cash outflow receive 3 Points. Listed insurance companies that disclose in addition to that sensitivities on cash in- and cash outflows receive 5 points.

344

Appendix C: Scoring documents for measuring risk disclosure performance

Post Crisis Check-List Summary: Criteria: B. I.

Accounting requirements - exact specification: B. I.

Following general accounting regulation require information about how financial instruments on the debit side of the balance sheet are measured. An entity shall disclose in the summary of significant accounting policies the measurement basis used in preparing the financial statements (IAS 1 par. 117a). It is important for an entity to inform users of the measurement basis used in the financial statements, because the basis on which an entity prepares the financial statements significantly affects users'analysis ((AS I par. 118).

The accounting standard IFRS 7 on the disclosure of financial instruments specifies theses requirements: An entity shall disclose the valuation techniques used, the assumptions applied in determining fair values of each lass of financial assets. (IFRS 7 par. 27a) An entity shall disclose whether fair values are determined in whole or in part directly by reference to published price quotations in an active market or are estimated using a valuation technique. (IFRS 7 par. 27b and IAS 39 AG 71 - 79

Results of the analysis:

# Name Categories of IAS Fair value determination Specification Quantification of 39 used and of these financial assets disclosure of their alternative measured at fair effect on the valuation value without balance sheet and technique active markets profittloss IFRS 7)

Directly Indirectly measured by measured reference to by an active alternative market valuation

technique I Old Mutual Yes Yes Yes No No 3 2 Standard Life Yes Yes No No No 1 3 Legal & Yes Yes Yes No No 3

General 4 CNP Yes Yes Yes No No 3

Assurances 5 Munich Re Yes Yes Yes Yes No 3 6 ING Yes Yes Yes No No 3 7 Hannover Re Yes Yes Yes Yes No 3 8 Generali Yes Yes No No No 1 9 AXA Yes Yes Yes Yes Yes 5 10 Aviva Yes Yes Yes Yes Yes 11 Allianz Yes Yes Yes Yes No 3 12 AEGON Yes Yes Yes Yes No 3 13 Zurich Yes Yes Yes No No 3

Summary: All listed insurance companies disclose the IAS 39 categories used and their effect on either earnings or equity. All companies disclose their directly measured references to an active market but two listed companies do not refer to alternative valuation techniques. Although eleven companies using them six listed insurance companies specify these valuation methods and only two quantify the financial instruments measured with these valuation techniques.

345

Quantification: Disclosure requirements about the IAS 39 categories and the fair value measurement on active markets can be considered as general information (1 point). Explanations about alternative valuation techniques or the specification of these techniques can be considered specific qualitative information (3 points). The information how many financial instruments are measured using these alternative valuation techniques is a quantitative information (5 points).

346

Appendix C: Scoring documents for measuring risk disclosure performance

Post Crisis Check-List Summary: Criteria: B. II.

Accounting requirements - exact specification: B. II.

Following general accounting regulation require information about how financial instruments on the debit side of the balance sheet are measured. An entity shall disclose in the summary of significant accounting policies the measurement basis used in preparing the financial statements (IAS 1 par. 1 17a). It is important for an entity to inform users of the measurement basis used in the financial statements, because the basis on which an entity prepares the financial statements significantly affects users'analysis (IAS 1 par. 118).

Results of the analysis:

# Name Categories of AS 39 used and disclosure of their effects on the balance sheet and profit/loss

Specific measuring techniques of fair values for financial liabilities

Disclosure about own credit risk considered in the valuation techniques of financial liabilities

Result

1 Old Mutual Yes Yes No 0 2 Standard Life No No No 0 3 Legal &

General No No No 0

4 CNP Assurances

No No No 0

5 Munich Re No No No 0 6 ING Yes Yes No 0 7 Hannover Re Yes Yes No 0 8 Generali Yes No No 0 9 AXA Yes No No 0 10 Aviva No No No 0 11 Allianz Yes Yes Yes 0 12 AEGON Yes Yes Yes 0 13 Zurich Yes Yes No 0

Summary: Eight out of thirteen listed insurance companies disclose the IAS 39 categories they use for financial liabilities and six of them specify the valuation techniques used to measure the fair value of their financial liabilities. Only two of the listed insurance companies disclose information about their own credit risk being considered in the valuation techniques of their financial liabilities.

Quantification: The disclosure on the valuation methods of financial liabilities is very limited. The reason might be that listed insurance companies do usually not depend on bonds as a financial instrument or that financial liabilities are connected only to certain insurance products (like unit linked products) being sold by the listed insurance companies. Due to these restrictions disclosure about financial liabilities differs strongly among the listed insurance companies and shall not be included in the analysis and therefore not quantified.

347

Appendix C: Scoring documents for measuring risk disclosure performance

Post Crisis Check-List Summary: Criteria: B. III.

Accounting requirements - exact specification: B. III.

Following general accounting regulation require information about reasons for measuring insecurities. An entity shall disclose the judgements that management has made in the process of applying the entity's accounting policies and that have the most significant effect on the amounts recognized in the financial statements (IAS 1 par. 122). An entity shall disclose information about the assumptions it makes about the future, and other major sources of estimation uncertainty at the end of the reporting period, that have a significant risk of resulting in a material adjustment to the carrying amounts of assets and liabilities within the next financial year (lAS 1 par. 125).

Results of the analysis:

# Name Statement that estimates and assumptions are used in the preparation of the financial statements

General reference that financial instruments that are not quoted on an active market

Other specific assumptions used for financial instruments in the financial statements

Specification of risk from change of assumptions of financial instruments

Quantification of risk from changes in assumptions (actual or pervious year)

1 Old Mutual Yes Yes No No No 1 2 Standard Life No No No No No -1 3 Legal &

General Yes Yes No No No I

4 CNP Assurances

Yes Yes Yes No No 3

5 Munich Re No No No No No -1 6 ING Yes Yes No No No 1 7 Hannover Re Yes Yes No No No 1 8 Generali Yes Yes No No No 1 9 AXA Yes Yes Yes Yes No 3 10 Aviva Yes Yes Yes No No 3 11 Allianz No No No No No -1 12 AEGON Yes Yes Yes Yes No 3 13 Zurich Yes Yes No No No 1

Summary: Ten out of thirteen listed insurance companies disclose the statement that estimates and assumptions are used in the preparation of their financial statements and all of them disclose general information about inactive market for financial instruments. Four listed insurance companies provide detailed and specific information about the assumptions used and the associated risk. No company quantifies the risk from changes in assumptions.

Quantification: No information provided leads to a negative judgement (-1 points). Statement of estimates and assumptions as well as reference to inactive markets can be considered as general information (1, point). Explanations about assumptions made and the resulting risk can be considered as specific qualitative information (3 points). The information about results from changes in assumptions in this calendar year or before can be considered a quantitative information (5 points).

348

349

Appendix C: Scoring documents for measuring risk disclosure performance

Post Crisis Check-List Summary: Criteria: B. I. crisis.

Accounting requirements - exact specification: B. I. crisis

There are not accounting regulations that require listed insurance companies to disclose information about the measuring methods of certain financial instruments that are affected by the financial crisis. However as the names of certain financial instruments have become public knowledge by information in the media, investors want to know how these critical financial instruments are measured especially when no valuation is possible on active markets. This information enables investors to make a judgement about the possible future risk of the listed insurance company and about how the management is handling the crisis using aggressive or conservative valuation methods.

Results of the analysis:

# Name Disclosure of asset types for which inactive market exists (qualitative by naming or quantitative by amount)

Specification about valuation method used instead of market prices to determine fair value

Disclosure about alternative sources of data that are used to determine the fair value

I Old Mutual No Yes Yes 3 2 Standard Life Yes Yes Yes 3 3 Legal &

General Yes No Yes 3

4 CNP Assurances

No No Yes 3

5 Munich Re No No - Yes 3 6 ING Yes Yes Yes 3 7 Hannover Re No Yes Yes 3 8 Generali Yes Yes Yes 3 9 AXA Yes Yes Yes 3 10 Aviva Yes No Yes 3 11 Allianz Yes Yes Yes 3 12 AEGON Yes Yes Yes 3 13. Zurich No No No -I

Summary: All but one listed insurance companies disclose specific information about alternative sources of data that are used to determine the fair value. Eight companies disclosed specific information about the valuation method. Only

, eight listed insurance . companies disclosed

general information of the asset types which are affected by inactive markets.

Quantification: No information disclosed about alternative valuation methods for financial instruments accounted for at fair value when the valuation process cannot rely on active markets leads to negative points (-1 point). General information about the asset types for which inactive markets exists can be considered a general information (1 point). Specifications about the method used instead or the information sources to determine the fair value can be considered a specific information (3 points). A quantification is not possible.

350

Appendix C: Scoring documents for measuring risk disclosure performance

Post Crisis Check-List Summary: Criteria: B. IV. crisis

Accounting requirements - exact specification: B. IV. crisis

The amendment to IAS 39 issued in October 2008 permits an entity to reclassify " Non-derivative financial assets at fair value through profit or loss category and " Non-derivative financial assets from the available-for-sale category

to the loans and receivables category (IAS 39 amendment IN8A) This entity shall disclose:

" The amount reclassified into and out of each category " For each reporting period until derecognition, the carrying amounts and fair values of all

financial assets that have been reclassified " If a financial asset was reclassified the facts and circumstances indicating that the

situation was rare " For the reporting period when the financial asset was reclassified the fair value gain or

loss on the financial asset recognised in profit or loss in that reporting period and in the previous reporting period (IFRS 7 amendment 12A)

Results of the analysis:

# Name General Negative Specific disclosure required when Result information about statement about reclassification option is used by the amendments and the use of entity option to reclassify reclassification

option Amount Specification of rare reclassified, circumstances and carrying quantification of amount and fair profit and loss effect value.

1 Old Mutual Yes Yes No No 3 2 Standard Life Yes Yes No No 3 3 Legal & Yes Yes No No 3

General 4 CNP Yes Yes No No 3

Assurances 5 Munich Re Yes Yes No No 3 6 ING Yes No Yes No 1 7 Hannover Re Yes Yes No No 3 8 Generali Yes Yes No No 3 9 AXA Yes Yes No No 3 10 Aviva No No No No -1 11 Allianz Yes Yes No No 3 12 AEGON Yes Yes No No 3 13 Zurich Yes Yes No No 3

Summary: One listed insurance company does not disclose any information. One company used the reclassification option but did not provide all necessary disclosure requirements. Eleven out of thirteen listed insurance companies provided general information and disclosed a negative statement that they did not make use of the reclassification.

Quantification: Information about the IFRS 7 and IAS 39 amendments can be considered as general information (1 point). When no information is provided this is considered with minus one point

351

(-1). Information required by IFRS 7 amendment 12A or a negative statement that the reclassification option is not used by the listed insurance company can be considered as specific qualitative information (3 points). Quantification that exceeds the disclosure requirements is not possible.

352

Appendix C: Scoring documents for measuring risk disclosure performance

Post Crisis Check-List Summary: Criteria: C. I. 1.

Accounting requirements - exact specification: C. I. 1.

German accounting requirements on risk disclosure are stated in GAS 15. It is assumed that all listed insurance companies can comply with these requirements by voluntarily disclosing such information. The financial crisis can be considered as a significant negative impact that should be addressed by the listed insurance companies in order to timely and sufficiently inform their investors

Management's expectations for the future development of the, results of operations and financial position of the group shall be presented and described at a minimum, as either a positive or a negative trend. The effects of the significant influencing factors shall be discussed (GAS 15 par. 89).

Results of the analysis:

# Name General information about future trends

Discussion of significant influencing factors that have a negative impact on the company

Quantification of future negative impact of stated influencing factors

No information disclosed

Results

1 Old Mutual Yes No No No 1 2 Standard Life Yes No No No 1 3 Legal &

General No No No Yes -1

4 CNP Assurances

No No No Yes -1

5 Munich Re No Yes No No 3 6 ING No Yes No No 3 7 Hannover Re No No No Yes -1 8 Generali Yes No No No I 9 AXA Yes No No No 1 10 Aviva No Yes No No 3 11 Allianz No Yes No- No 3 12 AEGON No Yes No No 3 13 Zurich Yes No No No 1

Summary: Three out of thirteen listed insurance companies disclose no information about future negative impact on their company. Five companies disclose general information about future trends. Five listed insurance companies discuss significant influencing factors and their negative impact on their company in the future. No company quantifies the negative impact.

Quantification: No information provided leads to a negative judgement (-1 points). General information about future trends leads to a positive judgement (1 point). The discussion of significant influencing factors that have a negative impact on the listed, insurance company can be considered as specific qualitative information (3 points). The quantification of future negative impact by stated influencing factors can be considered quantitative information (5 points).

.

353

Appendix C: Scoring documents for measuring risk disclosure performance

Post Crisis Check-List Summary: Criteria: C. I. 2.

Accounting requirements - exact specification: C. I. 2.

German accounting requirements on risk disclosure are stated in GAS 15. It is assumed that all listed insurance companies can comply with these requirements by voluntarily disclosing such information. It is recognized that as a result of the financial crisis it is harder to forecast results.

The expected development of the group for the following two financial years shall be described, together with significant opportunities and risks associated with this development. The expected volume of future capital expenditures and the expected financial effects shall be discussed (GAS 15 par. 84).

Results of the analysis:

# Name General Specific information Quantified results No information Results information about risk that might or dividends for the disclosed about significantly affect the following year or development group in the following next two financial of the group two financial years OR years in the negative statement that following two such an information is financial not disclosed years

1 Old Mutual No No No Yes -1 2 Standard Life No No No Yes -1 3 Legal & Yes No No No 1

General 4 CNP Yes No No No 1

Assurances 5 Munich Re No Yes No No 3 6 ING No Yes (ne g. statement) No No 3 7 Hannover Re No No Yes No 5 8 Generali No Yes No No 3 9 AXA No Yes (ne g. statement) No No 3 10 Aviva No No No Yes -1 11 Allianz No Yes No No 3 12 AEGON No No No Yes -1 13 Zurich Yes No No No 1

Summary: Four out of thirteen listed insurance companies, disclose no result forecast. Three companies disclose general information about developments in the next, two years. Two companies disclose a negative statement when they state that a result forecast is not possible and two companies disclose specific information about future risk. One, listed insurance company provides forecasts on results or dividends for the next one or two financial years.

Quantification: No information provided leads to a negative judgement (-1 points). General information about the development of the group within the future time period-leads to a positive judgement (1

point). The disclosure of specific risk information or a negative statement that no disclosure is

provided is measured with three points. The disclosure of future dividends or results can be

considered a quantitative information (5 points).

354

Appendix C: Scoring documents for measuring risk disclosure performance

Post Crisis Check-List Summary: Criteria: C. 11.1

Accounting requirements - exact specification: C. II. 1

German accounting requirements on risk disclosure are stated in GAS 5. It is assumed that all listed insurance companies can comply with these requirements by voluntarily disclosing such information. Interdependencies between the financial crisis and the insurance business model must be explained when an appropriate assessment by investors should be possible.

It is recommended that inter-dependencies between individual risks are described: it is mandatory to do so, where an appropriate assessment of the risk is otherwise not possible (GAS 5 par. 25)

Results of the analysis:

Name General Specific information Quantification of No information Results information about about risk about risk risk interdependencies interdependencies interdependencies interdependencies and how they are effects or

measured or diversification recognized in the effects company

I Old Mutual No no No Yes -1 2 Standard No No No Yes -1

Life 3 Legal & Yes No No no 1

General 4 CNP No No No Yes -1

Assurances 5 Munich Re No yes No No 3 6 ING Yes No No No I 7 Hannover No No No Yes -1

Re 8 Generali No No No Yes -1 9 AXA No Yes No No 3 10 Aviva Yes No No No 1 11 Allianz No Yes No No 3 12 AEGON No Yes No No 3 13 Zurich no No No Yes -1

Summary: Six out of thirteen listed insurance companies disclose no information about future risk interdependencies in their company. Three companies disclose general information about risk interdependencies. Four listed insurance companies specify information about risk interdependencies in the future. No company quantifies risk interdependency effects or diversification effects.

Quantification: No information provided leads to a negative judgement (-1 points). General information about risk interdependencies leads to a positive judgement (1 point). Specific information disclosed about interdependencies and how they are measured or recognized in the company can be considered as specific qualitative information (3 points). The quantification of risk interdependency effects or diversification effects can be considered quantitative information (5 points).

355

356

Appendix C: Scoring documents for measuring risk disclosure performance

Post Crisis Check-List Summary: Criteria: C. II. 2.

Accounting requirements - exact specification: C. II. 2.

For each type of risk arising from financial instruments, an entity shall disclose concentration of risk (IFRS 7 par. 34c). If the quantitative data disclosed as at the reporting date are unrepresentative of entity's exposure to risk during the period, an entity shall provide further information that is representative (IFRS 7 par. 35).

The requirements can be applied to financial instruments that are named in the financial crisis situation.

Results of the analysis:

# Name General Specific and Quantified data on No information Results information representative risk concentration disclosed about risk information about risk concentration concentration arising arising from from financial financial instruments named in instruments the financial crisis AND that are negative statement that named in the no risk concentration financial crisis exists

1 Old Mutual No No No Yes -1 2 Standard Life No No No Yes -1 3 Legal & No No No Yes -1 General 4 CNP No No No Yes -1 Assurances 5 Munich Re No Yes No No 3 6 ING No Yes No No 3 7 Hannover Re No No Yes No 5 8 Generali No No No Yes -1 9 AXA No Yes No No 3 10 Aviva No No No Yes -1 11 Allianz No No Yes No 5 12 AEGON No No Yes No 5 13 Zurich No No Yes No 5

Summary: Six out of thirteen listed insurance companies disclose no information about risk concentration. Three companies disclose specific information about risk concentrations. Four listed insurance companies provide quantitative information about risk concentration arising from financial instruments named in the financial crisis.

Quantification: No information provided leads to a negative judgement (-1 points). General information about risk concentration from financial instruments named in the financial crisis leads to a positive judgement (1 point). The disclosure of specific and representative information about risk concentration or a negative statement that no risk concentration exists is measured with three points. The disclosure of quantitative data on risk concentration can be considered quantitative information (5 points).

357

Appendix C: Scoring documents for measuring risk disclosure performance

Post Crisis Check-List Summary: Criteria: C. II. 3.

Accounting requirements - exact specification: C. 11.3.

German accounting requirements on risk disclosure are stated in GAS 5. It is assumed that all listed insurance companies can comply with these requirements by voluntarily disclosing such information.

In the case of risk threatening the existence of an enterprise, the forecast period is generally one year (GAS 5 par. 24).

Results of the analysis:

# Name General information about risk threatening the existence of the company

Specific information about risk threatening the existence of the company OR Negative statement

No information disclosed

Results

1 Old Mutual No No Yes -1 2 Standard Life No No Yes -1 3 Legal &

General No No Yes -1

4 CNP Assurances

No No Yes -1

5 Munich Re No Yes No 3 6 ING No No Yes -1 7 Hannover Re No Yes (negative statement) No 3 8 Generali No No Yes -1 9 AXA No Yes No 3 10 Aviva No No Yes -1 11 Allianz No No Yes -1 12 AEGON No No Yes -1 13 Zurich No Yes (negative statement) No 3

Summary: Nine out of thirteen listed insurance companies disclose no information about risk threatening the existence of the company. Two companies disclose specific information and two companies make a negative statement that no risk that threatens the existence of the company exists.

Quantification: No information provided leads to a negative judgement (-1 points). General information about risk threatening the existence of the company leads to a positive judgement (1 point). The disclosure of specific information about risk threatening the existence of the company or a negative statement that no such risk exists is measured with three points. A quantification of the risk is not possible.

358

Appendix D: Formulas, Variables, Results and data of the event study method

Variables: n Number of, nwance groups in tM research study (13) R, Me rate of reban on V. share pnca of an msunnce group i on day t R. the rats of reorn on 0. insurance marketbenirmark Indes BEIN SUR of Bloomberg-n dayt

a tM ntarupt tann of IM OLS market model regression of an Insurance group I b Or. systemabc nsk analysed In me OLS market model regression of stock I I summation veruble coefcient S. Standard error of on OLS market model repression of an insurance group I Ssquare residual nnana on ft market pcnf0)U rikulalsd dunnp the estmabon period T Number of days in me ssumaoon period (88) It started vnth BD and after two confounding events was reduced to 88 K Number of days ,n the event study penod(15) later mudded to(14)and supormod, 5ad to (13) ea confounding events Ba Mean rerun, of the knurance group I over the esbmabon period S. Mean return of" insurance market benc)snark index calculated over the osumaton Period

Formulas: AR, Abnormal R. Wm ARa"Ra-(a Db; Rr) SD, Standard Denabun SD . "(5; ((R00T(1a(1ff)a(SQUARE((R�r$a), 2)/5UM(50UARE((R�F$), 2))

SAR, Standardaed Abnormal Return SARi-AR J5D s CAR Cundabw Abnormal Retm for e. N firm CAR"ROOT(1M)'SUM(SARr) ACAR Average stndardued cumulaow abnormal returns ACAR-Iin*IIROOT((T. 2Y(T-4))'SUM(CAR) Z last shbsbo Z"ACAR1ROOT n)

Assumptions: Markets are effluent The event was unanboapated The value of CAR an independent and Idenbcaly dnhrbuted There were no cunfoundvp events dunnp the went window

Confounding ewnta: 7. active merger and aquiuncn process 2 duldend snnounumente 3 urusual stock marbt events 4 business spe a 6c events that rowed--I stock market reactions

Resulting research restrlcdons: To 1. Checked that ZEW-ZEPHYR M&A Index of world vnde M&A scOwbes has been on kt bweat level for but yeah m September 2008 nfun the ewnt took pi cs To 2 Checked by corporate webwdss To 3 September 14th has been a inple retch day with unusuaI thing activities, eliminated from two raaearch by supemwdR d oonables To 4: September 80, was influenced by bode in Chew and Human season in the US while now contn<ts were M1apobated in Manta Cano by the biggest nmeuranu compares, lead to moddod tomb..

359

Appendix D: Formulas, Variables, Results and data of the event study method

Ormral Ra*JH

z -7,5 "dm", Zmcd . 22,5 a" modNed Zs4"rmOd 39,5 Th. . -WW. b--. 1 nWrn YSipn1cant sot5t the near oKONanq of A1G MOa sgn5ant Ynpod on tho Va6as of the 1010 rth. d .. wams proga

AFAR ß, 1 modi5. d ACARmo6 4,3 auperma5i d ACARo pannoo 40.1

R"u is per Nauancs Gro p

kauarm Group All.. A%A Gsmra4 ANva ING Aegon CNP Lepel5Gemral Zuntl1 MunldoR. Old MI" StaroaMllfs MamcwtR.

CAR -116 6,1 . 0,6 4,6 -0,7 8,4 1,4 1.7 2,4 -0] 4,3 1,7 -0,6 mod6I. d CAR -16, ] 4,4 -2,4 1 "14,2 -14,1 -1,2 -1,7 -0,1 J -11.7 -1,1 -11,5 u . 0006.04 CAR -19 S, 7 -1,5 -6,6 "16,5 -19 0.2 "9,9 -2.5 ß, 4 -16,2 . 9,6 -13,1

Appendix D: ding d. y, - .

AIIlwnz

Formulas p rice ' ,

ýös. oe Data t . he In% B EINSUR Rat- s uared deviation h

1/ariables -ý-100 25.04.2008 129,9 0,0 0,0 q

0.0 0,0 1.0 0 O . re . In V

3 78 olume is

1 06 P)

r -9 28.04.2008 131.1 0,9 0,8 0,4 1.0 1.0

, 0.4 ,

2 12 . 4

Results and -98 29.04.2008 129.5 "1,2 -1,4 -0,2 1,4 1.0. -0,2

, 3 25

44 4

12 of the W -97 -=-98

30.04.2008

02 05 2008 130,3 0,6 1,5 -0,2 2,8 1.0 -0,2

. 2.75

46 41

G1C1 . . 133.3 2,2 2.6 0.6 7.8 1.0 0,6 3,93 43

event stud -84

05.05.2008 06 5

133,5 0,2 . 0,3 0,5 0,0 1,0 0,5 1,30 43 y

' .0 . 2008 131,6 -1,5 "1,7 "0,3 2,3 1,0 -0,3 3,28 47

method -93 - - -92

07.05.2008 131.8 0,2 0,2 0,1 0,1 1,0 0,1 2,26 46 08.05.2008 130,2 -1,2 -0,6 "0,8 0,2 1,0 "0,7 2,79 50

91 09.05.2008 128,6 -1.2 "1,1 -0,4 0,9 1,0 -0,4 2,65 51

., r -90 12.05.2008 129,3 0,5 0,0 0.6 0.0 1.0 0,6 0,95 50

-89 13 05.2008 129,4 0.1 "0,2 0,3 0,0 1.0 0,3 2,19 48 1-88 14.05.2008 129,2 . 0,2 0,2 -0,2 0,2 1.0 -0,2 2.06 47

"`-87 15.05.2008 128,5 -0.5 "0,1 "0,4 0,0 1,0 -0,4 2.37 46 -86 16 05.2008 329,5 0,8 0,0 0,8 0,0 1,0 0,8 3,17 44 -85 19.05.2008 129,4 -0,1 "0,2 0,1 0,0 1,0 0,1 3,10 42

^"-84 20.05.2008 129,2 Al -1,5 0,9 1,7 1.0 0.9 3,89 44 ---83 21.05.2008 126,0 "2.5 -2,2 . 1,0 4,1 1,0 -1,0 G, 04 46

-82 22.05.2008 119,3 -5,6 . 1.1 -4,9 0.8 1,0 -4,6 5.19 67 - "-81 23.05.2008 119,7 0,3 -0,9 I'D 0,5 1,0 1.0 3,43 67

- -80 26 05.2008 118,7 -0,8 -0,3 -0,5 0,0 1.0 -0,5 1.22 67 -79 27.05.2008 117.7 . 0,9 -0,2 -0,6 0,0 1.0 . 0,6 3,27 69 -78 28 05.2008 120,1 2.0 1,6 0,9 3,8 1,0 0,9 3,87 59 -77 29.05.2008 110,3 0,2 0,7 -0,1 0.7 1.0 -0,1 2,01 56 -76 30.05.2008 122,0 1,4 0,0 1,5 0,0 1,0 1,4 2,52 55 -75 02.06.2008 119,8 -1,9 '0,7 -1,3 0,2 1,0 -1,3 2,44 59

'-74 03.06.2008 118,7 "0,9 0,3 "0,9 0,2 1,0 -0,9 1,60 61 -73 04.06.2008 117,6 . 0,9 "0,6 -0,4 0,2 1,0 -0.4 2,40 63 -72 05.06.2008 117,7 0,1 -0,2 0,3 0,0 1,0 0,3 2,70 65

'-71 06.062008 115.7 -1,7 "2,5 0,0 5,3 1,0 0,0 4,03 67 '--7D 09.06.2008 114,3 -1,3 -0,5 -0,8 0,1 1,0 -0,8 3,11 71 "-69 10.06.2008 113,3 "0,9 "0,5 -0,5 0,1 1,0 "0,5 3,60 72 '-68 11.06.2008 112,1 "1,1 "1,7 0,1 2,2 1,0 0,1 4,26 67 -87 12062008 112,1 0,1 1,6 -0,9 3,1 1,0 -0,8 4,87 67 -66 13 06 2008 116,2 3,5 0,9 3,0 1,3 1,0 3,0 7,34 64 -8 16.06.2008 117,6 1,2 "0,4 1,6 0,1 1,0 1,5 3,65 63 `84 17.062008 118,6 0,8 0,0 0,9 0,0 1,0 0,9 3,40 61 "63 18.06.2008 117,3 -1,1 "1,3 -0,2 1,1 1,0 -0,2 2,69 64

"-62 19.06.2008 118,3 0,9 "0,9 1,5 0,4 1,0 1,5 5,46 67 - '-61 20.06.2008 117,2 -1,0 -1,6 0,2 2,0 1,0 0,2 6,92 74

-60 23.06,2008 116,8 -0,4 -1,3 0,6 1.3 1,0 0,6 3,34 73

-

-59 -56

24.06,2008 25,06.3008

115,6 118,5

"0,8 2,3

"0,4 1,0

-0,5 1,7

0,0 1,5

1,0 1,0

-0,5 1,7

3,76 3 38

79

74 -57 26.062008 115,6 "2,5 "3,4 -0,2 10,1 1,0 -0,2

, 3,90 78

-56 27.06,2008 112,3 "2,9 -1,4 "2,0 1,4 1,0 -1,9 4,34 31 -55 30 06.2008 111,2 -1,0 "0,2 -0,8 0,0 1,0 -0,8 3,72 78 -54 01.07.2008 108,9 -2.1 "2.8 "0,3 6,6 1,0 -0,3 4,61 81 "53 02.07.2008 109,6 0,6 0,1 0,7 0,1 1,0 0,7 3.58 80

'° "52 03.07.2008 110,5 0,9 1,7 "0,2 3,7 1,0 -0,2 4,77 61 -51 04.07 2008 110,0 -0,5 -2,5 1,2 5,2 1,0 1,2 2,00 81 "50 07.07,2008 110,8 0,7 1,3 0,0 2,2 1, D 0,0 2.72 80 "49 08.07,2008 110,1 -0,7 -1,3 0,3 1,3 1,0 0,3 4,10 80 -4e 09.07.2008 112,7 2,3 2,4 0,9 6,7 1,0 019 2,80 74 -47 10.07.2008 112,3 -0,4 -1,4 0,6 1,5 1,0 0,6 3,22 75 -46 11.07,2008 108,6 "3,4 "3,0 -1,4 7,6 1,0 -1,4 4,86 72 -45 14.07,2008 109,1 05

, 1,0 0,0 1,3 3.0 0.0 2,58 69

1 -44 15 07.2008 102,6 -6 ,4 "3,0 d, 3 8,0 1,0 -4,2 9,69 73 -43 16.07.2008 99.4 -3.2 0,2 "3.2 0,1 1,0 "3,2 8,72 68 -42 17.07,2008 106,6 6,8 2,9 5,0 9,7 1,0 4,8 8,32 62 -41 16.07.2008 109,1 2,3 3,5 0,1 13,6 1,0 0,1 5,41 61

' - -40 21.07.2008 111.3 2,0 0,2 2,0 0,1 1.0 2,0 3,44 58 -'- 49 22.07.2008 110.1 -1,1 -0,2 "0,9 0,0 1,0 "0,9 2,86 59

-38 23.07.2008 115,5 4,6 3,6 2,4 14,6 1,0 2.3 5,08 55 --- -37 24,07.2008 114,8 "0,6 -1.3 0,4 1,3 1,0 0,4 3,75 55

38 25.07,2008 107,9 "6.4 -4,1 -3,7 15,4 1,0 -3,5 9,07 57 -35 28.07.2008 107,2 "0,6 . 2,0 0,8 3,4 1,0 0,7 2,51 56

- " "34 29.07.2008 105,5 "1,7 0,1 -1,7 0,1 1,0 -1,6 3,66 56

" . 33 30.07.2008 108,7 2,9 2,4 1,5 6,6 1,0 1,4 2,62 56 '-

- -32 31.07.2008 109.3 0,6 0,0 0,7 0,0 1,0 0,7 2.52 58 -31 01.08 2008 108,8 -0,4 -0,3 -0,2 0,0 1,0 -0,2 2,09 60

30 04,08,2008 108,5 -0,3 . 0,6 0,2 0,2 1,0 0,2 1,67 61 -29 05.08.2008 112,4 3,5 4,5 0,7 12,2 1,1 0,7 3,94 59 -28 06 08.2008 113,6 1,0 0,4 0,8 0,3 1,0 0,8 3,37 80

- -27 07.082008 111,1 "2,3 0,1 -2,2 0,1 1,0 -2,2 5,41 61 08 08.2008 111,4 0,3 1,0 "0,3 1,5 1,0 -0,3 3,26 51

-- " -25 11.08.2008 113,7 2,0 1,4 1,2 2,6 1,0 1,2 2,90 59 -24 32.08.2008 113.9 0,2 -0.3 0,5 0,0 1,0 0.5 2,20 59

-23 13.08,2008 110,6 -3,0 "3,4 -0,7 10,3 1,0 -0,7 3,61 59 -22 14.08.2008 110,0 . 0,6 0,5 -0,6 0,5 1,0 -0,8 2,61 57 -21 15.08,2008 110,3 0,3 0,6 0,0 0,7 1,0 0,0 2,91 57 -20 28.08.2008 109,5 -0,7 -0,6 "0,3 0,2 1,0 "0,2 1,69 57

^ -19 19.08.2008 106,0 -3,3 "4,2 "0,6 16,0 1,0 -0,5 4,26 62 '-18 20.08.2008 105,9 "0,1 0,8 -0,5 1,0 1,0 "0,5 3,18 61 -17 21.08.2008 104,3 -1,5 "1,5 "0,4 1,8 1,0 -0,4 2,80 62

`-18 22.082008 106,9 2,4 2,8 0,7 9,0 1,0 0,7 3,36 57,5 - . 15 25.08.2008 109,2 2,1 0,0 2.2 0.0 1,0 2,2 3,15 57,5

-14 26 08,2008 109,2 0,1 0,0 0,1 0,0 1,0 0,1 4,02 59,5 -13 27.08.2008 109,5 0,3 "0,9 0,9 5 0, 1,0 0,9 3,03 61 -12 28.08

. 2008 112,2 2,4 2,7 0,8 ß, 2 1,0 O, 7 5,26 61

-11 0R 008 11a q 9 22 19 O1 10 9 '1 12 fin

"1 03.09.2008 114,1 -0,3 -0,1 -0,1 0,0 1,0 "0,1 "0,1 3,31 1.06 60 -9 02.09.2008 115,8 1,5 2,7 -0,1 8,5 1,0 -0,1 "0,1 4.85 1,06 57 -8 03 09 2008 115,2 "0,5 . 1,2 0,4 1,0 - 1,0 0,3 0.3 3,17 1.06 59

7 04 09.2008 112,6 -2,3 "2,7 "0,5 6,1 1,0 -0.5 "0,5 3,48 1,06 60 - '-^-8 05.09.2008 108,2 "4,1 -2,3 "2,6 4,3 1,0 -2,5 -2,5 5,54 1,06 63 - -5 08 09.2008 113,2 4,5 0,0 4,5 0,0 1,0 4,5 4,5 6,14 1,06 56

-4 09.09.2008 114,7 1,3 5,1 -1,9 28,2 1,1 -1,8 -1,8 5,29 1,06 60

" -3 1009.20 08 112,0 -2,4 "1,3 -1,5 1,1 1,0 -1,5 -1,5 4,31 1,06 58 -2 11.09.2008 110,1 -1,7 "1,4 -0,7 1,5 1,0 -0,7 -0,7 3,71 3,06 58 - 12.091008 330,4 0,3 as 0,0 0,5 1,0 0,0 0,0 3,09 3,06 60

15.09.2008 303,3 -6,8 "6,3 "2,7 37,9 1.1 -2,5 -2,5 Man

12,32 aging

1.06 81 16.09.2008 97,7 -5,7 "3,4 "3,5 10,4 3,0 -3,3 -3,3 11.83 1,06 90

-'3 17.09.2008 97,5 -0,3 -1,0 0,4 0,6 1,0 0,4 0,4 8,15 2,06 90 18.09.2008 90.6 -7,6 "1,1 "6,8 8 0,8 1,0 -6,7 -6,7 14,40 1,06 85 19.09 2008 100.9 10,3 11,0 3: 3 225 1,2 2,7 2,7 16.33 1,06 75

243.9 Sum SARk -11.9 325,09 S ource: Bloomberg and Goldman Sachs mod. SARk -16.3

supennod. f. SARk -19,0 CARt , R mean return of BEINSUR dunng the es timation penod . 45.1

-0,1912698 T is the number o f days In the es umehon penod considenn 9 confou modified CARS

" 88 "81,1 irr

0,011364 sUpermodifl. d CARt 5 standard error of the OL5 mar ket model regression -66,6

1,00868657

square of S standard error of the OLS market model regression 1.017,408

sum of (R,,. -$� squared) 243,3 -' - -

ALianz; ARit- Rit-("0.09+0,63 " Rmt)

y 361

Appendix D: Formulas, Variables, Results anc Data of the event study method

25.04.2008 22,9 0,0 0,0 0,0 0,0 1,0 0,0 9,0 8 3,68 65 28.04.2008 23,2 1,5 0,8 1,2 1,0 1,0 1,3 9,1 1 61 29.04.2008 23,0 -1,1 -1,4 -0,3 1,4 1,0 -0,3 7,36 63 30.04.2008 23,2 1,2 1,5 0,5 2,8 1,0 0,6 7 48 02.05.2008 24,2 3,9 2,6 2,7 7,8 1,0 2.8 ,

15,62 64 62

1 05.05.2008 24,0 -0,7 -0,3 -0,4 0,0 1,0 -0,4 5,90 62 06.05.2008 23,6 -2,0 -1,7 -0,9 2,3 1,0 -1,0 9,59 66 07.05.2008 23,7 0,4 0,2 0,5 0,1 1,0 0,5 9 87 08.05.2008 23,6 -0,4 -0,6 0,0 0,2 1,0 0,0

, 10,75

66 70 09.05.2008 22,9 -2.8 -1,1 -2,1 0,9 1.0 -2.2 10,76 73 12.05.2008 23,1 0,9 0,0 1,1 0,0 1,0 1.1 4,59 72 13.05.2008 23,3 0,8 "0,2 1,0 0,0 1,0 1,1 10,02 70 14.05.2008 23,3 0,1 0,2 0,1 0,2 1,0 0,1 9,42 71 2.5.0s

. 200s 23,2 -0,6 -0,1 -0,4 0,0 1,0 -0,5 8 18 73 16.05.2008 23,3 0,5 0,0 0,6 0,0 1,0 0,7 , 10 13

19.05.2008 23,1 -1,1 -0,2 "0,9 0.0 1,0 -0,9 ,

9 16 70

20.05.2008 22,8 -1,1 -1,5 "0,2 1,7 1,0 -0,2 .

10.37 0

71 21.05.2008 22,3 -2,2 -2.2 . 0,9 4,1 1,0 -1,0 14,86 75 22.05.2008 21,9 -3,9 -1,1 . 1,3 0.8 1,0 -1,3 13,26 84 23.05200 21,7 -0,8 "0,9 -0,2 0.5 1,0 -0,2 9,83 92 26.05.2008 21.5 -1,1 -0,3 -0,8 0,0 1,0 -0,8 4,06 92 27.05.2008 21.3 .0 ,6 -0,2 -0,4 0,0 1,0 -0,4 9,31 102 28.05.2008 22.0 3,1 1,8 2,3 3,8 1,0 2,4 14,51 94 29.05.2008 22,2 1,0 0,7 0,8 0,7 1,0 0,8 9,10 91 30.05.2008 22,3 0,5 0,0 0,6 0,0 1,0 0,6 30,71 89 02.06.2008 21,6 -3,3 -0,7 -2,9 0,2 1,0 -3,0 11,41 92 03.06.2008 21,3 -1,6 0,3 -1,6 0,2 1.0 -1,6 13,25 94 04.06.2008 21,0 -1,1 -0,6 -0,6 0,2 1,0 -0,7 10.72 100 05.06.2 008 21,3 1,0 -0,2 1,3 0,0 1,0 1,3 8,53 100 06.06.2008 20,9 -3,9 "2,5 . 0,4 5,3 1,0 -0,5 13,32 102 09.06.2008 20,3 -2,6 -0,5 -2,2 0,1 1,0 -2,3 10,22 105 10.06.2008 20,2 "0.6 "0,5 -0,2 0,1 1,0 -0,2 16,74 105 11.06.2008 20,4 0,6 -1,7 1,6 2,2 1,0 1,6 13,68 102 12.06.2008 20.5 0,5 1.6 -0,2 3,1 1,0 -0,2 13,84 100 13.06.2008 20,7 1,1 0,9 0,8 1,3 1,0 0,8 11,59 97 16.06.2008 21,1 1,9 -0.4 2,2 0.1 1,0 2,4 11,15 96 17.06.2008 21,2 0.6 0,0 0,7 0,0 1,0 0,7 9,02 92 18.06.2008 20,6 -2,8 -1,3 -2,0 1,1 1,0 -2.1 10,14 96 19.06.2008 20,4 -0,9 "0.9 -0,3 0,4 1,0 -0,3 10,50 98 20.06.2008 20,2 -1,2 "1,6 -0,2 2,0 1,0 -0,2 17,63 105 23.06.2008 19,8 -2,0 -1,3 -1,2 1,3 1,0 -1.2 9,95 104 24 06.2008 19,5 -1,5 "0,4 -1,2 0.0 1,0 -1,2 13,04 113 25.06.2008 19.9 2,0 1,0 1,6 1,5 1,0 1,7 8,75 107 26.06.2008 19.3 -3,3 -3,4 . 1,4 10,1 1,0 -1,4 14.59 112 27.06.2008 18,4 -5,1 -1,4 -4,2 1,4 1.0 -4,4 25,57 115 30.06.2008 18,3 -O, 1 -0,2 0,2 0,0 1.0 0,2 12,86 112 01.07.2008 17,7 -3,5 -2,8 -1,9 6,8 1,0 -2,0 17,70 115 02.07.2008 17,8 0,5 0,1 0,6 0,1 1,0 0,6 13,68 116 03.07.2008 17,9 0,8 1,7 0,0 3,7 1,0 0,0 19,98 115 04.07.2008 17,9 -0,2 -2,5 1,2 5,2 1,0 1,3 8,50 115 07.07.2008 18,0 0,4 1,3 -0,1 2,2 1,0 -0,1 9,67 11G 08.07.2008 17,6 -2,3 -1.3 -1,5 1,3 1,0 -1,5 36,58 117 09.07.2008 18.5 4,7 2,4 3,6 6.7 1,0 3,7 15,24 112 10.07.2008 18,3 -0,9 -1,4 0,0 1,5 1,0 0,0 13,92 117 11.07.2008 17,6 -4,2 -3,0 -2,5 7,6 1,0 -2,6 15,12 122 14.07.2008 17,7 0,9 1,0 0,5 1,3 1,0 0,5 11,19 120 15.07.2008 16,8 -5,1 -3,0 . 3,4 8,0 1,0 . 3,5 22,50 124 16.07.2008 16,4 -2,8 0,2 -2,7 0.1 1,0 -2,9 21,94 122 17.07.2008 17,8 7,8 2.9 6,4 9,7 1,0 6,6 22,04 111 18.07.2008 18,3 2.9 3,5 1,2 13,6 1,0 1,2 19,49 110 21.07.2008 18.7 1,9 0,2 2,0 0,1 1,0 2,1 11,94 107 22.07.2008 18,4 -1,6 -0,2 -1,4 0,0 1.0 -1,4 11.53 108 23.07.2008 19,5 5,9 3,6 4,1 14,6 1,0 4,2 18,84 102 24.07.2008 19,6 0,5 -1,3 1,3 1,3 1,0 1,4 13,20 100 25.07.2008 18,3 -7,2 -4,1 -4,9 15.4 1,0 -5,0 23,87 101 28.07.2008 18,0 "1,8 -2,0 -0,6 3,4 1,0 -0,7 11,42 100 29.07.2008 17,4 -3,1 0.1 -3,0 D. 1 1.0 -3,2 16,45 100 30.07.2008 18.3 4,7 2,4 3.6 6,6 1,0 3,7 14,33 99 31.07.2008 18,5 1,1 0, O 1,2 0,0 1,0 1,3 11,72 102 01.08.2008 18,4 -0,4 .0 ,3 -0.1 0,0 1,0 -0,1 10,83 305 04.08.2008 18,3 . 0,4 -0,6 0,0 0,2 1,0 0,0 6,80 107 05.08.2008 19,2 4,7 4,5 2,5 22,2 1,0 2,5 15,55 104 06.08.2008 19,8 2,6 0,4 2,5 0,3 1,0 2,7 13,92 103 07.08.2008 21.0 5,8 0,1 5,9 0,1 1,0 6,1 25,67 104 08.08.2008 21,5 2,2 1,0 1,8 1,5 1,0 1,9 20,86 104 11.08.2008 21,8 1,6 1,4 1,0 2,6 1,0 1,0 11,61 101 12.08.2008 21,7 -0,5 -0,3 -0,2 0,0 1,0 -0,2 13,92 99 13.08.2008 20,9 "3,8 -3,4 -1,9 10,3 1,0 -1,9 -

13,91 99 14.08.2008 20,7 -0.9 0,5 -1,0 0,5 3,0 -1,0 9,59 97 15.08.2008 21,3 2,8 0,6 2.6 0,7 1,0 2,7 10,31 97 18.08.2008 21,2 . 0,6 -0,6 -0,2 0,2 1,0 -0,2 7,61 97 19.08.2008 20,2 .5 '1 -4,2 -2,8 16,0 1,0 - -2,8 14,41 101 20.08.2008 39,9 -1,1 0,8 -1,4 1,0 1,0 -1,5 10,59 100 21.08.2008 19,7 -1,1 -1.5 -0,2 1,8 1.0 -0,2 10,20 302 22,08.2008 20,3 3,0 2,8 1,7 9,0 1,0 1.7 11,12 96.5 25.08.2006 20,5 0,9 0.0 1,1 0.0 1,0 1,1 6,37 96,5 26.08.2008 20,2 -1,7 0,0 .1 ,6 0.0 1,0 -1,6 9,33 98,5 27.08.2008 20,0 -0,9 -0,9 "0,3 0,5 - 1,0 -0,3 9.15 102 28.08.2008 20,7 3,3 2.7 2.0 8,2 1.0 2.1 14,91 300 29.08.2008 21,3 2,9 0.2 3,0 0,1 1,0 3,1 97 1 99 01.09.2008 21,2 -0,7 -0,1 "0,6 0.0 1,0 -O, 6 -0.6 09 .: 3,08 98 02.09.2008 22.3 5,1 2,7 3,8 8,5 1,0 3,9 3,9 24.70 3,68 94 03.09.2008 22,3 0.1 -1,2 0,9 1,0 1,0 0,9 0,9 13,15 3,68 95 04.09.2008 21,8 -2,4 "2,7 -0,8 6,1 1,0 -0,9 -0,9 14,89 3,68 96 05.09.2008 20,9 -4,5 "2,3 "3,2 4,3 1,0 -3,3 -3,3 18,63 3,68 99 08.09.2008 23,1 9,8 0,0 10,0 0,0 1,0 10,4 10,4 38,29 3,68 91 09.09.2008 23,0 -0 5,1 -3,0 28.2 1,0 -3,0 -3,0 19.73 3,68 94 10.09.2008 22,3 -3,4 -1,3 "2,6 1,1 1,0 "2,7 -2.7 15.64 3,68 93 11.09.2008 21,4 -4,1 -1,4 "3,3 1.5 1,0 -3,4 -3,4 20,91 3,68 93 12.09.2008 21,6 0,9 0,5 O8 05 1,0 0,8 0,8 12.93 3,68 99

15.09.2008 19,7 -9,6 "6,3 "6,1 37,9 1,0 -6,0 -6,0 40,30 3.68 127 16.09.2008 19,1 -3,2 -3,4 -1,3 10,4 1,0 -1,3 "1,3 34,04 3,68 145 17.09.2008 20.0 4 10 5,1 0.6 1,0 1 5,3 5,3 34,51 3. G8 145 18.09.2008 19,9 -0,2 -1,1 0,5 0,8 1.0 0,6 0,6 27,26 3,08 130 19.09.2008 23.3 14,3 11.0 8,7 125.4 1,2 7.4 7,4 56,92 3.68 120

243,9 Sum SARk 8,1 1130,00 " mod. SARI -2,4 Without 5.9.2008

eupermodil SARK -9.7 without 8.9.2008 and 19,9.2008 - CARt

Rm ean return of BEINSUR during the estimation period 31.3 -0,1912698

T Is the number of days In the estimation period modified CARt 88 -8,8

irr 0.011364 - supe

S standard error of the OLS3r(3&t model regression 5 1Ifiad CARL

0,949595 square of S standard error of the OLS market model regression

0,9017312 sum of (Rmt-Rm squared)

243.3

Appendix D: Formulas, Variables, Results and Data of the event study method

24,04.2008 28,6 0,0 0,0 0,2 0,0 1,7 0,1 5,43 1,89 49 25.04.2008 29,0 1,2 0,8 1,2 1,0 1,7 0,7 4,29 46 28,04.2008 28,7 -1,0 -1,4 -0,7 1,4 1,7 -0,4 7,28 44 29.04.2008 28,4 . 0,9 1,5 -0,9 2,8 1,7 -0,6 5,99 46 30.04.2008 28,5 0,2 2,6 0,0 7,8 1,7 0,0 11,45 46 02.05.2008 28,8 1,2 -0,3 1,4 0,0 1,7 0,8 11,19 43 05.05.2008 28,7 -0,5 -1,7 -O, 1 2,3 1,7 0,0 4,17 43 06.05.2008 28,4 -0,9 0,2 -0,8 0,1 1,7 -0,5 7,80 47 07,05.2008 28,5 0,4 -0,6 0,6 0,2 1,7 0,4 7,03 46 08.05.2008 28.4 -0,6 -1,1 -0,3 0,9 1,7 -0,2 8,90 51 09.05,2008 28,2 -0,7 0,0 -0,6 0,0 1,7 -0,3 8,21 51 12.05.2008 28,1 -0,1 -0,2 O, 1 0,0 1,7 0,0 8,24 50 13.05.2008 28,1 0,1 0,2 0,2 0,2 1,7 0,1 14,32 48 14.05.2008 28,1 -0,1 -0,1 0,1 0,0 1,7 0,1 10,59 47 15.05,2008 28,0 -0,3 0,0 -0,2 0,0 1,7 -0,1 7,17 46 16.05.2008 28,0 0,0 -0,2 0,2 0,0 1,7 0,1 8,68 44 19.05,2008 27,3 -2,8 -1,5 -2,4 1,7 1,7 -1,4 5,58 42 20.05,2008 27,0 -1,1 -2,2 -0,7 4,1 1,7 -0,4 7,02 44 21.05.2008 26,8 -0,7 -1,1 -0,4 0,6 3,7 -0,2 8,70 48 22.05.2008 26,4 -1,4 -0,9 -1,1 0,5 1,7 -0,7 11,71 57 23.05.2008 26,4 -0,1 -0,3 0,1 0,0 1,7 0,1 8,08 65 26.05.2008 26,2 -0,7 -0,2 -0,5 0,0 1,7 -0,3 4,35 55 27.05.2008 26,1 -0,3 1,8 -0,4 3,8 1,7 -0,2 10,48 69 28.05.2008 26,5 1,4 0,7 1,5 0,7 1,7 0,9 12,59 59 29.05,2008 26,5 0,1 0,0 0,3 0,0 1,7 0,2 4,78 56 30.05.2008 26,8 0,9 -0,7 1,2 0,2 1,7 0,7 5,72 55 02.06.2008 26,4 -1,3 0,3 -1,2 0,2 1,7 -0,7 4,31 58 03.06.2008 26,6 0,6 -0,6 0,8 0,2 1,7 0,5 4,16 59 04.00,2008 26,6 0,1 -0,2 0,2 0,0 1,7 0,2 4,52 53 05.06.2008 26,6 -0,2 -2,5 0,3 5,3 1,7 0,2 3,93 65 06.06.2008 26,2 -1,3 -0,5 -1,1 0,1 1,7 -0,7 6,06 69 09.06.2008 26,1 -0,6 -0,5 -0.4 0,1 1,7 -0,3 4,26 71 10.06.2008 25,9 -0,6 -1,7 -0,2 2,2 1,7 -0,1 4,86 72 11.06.2008 26,0 0,2 1,6 0,2 3,1 1,7 0,1 5,02 67 12.06.2008 26,0 -0,1 0,9 0,0 1,3 1,7 0,0 4,47 67 13.06.2008 25,8 -0,4 -0,4 -0,2 0,3 1,7 -O, 1 4.22 64 16.06.2008 25,8 -0,3 0,0 . 0,1 0,0 1,7 -0,1 5,57 63 17.06.2008 25,4 -1,6 -1,3 -1,2 1,1 1,7 -0,7 9,00 61 18.06.2008 25,1 -1,3 -a, 9 -1,0 0,4 1.7 -0,6 6,78 64 19.06.2008 25,1 0,3 -1,6 0,7 2,0 1,7 0,4 4,78 (, 7 20.06.2008 24,8 -1,2 -1,3 -0,9 1,3 1,7 -0,5 10,26 74 23.06.2008 24,7 -0.5 -0,4 -0,3 0,0 1,7 -0,2 4,81 74 24.06.2008 24,4 -1,2 1,0 -1,2 1,5 1,7 -0,7 6,67 82 25.06.2008 24,6 0,8 -3,4 1,4 10,1 1,7 0,8 4,38 78 26.06.2008 24,7 0,1 -1,4 0,4 1,4 1,7 0,3 5,71 82 27.06.2008 24,2 -1,7 -0,2 -1,5 0,0 1,7 -0,9 6,99 85 30.06.2008 24,2 -0,3 -2,8 0,2 6,8 1,7 0,1 6,56 82 01.07.2008 24,0 -0,7 O, 1 -0,5 0,1 1,7 -0,3 6,87 86 02.07,2008 24,1 0,5 1,7 0,4 3.7 1,7 0.2 4,97 86 03.07.2008 24,0 -0,5 -2,5 0,0 5,2 1,7 0,0 7,17 87 04.07.2008 23,8 -0,8 1,3 -0,8 2,2 1,7 -0,5 5,65 87 07,07.2008 23,7 -0,6 -1,3 -0,2 1,3 1,7 -0.1 5,36 87 08.07.2008 23,4 -1,2 2,4 -1,3 6,7 1,7 -0,8 7,62 87 09,07.2008 23.6 0,8 -1,4 1,1 1,5 1,7 0,7 6,25 80 10.07.2008 23,6 0,1 -3,0 0,6 7.6 1,7 0,4 6,16 80 11.07.2008 23,0 -2,6 1,0 -2,5 1,3 1,7 -1,5 10,61 78 14.07.2008 23,1 0,5 -3,0 1,0 8,0 1,7 0.6 5,08 75 15.07,2008 22,6 -2,3 0,2 -2,2 0,1 1,7 -1,3 9,52 77 16.07.2008 22,3 -1,5 2,9 -1,7 9,7 1,7 -1,0 10,57 72 17.07.2008 22,6 1,3 3,5 1,0 13,6 1,7 0,6 8,63 66 18.07.2008 22,6 0,0 0,2 0,1 0,1 1,7 0,1 7,44 65 21.07.2008 22.8 1,2 -0,2 1,4 0,0 1,7 0.8 5,78 61 22.07.2008 22,7 -0,7 3,6 -1,0 14,6 1,7 -0,6 7,72 61 23.07.2008 23,3 2,5 -1,3 2,9 1,3 1,7 1,7 6,44 55 24.07.2008 23.3 0,0 -4,1 0,7 15,4 1,7 0,4 5,05 55 25,07.2008 22,6 -2,8 -2,0 -2,4 3,4 1,7 -1,4 12,32 57 28.07.2008 22,3 -1,6 0,1 -1,5 0.1 1,7 -0,9 7,35 57 29.07.2008 22,2 -0,4 2,4 -0,5 6,6 1,7 -0,3 6,57 57 30.07.2008 22,6 1,8 0,0 2,0 0,0 1,7 1,2 6.38 57 33.07,2008 22,5 -0,3 -0,3 -0,1 0,0 1,7 -0,1 9,77 59 0208 . 2008 22,3 -1,2 -0,6 -0,9 0,2 1,7 -0,6 6,02 61 04.08,2008 22,2 -0,5 4,5 -0,9 22,2 1,7 -0,5 3,46 62 05.08,2008 22,8 2,7 0,4 2,8 0,3 1,7 1,7 7,27 GO 06.08,2008 22,9 0,7 0.1 as 0,1 1,7 0, S 5,04 59 07.08.2008 23,0 0,3 1,0 0,3 1,5 1,7 0,2 4,38 62 08.08.2008 22,8 -0,8 1,4 -0,8 2,6 1,7 -0,5 3,66 62 11.08.2008 23,2 1,6 -0,3 1,8 0,0 1,7 1,1 4,07 60 12.08.2008 23,2 0,1 -3,4 0,7 10,3 1,7 0,4 3,66 58 13.08.2008 23,1 -0,7 0,5 -0,6 0,5 1,7 -0,3 4,79 59 14.08,2008 22,9 -1,0 0.6 -0,9 0,7 1,7 -0,5 3,67 57 18.08.2008 22,8 -0,2 -0,6 0,0 0,2 1,7 0,0 3,67 57 19.08.2008 22,6 -1,2 -4,2 -0,4 16,0 1,7 -0,3 4,81 62 20.08,2008 22.4 -0,6 0,8 -0,5 1,0 1,7 -0,3 3,57 61 21.08.2008 22,2 -1,1 -1,5 -0,7 1,8 1,7 -0,4 4,34 62 22.08.2008 22,4 0,7 2,8 0,5 9,0 1,7 0,3 3,75 58,5 25.08.2008 22,4 0,0 0,0 0,2 0,0 1,7 0,1 1,86 58,5 26.08.2008 22,2 -0,6 0,0 -0,5 0,0 1,7 -0,3 3,89 60,5 27.08.2008 22,2 -0,1 -0,9 0,2 0,5 1.7 0,1 3,45 62 28.08 22,6 1,7 2.7 1,5 8,2 1,7 0,9 5,33 62 29.08 2008 22, $ 1, O 0.2 1,1 O, 1 1,7 0,7 37 15,1 01.09.2008 22,7 -0,4 -0.1 -0,2 0,0 1,7 -0,2 -0,2 3,31 1,89 G1 02.09.2008 23,1 1,7 2,7 1,5 8,5 1,7 0,9 0,9 6,79 1,89 58 03.09.2008 23,1 0,1 -1,2 0,4 1,0 1,7 0,3 0,3 3,33 1,89 60 04.09.2008 22,9 -0,9 -2,7 -0,4 6,1 1,7 -0,2 -0,2 5,03 1,89 61 05.09.2008 22,2 -3,2 -2,3 -2,7 4,3 1,7 -1,6 -1,6 8,36 1,89 64 08.09.2008 22,8 2,4 0,0 2,6 0,0 1,7 1,6 1,6 9,30 1,89 57 09.09.2008 22,9 0,5 5,1 0,0 28,2 1,8 0,0 0,0 6.30 1.89 61 10.09.2008 22,7 -0,6 -1,3 -0,3 1,1 1,7 -0,2 -0,2 5,61 1.89 59 11.09.2008 22,6 -0,6 -1,4 -0,2 1,5 1,7 -0,1 -0.1 5.32 1.89 59 12 09.2008 22,5 ý1 0,5 010 0,5 1,7 0,0 010 5,20 1,89 61

15.09.2008 21,7 -4.1 -6,3 -3,1 37,9 11 1,8 11 -1,7 -1,7 1 15,15 1,89 85 16.09.2008 21,2 -2,3 -3,4 -1,7 10,4 1,7 -1.0 -1,0 14,22 1,89 90 17.09.2008 21,4 0.8 -1.0 1,1 0,6 1,7 0,7 0,7 8,25 1,89 95 18.09.2008 20,9 -2,4 -1,1 -2,1 0,8 1,7 -1,3 -113 12,77 1,89 85 19.09.2008 22,1 5,6 11.0 4,3 125,4 2,0 2,1 2,1 24,98 1,89 75

243,9 Surn SARI, -0,8 580,60 mod. SARI, -2,4 Without 5.9.2008

supermodit SARI, -4,5 without 8.9.2008 an d 19,9,2008 CARt

R mean return of BEINSUR during the estimation period -3,3 -0,19127

T Is the number of days In the estimation penod modified CARt 88 -9.0 irr

0,011364 S standard error of the OLS markr3t el regression

sup. rmodl9. d -16.3

CARL

1,649947 square of S standard error of the OLS market model regression

2,7223 sum of (Rmt-Rm squared)

243,3

Appendix D: Formulas, Variables, Results and Data of the event study method

Aviva _. ý . ý_.

24.04.2008 25.04.2008 28.04.2008 29.04.2008 30.04.2008 01.05.2008 02.05.2008 06.05.2008 07.05.2008 08.05.2008 05>. 05 . 2008 12.05.2008 13.05.2008 14.05.2008 15.05.2008 16.05.2008 19.05.2008 20.05.2008 21.05.2008 22.05.2008 23,05.2008 27.05.2008 28,05.2008 29.05.2008 30.05.2008 02.06.2008 03.06.2008 04.06.2008 05.06.2008 06.06.2008 09.06.2008 10.06.2008 11.06.2008 12.06.2008 13.06.2008 16.06.2008 17.06.2008 18.06.2008 19.06 20.06.2008 23.06.2008 24.06.2008 25.06.2008 26.06.2008 27.06.2008 30.06.2008 01.07.2008 02.07.2008 03.07.2008 04.07.2008 07.07.2008 08.07.2008 09.07.2008 10.07.2008 11.07.2008 14.07.2008 15.07.2008 16.07.2008 17.07.2008 18.07.2008 21.07.2008 22.07.2008 23.07.2008 24.07.2008 25.07.2008 28.07.2008 29.07.2008 30.07.2008 3 1.07.2008 01.08.2008 04.08.2008 05.08.2008 06.08.2008 07.08.2008 08.08.2008 11.08.2008 12.08.2008 13.08.2008 14.08.2008 15.08.2008 18.08.2008 19.08.2008 20.08.2008 21.08.2008 22.08.2008 26.08.2008 27.08.2008 28.08.2008

02.09.2008 03.09.2008 04.09.2008 05 . 09 . 2008 08.09.2008 09.09.2008 10.09.2008 11.09.2008

15.09.2008 26.0

9.2008 2617'C'9 200

18.09.2008 19.09.2008

635,4 0,0 0,0 0,0 0,0 1,7 0,0 15,40 3,06 623,3 -2,0 0,8 -1,8 1,0 1.7 -1,1 11,86 623,9 0,1 -1,4 0,3 1,4 1,7 0,2 8,07 635,7 1,9 1,5 2,0 2,8 1,7 1,2 4,71 623,7 -1,9 2,6 -1,8 7.8 1,7 -1,1 7,42 628,3 0,7 -0,3 0,9 0,0 1,7 0,6 7,59 628,4 0,0 -1,7 0,2 2,3 1,7 0.1 2,55 656,6 4,3 0,2 4,5 0,3 1,7 2,7 12,97 653,1 -0,5 -0,6 -0,3 0,2 1,7 -0,2 7,31 649,4 -0,6 -1,1 -0,4 0,9 1,7 -0,2 8.29 642,6 -1,1 0,0 -0,9 0,0 1,7 -0,5 7,97 635,3 -1,1 -0.2 -1,0 0,0 1.7 -0,6 5,60 637,9 0,4 0,2 0.6 0,2 1,7 0,3 4,16 634,8 -0,5 -0,1 -0,3 0,0 1,7 -0,2 7,49 643,8 1,4 0,0 1,6 0,0 1.7 1.0 9,50 634,1 -1,5 -0,2 -1,3 0.0 1,7 . 0,8 6,25 647,8 2,1 -1,5 2,3 1,7 1,7 1,4 10,70 646,5 -0,2 -2,2 0,0 4,1 1,7 O, O 4,81 633,3 -2,1 -1,1 . 1,9 0,8 1,7 -1,1 10,82 610,7 -3,7 -0,9 -3,5 0,5 1,7 -2,1 11,35 605,2 -0,9 -0,3 -0,7 0,0 1,7 -0,4 9,80 616,5 1,8 -0,2 2,0 0,0 1,7 1,2 9,08 608,7 -1,3 1,8 -3,2 3,8 1,7 -0,7 7,71 621,9 2,1 0,7 2,3 0,7 1,7 1,4 9,38 626,8 0,8 0,0 1,0 0,0 1,7 0,6 8,43 636,9 1,6 -0,7 1,8 0,2 1,7 1,1 9,86 620,9 -2,6 0,3 -2,4 0,2 1,7 -1,4 7,41 623,1 0,4 -0.6 0,6 0,2 1,7 0,3 5.87 619,9 -0,5 -0,2 -0,3 0,0 1,7 -0,2 7,67 618,0 -0,3 -2,5 0,0 5,3 1,7 O, 0 8,00 615,1 -0,5 -0,5 -0,3 0,1 1,7 -0,2 11,62 601,0 -2,3 -0,5 -2,1 011 1,7 -1,3 7,80 588,1 -2,2 -1,7 -2,0 2,2 1,7 -1,2 8,01 577,0 -1,9 1,6 -1,8 3,1 1,7 -1,1 13,97 587,4 1,8 0,9 1,9 1,3 1,7 1,2 9,32 585,5 -0,3 -0.4 -0,1 0,1 1,7 -0,1 7,33 588,2 0,5 0,0 0,6 0,0 1,7 0.4 7,82 590,2 0,3 -1,3 0,5 1,1 1,7 0,3 8.91 574,9 -2,7 -0,9 -2,4 0,4 1,7 -1,5 8.09 566,4 -1,5 -1,6 -1,3 2,0 1,7 -0.8 10,10 555,7 -1,9 -1,3 . 1,7 3,3 1,7 -1,0 11,89 545,1 -1,9 -0.4 -1,8 0,0 1,7 -1,1 9,98 533,2 -2,2 1,0 -2,1 1,5 1,7 -1,2 10,22 530,7 -0,5 -3,4 -0.2 10,1 1,7 -0.3 7, G3 515,0 -3,0 -1,4 -2,8 1,4 1,7 -1,7 9, G3 503,9 -2,2 -0,2 -2,0 0,0 1,7 -1,2 11,84 498,8 -3,0 -2,8 -0,8 6,8 1,7 -0.5 13,26 482,0 -3,5 0,1 -3,3 0,1 1,7 -2,0 19,55 492.9 2,2 1,7 2,3 3.7 1,7 1,4 16,01 495,9 0,6 -2,5 0,8 5,2 1,7 0,5 16,96 489,5 -1,3 1,3 -1,2 2,2 1,7 -0,7 8,23 492,0 0,5 -1,3 0,7 1,3 1,7 0,4 10,05 491,3 -0,2 2,4 0,0 6,7 1,7 0,0 10,90 516,1 4,8 -1,4 5,0 1,5 1,7 3,0 11,52 503,1 -2,6 -3,0 -2,3 7,6 1,7 -3,4 10,75 494,6 -1,7 1,0 -1,6 1,3 1,7 -0,9 12,14 498,9 0,9 -3,0 1,1 8,0 1,7 0,7 9,01 478,5 -4,3 0,2 -4,1 0,1 1,7 -2,5 12.06 4757 -0,6 2,9 -0,5 9,7 1,7 -0,3 17,09 504,1 5,6 3,5 5,7 13,6 1,7 3,3 15,47 525,4 4,0 0,2 4,2 0,1 1,7 2,5 34,15 S21 ,4 -0,8 -0,2 -0.6 0.0 1,7 -0,3 16,00 504,9 -3,3 3,6 -3,2 14,6 1,7 -1,9 10,82 529,4 4,6 -1,3 4,8 1,3 1,7 2,9 11,57 521,7 -1,5 -4,1 -1,2 15,4 1,7 -0,7 9,89 486,7 -7,2 -2,0 -6,9 3,4 1,7 -4,1 20.91 475,5 -2,4 0,1 -2,2 0.1 1,7 -1.3 9,98 468,7 -1.4 2,4 -1,3 6,6 1,7 -0,8 13,65

500,2 6,3 0,0 6,5 O, 0 1,7 3,9 20,98 500.9 0,1 -0,3 0,3 0,0 1,7 0,2 15,67

502,6 0,3 -0,6 0,5 0,2 1,7 0,3 8,19 494,8 -1,6 4,5 -1,5 22,2 1,7 -0,9 8.96 522,4 5,3 0,4 5,4 0,3 1,7 3,3 14,36 517,0 .1 ,0 0,1 -0,9 0,1 1.7 -0,5 14,66 517,6 0,1 1,0 0,3 1,5 1,7 0,2 11,87 526,1 1,6 1,4 1,7 2,6 1,7 1,0 12,40 541,2 2,8 -0,3 3,0 0,0 1.7 1,8 9,67 548,2 1,3 -3.4 1,6 10,3 1,7 0,9 9,32 526,9 -4,0 0,5 -3,9 0,5 1,7 -2,3 16,49 526,9 O, 0 0,6 0,2 0,7 1,7 0,1 7,85 533,5 1,2 -0,6 1,4 0,2 1,7 0,9 9,12 526,1 -1,4 -4,2 -1,1 16,0 1.7 -0,6 4,67 498,6 -5,5 0,8 -5,4 1,0 1,7 -3,2 10,59 497,0 -0,3 -1,5 -0,1 1,8 1,7 0,0 8.99 483,1 -2,9 2,8 -2,8 9,0 1,7 -1,6 12,98 499,1 3,2 0,0 3,4 0.0 1,7 2,0 9,28 498,5 -O, 1 O, 0 0,1 0,0 1.7 0,0 G. 85 499,5 0,2 -0,9 0,4 0,5 1,7 0.2 6,38 516,0 3,2 2.7 3,3 8,2 1,7 1,9 6.03 5108 -0,2 0.2 0.0 0.1 1,7 0,0 18,04 520,4 1,1 -O, 1 1,3 0,0 1,7 0.8 0,8 6,24 3,06 538,8 3,4 2,7 3,5 as 1,7 2,1 2,1 8,56 3,06 531,3 -1,4 -1,2 . 1,2 1,0 1,7 -0,7 -0,7 11,93 3,06 525,7 -2,7 -O, 8 6,1 1,7 -0,5 -0,5 10,95 3,06 501,6 -4,8 -2,3 -4,6 4,3 1,7 -2,7 -2,7 12,25 3,06 532,0 5,7 0,0 5.9 0,0 1,7 3,5 3.5 6.63 3,06 547,4 2,8 5,1 2,6 28,2 1,8 3,6 1.6 13,64 3,05 538,4 -1,7 -1,3 -1,5 1,1 1,7 -0,9 -0,9 11,01 3,066 528,4 -1.9 -1.4 -1.7 1,5 1,7 -1.0 -1.0 7.22 3,06 532 3 0,7 0,5 0,9 0,5 1,7 0,5 0,5 11,25 3 06

494,4 -7,7 -6,3 -7,3 37,9 1,8 -4,1 -4,1 21,11 3,05 463,1 -6,7 -3,4 "6,5 10,4 1,7 -3,8 -3,8 25,60 3,066 469,9 1,4 -1,0 1,6 0,6 1,7 1,0 1,0 18.90 3,06 478,0 1,7 -1,2 1,9 0,8 1,7 1,1 3,1 28,63 3,06 567,0 15,7 11,0 35,5 125,4 2.0 7,6 7.66 36,24 3,06

243.9 Sum SARK 4,6 937,53 mod. SARI 1.0 With out 5.9.2008

supermodit SARk -0,0 without 8.9.2008 an d 19.0.2008 CARt

R mean return of BEINSUR during the estimation period 17,8 -0,19127

T Is the number of days In the estimation period modl6. d CARt 88 3,9

1 /T 0.011364

S standard error of the OLS markeG64 re. ressIon a6p. rmodl6. d

-23 7 CARt

1 , 853157 ,

sq uare of S standard error of the OLS market model regression 2.7329

su m of (Rmt-Rm squared) 243,3

81 82 81 82

82

84 94

201 107

99 97 94 97 99

107 108 110 113 113 108 107 102 101

96 101 103 112 111 120 115 120 126 121 125 124 127 127 126 127 121 122 123 120 124 121 110

10 10

101 103 107 109 305 104 106 107 104 102 103 102 102 102 lOG 104 105

995

101,, 5 105

Appendix D: INC. Formulas, pr

share ice on c

P" hange

P- change In A

AR,, bnormal

P- -S moon t.

D,, S ndard

AR AR S "

um SAR V olumon ot M InIm 6 Y C6 Variables Date th

25.0 4.2008 e In ' % har. % Return Return dev iatin 1 haree In v olume Is fbP) , 18.4 O, 0 O, O 0,0 0,0

,1 0.0 18.02 5 62 70 Results and 28.04.2008 18,7 1,7 0,8 1,3 1,0 1,1 1,2 15,12 ,

65 29.04.2008 18,6 -0,8 -1,4 0,1 1,4 1,1 0,1 15 29 Data of the 30.04.2008 18.7 0,8 1.5 0,1 2.8 1.1 0.1 ,

13 06 67

02.05.2008 19,3 2,9 2,6 1,6 7,8 1,1 1,4 , 20 50

71

event study 05.05.2008 19,2 -0,3 -0,3 0,0 0,0 1,1 0,0 , 5 88

56

06,05.2008 19,0 -1,2 -1,7 -0,1 2,3 1.1 -0.1 ,

14,22 56 58 method 07

. 052008 19,3 1,3 0,2 1,3 0,1 1,1 1,2 13 18 59 08.05.2008 19,0 -1,2 -0,6 -0,7 0,2 1,1 -0,7 ,

16 11 09.05.2008 18,7 -2,0 -1,1 -1,3 0,9 1,1 -1,1 ,

18 15 59

12.05 2008 18,9 1,1 0,0 1,3 0,0 1,1 1,1 , 8 49

60

13.05.2008 18,9 0,0 -0,2 0,2 0,0 1,1 0,2 , 12 69

60

14.05.2008 19,4 2,6 0,2 2,6 0,2 1,1 2,3 , 25 35

60

15,05.2008 19,7 1,9 -0,1 2,1 0,0 1,1 1,9 , 18 78

60

16.05.2008 19,7 -0,4 0,0 -0,4 0,0 1,1 -0,3 ,

26 11 59

19.05.2008 19,6 -0,3 -0,2 -0,1 0.0 1,1 -0,1 ,

9 32 59

20.05.2008 19,5 -0,5 -1,5 0,5 1,7 1,1 0,4 , 20 64

59

21.05.2008 19,1 -2,0 -2,2 -0,6 4,1 1,1 -0,6 ,

18 33 56

22.05.2008 18,8 -1,6 -1,1 -0,9 0,8 1,1 -0,8 ,

15 55 53

23.05.2008 18,6 -1,0 . 0,9 -0,4 0,5 1,1 -0,4 ,

14 54 60

26.05.2008 18,5 -0,8 -0,3 -0,5 0,0 1,1 -0,4 ,

4 41 63

27.05.2008 18,3 -1,0 -0,2 -0,8 0.0 1,1 -0,7 ,

13 51 63

28.05.2008 18,7 2,3 1,8 1,2 3,8 1,1 1,1 , 16,52

64 59 29.05.2008 18,6 -0,7 0,7 -1,0 0,7 1,1 -0,9 11,41 55 30.05.2008 18,9 1,6 0,0 1.7 0,0 1,1 1,5 12 91 5 02.06.2008 18,6 -1,7 -O, 7 -1,2 0,2 1.1 -1,1

, 12 95

3

03.06.2008 18,4 -O, 8 0,3 -0,9 0,2 1,1 -0,8 ,

14 45 53

04.06.2008 18,3 -0,4 -0,6 0,0 0,2 1.1 0,0 , 15 61

53

05.06.2008 18,6 1,5 -0,2 1,7 0.0 1,1 1,5 , 14 37

58

06.06.2008 18,1 -3,0 -2,5 -1,5 5,3 1,1 -1,4 ,

26 49 58

09.06.2008 17,6 -3,0 -0,5 -2,6 0,1 1,1 -2,3 .

18 18 as

10.06.2008 17,3 -1,3 -0.5 -0,9 0,1 1,1 -O, 8 , 20 11

65

11.06.2008 17,2 -0,5 -1,7 0,5 2,2 1.1 0,4 , 23 90

66

12.06.2008 17,3 0,1 1,6 -0,7 3,1 1,1 -0,6 ,

26 18 64

13.06.2008 17,4 1,1 0,9 0,6 1,3 1,1 0,6 ,

13 58 65

16.06.2008 17,7 1,4 -0,4 1,7 0,1 1,1 1.6 , 15 50

63 17.06.2008 17,8 0,8 0,0 0,9 0,0 1,1 0,8 ,

14 71 59

18.06.2008 17,3 -2,9 -1,3 -2,1 1,1 1,1 -1,9 .

15 30 59

19.06.2008 17,0 -1,8 -0,9 -1,2 0,4 1,1 -1,1 ,

19 35 61

20.06.2008 16,5 -2,9 -1,6 -1,9 2,0 1,1 -1,7 ,

36 41 61

23.06.2008 16.3 -1,2 -1,3 -0,4 1,3 1,1 -0,3 ,

18 38 71

24.06.2008 16,3 0,0 -0,4 0,3 0,0 1,1 0,3 , 29 66

72

25.06.2008 16,7 2,3 1,0 1,8 1,5 1,1 1,6 , 19 06

80

26.06.2008 16,0 -4,2 -3,4 -2,3 10,1 1,1 -2,0 ,

28 23 73

27.06.2008 15,6 -2,6 -1,4 -1,7 1,4 1,1 -1,5 ,

22 61 78

30.06.2008 15,4 -1,7 -0,2 -1,5 0,0 1,1 -1,3 ,

22 79 BG

01.07.2008 15,1 -1,9 -2,8 -0,2 6,6 1,1 -0,2 ,

30 98 85

02.07.2008 15,4 1,9 0,1 2,0 0,1 1,1 1,8 , 23 36

87

03.07.2008 15,5 0,8 1,7 -0,1 3,7 1,1 -0,1 ,

27,27 84 85 04.07.2008 15,4 -0,6 -2,5 0,9 5,2 1,1 0,8 11,56 85 07.07.2008 15,4 0,0 1,3 -0,7 2,2 1,1 -0,6 13,66 85 08.07.2008 15,1 -2.2 -1,3 -1,4 1,3 1,1 -1,2 22,66 87 09.07.2008 15,7 4,0 2,4 2,8 6,7 1,1 2,5 18,86 83 10.07.2008 15,5 -1,5 -1,4 -0,6 1,5 1,1 -0,5 23,15 83 11.07.2008 15,0 -3,2 -3,0 -1,5 7,6 1,1 -1,3 27,10 as 14.07.2008 14,9 -0,5 1,0 -0,9 1,3 1,1 -0,8 18 71 8 15.07.2008 14,2 -5,2 -3,0 -3,4 8,0 1,1 -3,0

. 36 66

0 85 16.07.2008 14,1 -0,7 0,2 -0,7 0,1 1,1 -0,6

, 31,40 90 17.07.2008 15,1 6,8 2,9 5,3 9,7 1,1 4,6 28,18 80 18.07.2008 15,5 2,7 3,5 0,8 13,6 1,1 0,7 27,62 BO 21.07.2008 16,3 4,7 0,2 4,7 O, 1 1,1 4,2 26,02 77 22.07.2008 16,1 -1,6 -0,2 -1,4 O, 0 1,1 -1,3 23,00 77 23.07.2008 16,9 4,8 3,6 2,9 14,6 1.2 2,5 21,08 72 24.07.2008 16,9 0,1 -1,3 0,9 1.3 1,1 0,8 19,92 72 25.07.2008 16,0 -5,5 -4,1 -3,1 15,4 1,2 -2,7 30,54 73 28.07.2008 15,9 -0,8 -2,0 0,4 3,4 1,1 0.4 16,52 73 29.07.2008 15,5 . 2,2 0,1 -2.2 0,1 1,1 -2,0 20,08 73 30.07.2008 16,0 3,0 2,4 1,8 6,6 1,1 1,6 22,36 73 31.07.2008 16,1 0,5 0,0 0,6 0,0 1,1 0,5 23,55 76 01.08.2006 16,2 0,4 -0,3 0,7 0,0 1,1 0,6 17,75 78 04.08.2008 16,3 0,7 -0,6 1,2 0,2 1,1 1,1 18,24 81 05.08.2008 16,7 2,6 4,5 0,2 22,2 1,2 0,1 21,25 80 06.08.2008 17,1 2,4 0,4 2,3 0,3 1,1 2,1 21 40 80 07.08.2008 17,1 O, 1 0,1 0,2 0,1 1,1 0,2

, 19,79 81 08.08.2008 17,0 -1,2 1,0 -1,6 1,5 1,1 -1,5 18,72 81 11.08.2008 17,3 2,1 1,4 1,4 2,6 1,1 1,3 13,59 79 12.08.2008 17,6 1,4 -0,3 1,7 0,0 1,1 1,5 19,48 79 13.08.2008 17,5 -0,2 -3,4 1.8 10,3 1,1 1.6 29,28 79 14.08.2008 16,8 -4,6 0,5 -4,8 0,5 1,1 -4,3 21,66 79 15.08.2008 16,9 0,8 0,6 0,6 0,7 111 0,5 17.84 79 18.08.2008 16,9 -0,2 -0,6 0,3 0,2 1,1 0,2 11.90 78 19.08.2008 16,2 -4,2 -4,2 -1,8 16,0 1,2 -1,6 25,86 85 20.08.2008 15,9 -1,7 0,8 -2.1 1,0 1,1 -1,9 16,40 85 21,08.2008 15,7 -1,5 -1,5 -0,6 1,8 1,1 -0,5 15,52 92 22.08.2008 16,0 2,2 2,8 0,7 9,0 1,1 0,6 13,97 91 25.08.2008 16,1 0,4 0,0 0,6 0,0 1,1 0,5 6,17 91 26.08.2008 15,9 -1,2 0,0 -1,2 0,0 1,1 -1,0 20 98 93 27.0820013 16,0 0,3 -0,9 0,9 0,5 1,1 0,6 , 10,19 93 28.08.2008 16.2 1,8 2,7 0,4 8,2 1,1 0,3 18,47 93 2 08.2008 16.4 1,1 0.2 1,2 all 1,1 1,0 14 37 1 01.09.2008 16,5 0,2 -0,1 0,3 0,0 1,1 0,3 0,3

. 8,11 5 62 91 02.09.2008 16,9 2,6 2,7 1,1 8,5 1,1 1,0 1,0 22,13

, 5 62 88 03.09,2008 16,8 -0,7 -1,2 0,1 1,0 1,1 0,1 0,1 12,10

, 5 62 88 04.09.2008 16,4 -2,5 -2,7 -0,9 6,1 1,1 -0,8 -0,8 20,33 ,

62 5 90 05.09.2008 15,8 -3,4 -2,3 -2,0 4,3 1,1 -1,8 -1,6 25,54 ,

5 62 93 08.09.2008 16,9 6,1 0,0 6,2 0,0 1,1 5,5 5,5 40,93 ,

5 62 85 09.09.2008 16,9 0,4 5,1 -2,4 28,2 1,2 -2,0 -2,0 22,24 , 62 5 90 10.09.2008 16,5 -2,3 -1,3 -1,5 1,1 1,1 -1,4 -1,4 21,59

, 5 62 90 11.09.2008 16.0 -3.1 -1,4 -2,2 1,5 1,1 -1,9 -1,9 24,75

, 5 62 90 12.09.20Oß 16,3 ], 7 n. 5

_ 0,5 1,1 1.3 1.3 20.24

, cI eý

15.09.2008 15,4 -6,0 -6.3 -2,4 37,9 1,2 -2,0 -2,0 47,74 5 62 115 16.09.2008 14.1 . 8,9 -3.4 . 6,9 10,4 1,1 -6,1 -6,1 70,85 ,

5 62 130 17.09.2008 14,0 -1,1 -1,0 -0,5 0,6 1,1 -0,4 -0,4 57,03 ,

5 62 140 18.09.2008 13,2 -6,1 -1,1 -5,4 0,8 1,1 -4,8 -4,8 41.64 ,

5 62 140 19.09.2008 15,0 12,0 11,0 5,9 125,4 1.4 4.3 4.3 80.36 ,

5,62 130 243.9 Sum SARIt -8,7 1722,95 mod. SARI -14,2 Without 5.9.2008

supermodu SARk -18,5 Without 8.9.2008 an d 19.9.2008 R mean return of BEINSUR during the esbmation period

CARL -33 7

-0,19127 , T is the number of days In the estimation period modified CARt

88 -53 2

1/T . 0,011384

S standard error of the OLS markets 5I

repression SUp"rmodiflod CARt

-66 6 1,111773 , square of S standard error of the OLS market model regression

1,2360 sum of (Rmt-Rm squared)

243,3

Appendix D: Formulas, Variables, Results and Data of the event study method

AEGON

25.04.2008 10,1 0,0 0,0 0,0 0,0 1,1 0,0 11,56 3,93 116 28.04.2008 10,2 16 0,8 1,3 1,0 1,1 1,1 11,40 111 29.04.2008 10,1 -0,8 -1,4 -0,1 1,4 1,1 0,0 7,95 115 30,04,2008 10,2 0,5 1,5 -0,2 2,8 1,1 -0,2 10,44 120 02.05.2008 10,6 3,9 2,6 2,6 7,8 1,2 2,2 14,02 100 05.05.2008 10,7 0,4 -0,3 0,6 0,0 1,1 0,6 6,33 100 06.05.2008 10,4 -2,9 -1,7 -2,0 2,3 1,1 . 1,8 17,34 107 07.05.2008 10,2 -1,3 0,2 -1.4 0,1 1,1 -1,2 17.82 116 08.05.2008 10,2 -0,1 -0,6 0,3 0,2 1,1 0,3 11,50 120 09.05.2008 10,1 -1,2 -1,1 -0,5 0,9 1,1 -0,5 15,67 119 12.05.2008 10,1 0,5 0,0 0,5 0,0 1,1 0,5 6,75 117 13.05.2008 10,1 -0,3 -0,2 -0,2 0,0 1,1 -0,1 9,95 115 14.05.2008 10,1 -0,2 0,2 -0,2 0,2 1,1 -0,2 9,63 115 15.05.2008 30,1 0,2 -0,1 0,3 0,0 1,1 0,3 8,99 111 16.05.2008 10,2 1,3 0,0 1,3 0,0 1.1 1,2 10,61 109 19.05.2008 10,1 -1,1 -0,2 -0,9 0,0 1,1 -0,8 6,85 105 20.05.2008 10,0 -0,9 -1,5 -O, 1 1,7 1,1 -0,1 6,60 104 21.05.2008 9,9 -1,6 -2,2 -0,4 4,1 1.1 -0,4 11,85 107 22.05.2008 9,7 -2,0 -1,1 -1,4 0,8 1.1 -1,3 10,50 119 23.05.2008 9.6 -0,5 -0,9 0,0 0,5 1,1 0,0 9,04 127 26 05.2008 9,5 -1,4 -0,3 -1,2 0,0 1,1 -1,0 3,87 127 27.05.2008 9,5 -0,4 -0,2 -0,2 0,0 1,1 -0,2 8,17 129 28.05.2008 9,7 2,6 1,6 1,8 3,8 1,1 3,6 12,05 119 29.05.2008 9,7 -0,1 0,7 -D, 3 0,7 1,1 -0,3 8,46 114 30.05.2008 9,9 1,3 0,0 1,3 0,0 1,1 1,1 9,85 108 02.06.2008 10,0 1,4 -0,7 1,8 0,2 1,1 1,6 17,63 111 03.06.2008 9,8 -1,5 0,3 0,2 1,1 -1.4 9,60 114 04.06.2008 9,6 -2,3 -0,6 -1,9 0,2 1,1 -1,7 13,68 120 05.06.2008 9,6 0,2 -0,2 0,3 0,0 1,1 0,3 11,53 121 06.06.2008 9.4 -2,6 -2,5 -1,3 5,3 1,1 -1,1 15,13 127 09.06.2008 9,1 -2,9 -0,5 -2,6 0,1 1,1 -2,3 9,99 124 10,06.2008 9,0 -1,3 -0,5 -1,0 0,1 1,1 -0,9 12,63 125 11.06.2008 9,1 1,4 . 1,7 2,3 2,2 1,1 2,0 16.43 120 12.06.2008 9,2 0,7 1,6 -O, 1 3,1 1,1 -O, 1 34,01 116 13.06.2008 9,3 0,7 0.9 0,2 1,3 1,1 0,2 8,32 Ill 16.06.2008 9,3 0,1 -0,4 0,3 0,1 1,1 0,3 9,90 110 17.06.2008 9,3 0,3 0,0 0,4 0,0 1,1 0,3 9,86 100 18.06.2008 9,0 -3,1 -1,3 -2,4 1,1 1,1 -2,1 9,74 104 19.06.2008 8, B -2,9 -0,9 -2,5 0,4 1,1 -2,2 18,73 107 20.06.2008 8,6 -2,5 -1,6 -1,6 2,0 1,3 -1,4 19,99 319 23.06.2008 8,4 -1,9 -1,3 -1,2 1,3 1,1 -1,0 13,24 118 24,06.2008 8,3 -1,4 -0,4 -1,2 0,0 1,1 -1.0 15,16 130 25.06.2008 8,5 3,0 1,0 2, S 3,5 1,1 2,2 9,57 123 26.06.2008 8,5 -0,3 -3,4 1,4 10,1 1,2 1,2 28,36 129 27.06.2008 8,3 -1,9 -1,4 -1,2 1,4 1,1 -1,0 20,05 135 30.06.2008 8,4 0,6 . 0,2 0,7 0,0 1,1 0,7 37,71 129 01.07.2008 8.1 -3,3 -2,8 -1,8 6,8 1,1 -1 2067

, 136 02.07.2008 8,2 0,7 0,1 0,8 0,3 1,1 0.7 13,49 136 03.07.2008 8,2 0,7 1.7 -0,2 3,7 1,1 -0,1 16,43 140 04.07.2008 8,2 -0,9 -2.5 0,4 5,2 3,1 0,4 7.51 140 07.07.2008 8,1 -0,3 1,3 -0,9 2,2 1,1 -0,8 9,59 139 08.07.2008 7,9 -3,1 -1,3 -2,4 1,3 1,1 -2,1 15,65 145 09.07.2008 8,2 3,8 2,4 2,6 6,7 1,1 2,3 12,29 137 10.07.2008 8,1 -1.2 . 1,4 -0,4 1,5 1,1 -0,4 9,90 139 12.07.2008 7,9 -3,0 -3,0 -1,4 7,6 1,2 -1,3 15,31 138 14.07.2008 7,9 0,0 1,0 -0,4 1,3 1,1 -0,4 12,04 136 15.07.2008 7,4 -6,2 -3,0 -4,6 8.0 1,2 -4,0 29,07 142 16.07.2008 7,3 -1,2 0,2 -1,2 0,1 1,1 -1,1 21,07 138 17.07.2008 7,8 5,6 2,9 4,2 97 1,2 3,6 16,34 130 18.07.2008 8,0 2,5 3,5 0,7 13,6 1,2 0,6 19,40 129 21.07.2008 8,2 2,8 0,2 2,7 0,1 1,1 2,4 12,02 124 22.07.2008 7,9 -3,8 -0,2 -3,7 0,0 1,1 -3,2 17,70 125 23.07.2008 8,2 4,2 3.6 2,4 14,6 1,2 2,0 18,25 118 24.07.2008 8,2 -0,2 -1,3 0,5 1,3 1,1 0,5 11,17 117 25.07.2008 7,7 -7,1 -4,1 -5,0 15,4 1,2 -4,3 21.41 118 28,07.2008 7,4 -3,5 -2,0 -2,5 3,4 1,1 -2,2 14,98 118 29.07.2008 7,3 -2,2 0,1 -2,2 0,1 1,1 -2,0 15,40 120 30.07.2008 7,5 2,6 2,4 1,5 6,6 1,1 1,3 15,02 121 31.07.2008 7,5 0,6 0,0 0,7 O, 0 1,1 0,6 14,82 123 01.08.2008 7,5 0,2 . 0,3 0,4 0,0 1,1 0,4 10,50 129 04.08.2008 7,6 1,7 -0,6 2,1 0,2 3,1 1,8 12,81 133 05.08.2008 8,0 4,3 4,5 2,1 22,2 1,2 1,7 15,60 129 06,08.2008 8,2 2,5 0,4 2,3 0,3 1,1 2,0 14,10 128 07.08.2008 7.7 -6,0 0,1 -6,0 0,1 1,1 -5,3 30,42 132 08.08.2008 7,6 -1,7 1,0 -2,2 1,5 1,1 -1,9 21,36 134 11.08.2008 8,0 5,5 1,4 4,8 2,6 1,1 4,2 24,36 131 12.08.2008 8.3 2,8 .0 ,3 3,0 O, 0 1,1 2,7 17,30 129 13.08.2008 8,1 -2,3 -3,4 -0,6 10,3 1,2 -0,5 15,70 131 14.08.2008 8.1 -0,2 0,5 -0,5 0,5 1,1 -0,4 14.72 130 15.08.2008 8,2 1,8 0.6 1.5 0,7 1,1 1,3 11,40 130 18.08.2008 8,1 -0,9 -0,6 -0,5 0,2 1,1 -0,5 7,17 130 19.08.2008 7,9 -3,4 -4,2 -1,2 16,0 1,2 -1,1 11,28 136 20.08.2008 7,8 -1,0 0,8 -1,4 1,0 1,1 -1,2 8,33 136 21,08.2008 7,7 -1,1 -3,5 -0,3 1.8 1,1 -0,2 7,93 241 22.08.2008 7,9 2,5 2,8 1,1 9,0 1,2 1,0 7,81 235,5 25.08.2008 7,9 0,1 0,0 0,2 0,0 1,1 0,2 3,46 135,5 26.08.2008 7,8 -1,2 0,0 -1.2 0,0 1.1 -1,0 6,38 137,5 27.08.2008 7,8 -0,3 -0.9 0,2 0,5 1.1 0,2 6,74 143 28.08.2008 8,0 2,1 2,7 0.8 8,2 1,2 0,7 9,81 143 29 08 2008 ß1 1,6 0.2 1,5 0.1 1,1 1,4 9 36 141 01.09.2008 8,1 -0,1 -0,1 0,0 0,0 1,1 0,0 0.0 6,15 3,93 341 02.09.2008 B, 4 3,9 2,7 2,5 8,5 1,2 2,2 2,2 15,26 3,93 137 03.09.2008 8,3 -0,6 -1,2 0,1 1,0 1,1 0,0 0,0 8,74 3,93 141 04.09.2008 8,2 -1,7 -2,7 -0,3 6,1 1,1 -0,2 -0,2 12,75 3,93 145 05.09.2008 7,9 -3,7 -2,3 -2,5 4,3 1,1 -2,2 -2,2 13,33 3,93 148 08.09.2008 as 6,4 0,0 6,4 0,0 1,1 5,7 5,7 18,45 3,93 140 09.09.2008 8,6 2,3 5,1 -0,3 28,2 1,2 -0,2 -0,2 19,53 3.93 152 10.09.2008 8,5 -1,5 -1,3 -0,8 1,1 1,1 -0,7 -0,7 11,27 3,93 355 11.09.2008 8,3 -3,0 -1,4 -2 1,5 1,1 -2,0 -2,0 12,88 3,93 159 1209.7908 ß5 2.2 05 2.0 0.5 t, 1 1,7 1 ,7 12 94 393 161 `,,. _.. , ý.. ýý,., ý .-^. ýc . FTa: ^j,,...,, T rý., r . ný..., ýý.

ý"" 15.09.2008 7,6 -10,7 6,3 -7,5 37,9 1,2 6,1 6,1 37,40 3,93 198 16.09.2008 7,2 -6,5 -3,4 -4,7 10,4 1,2 -4,1 -4,1 31,03 3,93 238 17.09.2008 6,9 -3,3 -1,0 -2,8 0,6 1,1 -2,4 -2,4 31,13 3,93 240 18.09.2008 6,5 -6,3 -1,1 -5,6 0,8 1,1 -5,0 -5,0 25,85 3,93 219 19.09.2008 7,5 12,4 11,0 6,8 125,4 1,4 4,9 4,9 44,27 3,93 200

243,9 Sum SARk -8,4 1204,52 mod. SARI, -, 4,1 With out 5.9.2008

supermodd SARk -19,0 without 8.9.2008 and 19.9.2008 CARt

R mean return of BEINSUR during the estimation period -32,6 -0,19127

T Is the number of days in the estimation period modlfl"d CARt

66 366 -52.7 0,011364 supermodlfied CARt S standard error of the OLS market model regression -68,3 1.127421 square of S stand ard error of the OLS market model regression

1,2711 sum of (Rmt-Rm squared)

Appendix D: Formulas, Variables, Results and Data of the event study method

GNP Assurances

25.04.2008 71,5032 0 0 O 0,0 1,3 0, O 0,32 0,08 N/A 28.04.2008 71,8332 0,5 0,8 0,2 1,0 1,3 0,1 0,46 29.04.2008 71,1356 -1,0 -1,4 0,2 1,4 1,3 0,1 0,25 30.04.2008 71,6222 0,7 3,5 0,0 2.8 1,3 0,0 0,28

' 02.05.2008 73,2991 2,3 2,6 0,8 7,8 1,3 0,6 0,22 05.05.2008 73,4586 0,2 -0,3 0,7 0,0 3,3 0,1 0.20 05.05.2008 73,0289 -0,6 -1,7 0,8 2,3 1,3 0,6 0,35 07.05.2008 72,801 -0,3 0,2 -0,2 0,1 1,3 -O, 1 0,26 08.05.2008 72,5664 -0,3 -0,6 0,3 0,2 1,3 0,3 0,17 09.05.2008 71,7138 -1,2 -1,1 -0,2 0,9 1,3 -0,2 0,20 12.05.2008 72,9519 1,7 0,0 2,0 0,0 1.3 1,5 0,20 13.05.2008 72,9325 0,0 -0,2 0,4 0,0 1,3 0,3 0,29 14.05.2008 72,4654 -0,6 0,2 -0,6 0,2 1,3 -0,4 0,44 is . 05.2008 73,6949 1,7 -0,1 2,0 0,0 1,3 1.5 0,27 16.05.2008 75,5224 2,4 0,0 2,6 0,0 1,3 2,0 0.23 19.05.2008 76,3235 1,0 -0,2 1,4 0,0 1,3 1,1 0,32 20.05.2008 76,6962 0,5 -1,5 1,7 1,7 1,3 1,3 0,25 21.05.2008 76,535 -0,2 -2,2 1,5 4.1 1,3 1,1 0,23 22.05.2008 74,8414 -2,3 -1,1 -1,3 0,8 1,3 -1,0 0,20 23.05.2008 74,2281 -0,8 . 0,9 0,0 0,5 1,3 0,0 0,27 26.05.2008 73,9972 -0,3 -0,3 0.1 0,0 1,3 0,1 0,07 27.05.2008 73,5409 -0,6 -0,2 -0,2 0,0 1,3 -0,2 0,21 28.05.2008 75,4786 2,6 3,8 1,7 3,8 1,3 1,3 0,19 29.05.2008 75,3111 -0,2 0,7 -0,4 0,7 1,3 -0,3 0,26 30.05.2008 75,0155 -0,4 0,0 -0.2 0,0 1,3 -0,1 0,51 02.06.2008 74,7227 -0,4 -0,7 0,3 0,2 1,3 0,2 0,35 03.06.2008 74,9442 0,3 0,3 0,4 0,2 1,3 0,3 0.22 04.06.2008 74,4553 -0,7 -0,6 0,0 0.2 1.3 0.0 0,28 05.06.2008 74,7407 0,4 -0,2 0,7 0.0 1,3 0,6 0,16 06.06,2008 74,5291 . 0,3 -2,5 1,6 5,3 1,3 1,2 0,21 09.06.2008 73,1155 -1,9 -0,5 -1,4 0,1 1,3 -1,0 0,17 10.06.2008 71,9111 -1,7 -0,5 -1,1 0,1 1,3 -0,8 0,27 11.06.2008 72,9866 1,5 -1,7 2,8 2,2 1,3 2,1 0,46 12.06.2008 72,4656 . 0,7 1,6 -1,5 3,1 1,3 -1,2 0,23 13.06.2008 71,3668 -1,5 0,9 -1.9 1,3 1,3 -1,5 0,36 16.06.2008 72,1368 1,1 -0,4 1,6 0,1 1,3 1.2 0,17 17.06.2008 73,0404 1,2 0,0 1,5 0,0 1,3 1.1 0,21 18.06,2008 72,6863 -0,5 -1,3 0,6 1,1 1,3 0,4 0,26 19.06.2008 72,5106 -0,2 -0,9 0,6 0,4 1,3 0,4 0,32 20.06.2008 71,9558 -0.8 -1,6 0,5 2.0 1.3 0.4 0,31 23.06.2008 71,21 -1,0 -1,3 0,0 1,3 1.3 O, 0 0,27 24.06.2008 70,5734 -0,9 -0,4 -0,4 0,0 1,3 -0.3 0,36 25.06.2008 70,4321 -0,2 1.0 -0,6 1,5 1,3 -0,5 0,25 26.06.2008 70,0906 -0,5 -3,4 1.9 10,1 1,3 1,5 0,40 27.06.2008 68,0614 -3,0 -1,4 -1,9 1,4 1,3 -1.4 0,31 30.06.2008 67,7209 -0,5 -0,2 -O, 1 0,0 1,3 -0,1 0,27 01,07.2008 66,2062 -2,3 -2,8 -0,2 6,8 1,3 -0,2 0,31 02.07.2008 66,8063 0,9 0,1 1,1 0,1 1.3 0.8 0,28 03.07.2008 65,8942 -1,4 1,7 -2,3 3,7 1,3 -1,7 0,46 04.07.2008 65,8544 -011 -2,5 1.8 5,2 1,3 1,4 0,22 07.07.2008 65.4242 -0,7 1,3 -1,3 2,2 1,3 -1,0 0,28 08.07.2008 64,3862 -1,6 -1,3 -0,5 1,3 1,3 -0,4 0,33 09.07.2008 65,727 2,0 2,4 0,7 6,7 1,3 0,5 0,29 10.07.2008 67,9667 3,3 -1,4 4,5 1.5 1,3 3,4 0,57 11.07.2008 66,9464 -1,5 -3,0 0,6 7,6 1,3 0,5 0,24 14.07.2008 65,9954 -1,4 1.0 -1,8 1,3 1,3 -1,4 0,29 15.07.2008 63,7754 -3,5 -3,0 -1,3 8,0 1,3 -1.0 0,26 16.07.2008 63,5443 -0,4 0,2 -0,2 0,1 1,3 -0.2 0.32 17.07.2008 66,7508 4,8 2,9 3,1 9.7 1,3 2.4 0.29 18.07.2008 68,9652 3,2 3,5 1,2 13,6 1,3 0,9 0,32 21.07.2008 70,3413 2.0 0,2 2,1 0,1 1,3 1,6 0.34 22.07.2008 69,6506 -3,0 -0,2 -0,6 0,0 1,3 -0,5 0,26 23.07.2006 71,8322 3,0 3,6 0,9 14,6 1,3 0,7 0,18 24.07.2008 72,2928 0,6 -1,3 1,7 1,3 1,3 1,3 0,19 25.07.2008 69,8252 -3,5 -4,1 -0,6 15,4 1,3 -0,5 0.25 28.07.2008 68,02 -2,7 -2,0 -1,1 3,4 1,3 -0,8 0,26 29.07.2008 66,4764 -2,3 0,1 -2,2 0,1 1,3 -1,7 0,23 30.07.2008 67,7906 1,9 2,4 0,6 6,6 1,3 0,5 0,21 31.07.2008 68,0127 0,3 0,0 0,6 0,0 1,3 0.5 0,25 01.08.2008 68,8799 1,3 -0,3 1,7 0,0 1,3 1,3 0,25 04.08.2008 68,4985 -0,6 -0.6 0,1 0,2 1,3 0,1 0,18 05.08.2008 70,8278 3,3 4,5 0,6 22,2 1,4 0,4 0,30 06.08.2008 72,4145 2,2 0.4 2,2 013 1,3 1,7 0,21 07.08.2008 73,0526 0,9 0,1 1,1 0.1 1.3 0.8 0,18 08.08.2008 73,136 0,1 1,0 -0,3 1,5 1.3 -0,2 0,22 11.08.2008 74,1058 1,3 1.4 0,6 2,6 1,3 0,5 0,18 12.08.2008 74,6146 0,7 -0,3 1,1 0,0 1,3 0,9 0,21 13.08.2008 73,4042 -1,6 -3,4 0,8 10,3 1,3 0,6 0,26 14.08.2008 74,8024 1,9 0,5 1,8 0,5 1,3 1,4 0,23 15.08.2008 75,8216 1,3 0.6 1.2 0,7 1,3 0,9 0,17 38.08.2008 75,9426 0.2 -0.6 0.8 0,2 1,3 0,6 0,12 19.08.2008 75,1424 -1.1 -4,2 1,9 16.0 1.3 1.4 0,19 20.08.2008 74,8358 -0,4 as -0,7 1,0 1,3 -0,5 0,22 21.08.2008 74,3325 -0,7 -1,5 0.6 1,8 1,3 0,4 0,25 22.08.2008 75,7681 1,9 2,8 0,3 9,0 1,3 0.2 0,20 25.08.2008 76,3445 0,8 0.0 1,0 0,0 15 0,8 0.05 26.08.2008 75,9607 -O, 5 0,0 -0,3 O1O 1.3 -0,2 0,18 27.08.2008 74,7843 -1,6 -0,9 -0.8 0,5 1,3 -0,6 0,31 28.08.2008 76,0025 1,6 2,7 0,1 8,2 1,3 0,1 0,23 29 08.2008 77.7139 2,2 0.2 2,3 0.1 1,3 118 0.24 01.09.2008 77,6461 -0,1 -O, 1 0,2 0,0 1.3 0,1 0,1 0,11 0.08 02.09.2008 78,5799 1,2 2,7 -0.3 8,5 1,3 -0.3 . 0,3 0,31 0,08 03.09.2008 78,5905 0,0 -1,2 110 1,0 1.3 0.8 0,8 0,30 0,08 04.09.2008 78,4745 -0,1 -2.7 1,8 6,1 1,3 1,4 1,4 0,33 0,08 05.09.2008 75,7439 -3,6 -2.3 -1,9 4,3 1,3 -1,4 -1,4 0,32 0,08 08.09.2008 78,1939 3,1 0,0 3,4 O, 0 1,3 2.6 2.6 0,32 0108 09.09.2008 77,521 -0,9 5,1 -4.0 28,2 1.4 -2,9 -2,9 0,37 0,08 10.09.2008 76,6903 -1,1 -1,3 0.0 1,1 1.3 O, 0 0,0 0,44 0,08 11.09.2008 75,2098 -2,0 -1.4 -0,8 1,5 1,3 -0,6 -0,6 0,33 0,08 12.09.2003 76,5314 1,7 P, 5 2,6 0,5 1,3 1,3

T1.3 0,23 0,08

15.09.2008 74,8238 -2,3 -6.3 2,1 37,89712 1,4 1,5 1,5 0,33 0.08 16.09.2008 74,1057 -1,0 -3,4 1.5 10,37321 1,3 1.1 1.1 0,39 0,08 17.09.2008 73,5954 -0,7 -1,0 0,2 0,590428 1,3 0,1 0,1 0,42 0,08 18.09.2008 72,0602 -2,1 -1,1 -1,2 0,831295 1,3 -0,9 -0,9 0,36 0,08 19.09.2008 75,5629 4,6 11,0 -2,3 125,4039 1,6 -1,4 -1,4 0,52 0,08

243,9 Sum SARk 1,4 23,49 mod. SARI -1,2 Without 5.9.2008

aupermodit SARk 0,2 without 8.9.2008 and 19.9,2008 CARL

R mean return of BEINSUR during the estimation period 5,3 -0,19127

T Is the number of days In the estimation period modlflod CARt 88 -4.7

irr 0.011364 super

S standard error of the OLS marke Rlel

regression Ö 7Ifloq CARt

1.286832 square of S standard error of the OLS market model regression

1,6559 sum of (Rmt-Rm squared)

243.3

Appendix D: Formulas, Variables, Results and Data of the event study

I method

legal & General '- T

24.04.2008 25.04.2008 28.04.2008 29.04.2008 30.04.2008 01.05.2008 02.05.2008 06.05.2008 07.05.2008 08.05.2008 09.05.2008 12.05 13.05,2008 14 05.2008 15.05.2008 16.05.2008 19.05.2008 20.05.2008 21.05.2 008 22.05.2008 23.05.2008 27.05.2 DO" 28.05.2008 29.05.2 008 30.05.2 008 02.06.2008 03.06.2008 04.06.2008 05.06.2008 06.06.2 008 09.06.21) 11 10.06.2 Doll 11.06.2008 12.06.2008 13.06.2008 16.06.2008 17.06.2008 18.06.2 008 19.06.2008 20.06.20 08 23.06.2 008 24.06.2008 25.06.2 008 26.06.2008 27.06.2008 30.06.2008 01.07.2 o08 02.07.2008 03.07.2008 04.07.2008 07.07.2008 08.07.2 "0" 09.07.2008 10.07.2008 11.07.2 008 14.07.2008 15.07.2008 16.07.2008 17.07.2008 18.07.2008 21.07.2008 22.07.2008 23.07.2008 24.07.2008 25.07.2008 28.07.2008 29.07.2008 30.07.2008 31.07.2 "0" 01.08.2008 04.08.2008 05.08.2008 06 08.2008 07.08.2 008 08,08.2008 11.08.2008 12.08.2008 13.08.2008 14.08.2008 15.08.20 08 18,08.2008 19.08.2008 20.08.2008 21.08.2 008 22.08.2008 26.08.2008 27.08.2008 28.08.2008

02.09.2008 03.09.2008 04.09.2008 05.09 . 2008 08.09.2008 09.09.2008 10.09.2008 11.09.2008

15.09.2008 16.09.2008 17.09.2008 18.09.2008 19.09.2008

125,5 0.0 0,0 0,0 0,0 1.7 0,0 35,08 9,24 124,7 -0,6 0,8 -0,5 1,0 1,7 . 0,3 32,10 126,9 1,7 -1,4 2,0 1,4 1,7 1,2 20,05 128,5 1,2 1,5 1,4 2,8 1,7 0.8 18,25 126,6 -1,5 2,6 -1,5 7,8 1,7 . 0,9 16,11 126,5 0,0 -0,3 0,2 0,0 1,7 0,1 17,64 125,1 -3,1 -1,7 -0.6 2,3 1,7 -0,5 10,64 127,2 1,6 0,2 1,8 0,1 1,7 1,1 35,99 127,7 0.4 -0.6 0,6 0,2 1,7 0,4 40,48 127,3 -0,3 -1,1 -0,1 0,9 1,7 0,0 21,92 127,1 -0,2 0,0 0,0 0,0 1,7 0,0 25,06 124,9 -1,8 -0,2 -1,6 0,0 1.7 -1,0 17,45 125,1 0,2 0,2 0,4 0,2 1,7 0,2 16,46 124,1 -0,8 -0.1 -0,7 0,0 1,7 -0,4 27,48 123,5 -0,4 0,0 -0,3 0,0 1,7 -0,2 40,18 122,2 -1,1 -0,2 -0.9 0,0 1,7 -0,6 26,12 124,0 1,5 -1,5 1,7 3,7 1,7 3,0 23,03 124.1 0,1 -2,2 0,3 4,1 1,7 0,2 21,12 123,3 -0,6 -1,1 -0,4 0,8 1,7 . 0,2 25,46 120,2 -2,6 -0,9 -2,4 0,5 1,7 -1,5 36,78 138,3 -1,6 -0,3 -1.4 0,0 1,7 -0,8 33,24 119,2 0.8 -0,2 1.0 0,0 1,7 0.6 27,14 117,7 -1,2 1,8 . 1,1 3,8 1,7 -0,7 20,52 119,7 1,6 0,7 1,8 0,7 1,7 1,1 18,13 119,0 -0,6 0.0 -0,4 0,0 1,7 -0,3 33,30 120,7 1,4 -0,7 1,6 0,2 1,7 1,0 34,21 117,4 -2,8 0,3 -2,7 0,2 1,7 -1,6 26,86 118,5 0,9 -0,6 1,1 0,2 1,7 0,7 31,62 117,9 -0,5 -0.2 -0,3 0,0 1,7 -0,2 27,34 118,6 0,6 -2,5 0,8 5,3 1,7 as 28.13 117,5 -0.9 -0,5 -0,7 0.1 1,7 -0,4 29,24 115,5 -1,8 -0,5 -1,6 0,1 1.7 -1,0 23,86 113,9 -1,4 -1,7 -1,1 2,2 1,7 -0,7 38,83 112,4 -1,3 1,6 -1,2 3,1 1,7 -0,7 47,74 112,7 0,3 0,9 0,4 1,3 1,7 0,2 48,36 112,5 -0,2 -0,4 0,0 0,1 1,7 0,0 17.72 114,0 1,3 0,0 1.5 0,0 1.7 0,9 29,22 116,3 2,0 -1,3 2,2 1,1 1,7 1,4 24,43 111,8 -4,0 -0,9 -3,8 0,4 1,7 -2,3 66,36 308,8 -2,8 -1,6 -2,5 2,0 1,7 -1,5 33,20 107,2 -1,5 -1,3 -1,3 1.3 1,7 -0,8 41,63 105,8 -1,3 -0,4 -1,1 0,0 1,7 -0,7 22,48 105,8 0,0 1,0 0,1 1,5 1,7 0,1 32,41 105,7 0,0 -3,4 0,3 10,1 1,7 0,2 25,80 101,6 -4,0 -1,4 . 3,8 1,4 1,7 -2,3 33,45 100,1 -1,5 -0,2 . 1.3 0,0 1,7 . 0,8 26,43

99,7 -0,3 -2,8 0,0 6,8 1,7 0,0 26 11 97,6 -2,2 0,1 -2,0 0,1 1,7 -1,2 28,17 99,5 1,9 1,7 2,0 3,7 1,7 1,2 35,41 99,2 -0,3 -2,5 0,0 5,2 1,7 0.0 51.15 95,2 -4,3 1,3 -4,1 2,2 1,7 -2,5 3076 , 95,8 0,7 -1,3 0,9 1,3 1,7 0,5 35 , 77 95,3 -0,5 2,4 . 0,4 6,7 1,7 -0,3 23,22

100,8 5,5 -1.4 5.7 1,5 1,7 3.4 37,52 98,9 -1,9 -3,0 -1.6 7,6 1,7 -1,0 33,10 95,5 -3,6 1.0 -3,4 1,3 1,7 -2.1 41,26 95,8 0,3 -3,0 0,6 8,0 1,7 0,4 29,79 91,3 -4,9 0,2 -4,7 0,1 1,7 -2,9 27,51 91.2 -0,1 2,9 0, D 9,7 1,7 0,0 34,04 95.3 4,2 3,5 4,3 13,6 1,7 2,5 44,57 99,8 4,5 0,2 4,7 0.1 1,7 2,8 44,45

100,2 0,4 -0,2 0,5 0,0 1,7 0,3 25,07 99,9 -0,3 3,6 -0,2 14,6 1,7 -0,1 31,73

105,6 5,5 -1,3 5.7 1,3 1,7 3,4 41,50 106,2 0,5 -4,1 0,8 15,4 1,7 0,5 37,92

99,0 -7,2 -2,0 -7,0 3,4 1,7 -4,2 39,32 96,4 -2.7 0,1 . 2,6 0,1 1,7 -1,5 23,22 93,0 -3,6 2,4 -3,5 6,6 1,7 -2,1 29,08 98.1 5,2 0,0 5,4 0,0 1,7 3.2 34.66 97.2 -0,9 -0,3 -0,7 0,0 1,7 -0,4 29,06 97,1 -0,1 . 0,6 O, 1 0,2 1,7 0,1 23,95 97.1 0,0 4,5 0,0 22,2 1,7 0,0 15,22

105,0 7,5 0,4 7,7 0,3 1,7 4,6 75,44 105,8 as 0,1 1,0 0,1 1,7 O, 6 51,96 108,7 2,6 1,0 2,8 1,5 3,7 1,7 47,97 109,4 0,7 1,4 0,8 2,6 1,7 0,5 33,61 110,1 0,6 -0,3 0,8 0,0 1,7 0,5 26,52 11110 0,8 -3,4 1,3 10,3 1,7 0,6 24,21 106,1 -4,6 0,5 -4,4 OS 1,7 -2,7 28,06 105,7 -0,4 0,6 -0,3 0,7 1,7 -0,2 31: 98 105,4 -0,3 -0,6 -0,1 0,2 1,7 -O, 1 22,95 103,2 -2,1 -4,2 -1,8 16,0 1,7 -1,0 20,67

97,6 -5,7 0.8 -5,5 1,0 1,7 -3,3 27,90 96,2 -1,4 -1,5 -1,2 1,8 1,7 -0,7 38,12 95,4 -0,9 2,8 -0,8 9,0 1.7 -0,5 23,57 97,9 2,6 0,0 2,8 0,0 1,7 1,7 64,30 96,6 -1,4 0,0 -1,2 0,0 1,7 -0.7 23,61 96,3 -0,3 -0,9 -0.1 0,5 1.7 -0,1 44,21

100,5 4,3 2,7 4,4 8,2 1,7 2,6 24,12 100 9 0,3 0.2 0,5 0.1 1,7 0,3 45.18 101,7 0,8 -0,1 1,0 0,0 1,7 0,6 0,6 17,88 9,24 104,6 2,7 2,7 2,8 8,5 1,7 1,7 1,7 20,22 9,24 300,1 -4,4 -1,2 -4,2 1.0 1.7 -2,5 -2,5 29.87 9,24

974 29 -2,7 -2,6 6,1 1,7 -1,5 -1,5 28,36 9,24 94,5- -3,0 -2,3 -2,7 4,3 1,7 -1,6 -1,6 28,67 9,24

100.0 54 0,0 5.6 010 1,7 3,4 3,4 15,71 9,24 103,1 3,1 5,1 3,0 28,2 1,8 1,7 1,7 41.31 9,24 100,8 -2.3 -1,3 -2.1 1.1 1,7 -1,3 -1,3 32,13 9,24

98,1 -2,7 -1,4 -2,5 1,5 1,7 -1,5 -1,5 23,47 9,24 98.

M1 ii 0.3 O. `

s_..... a r r, 04 as 1,7 0,2 D2 23,93 9 24

_ 93,5 . -5,2 -6,3 -4,8 37,9 1,8 -2,7 nm

-2.7 mi

53,05 9,24 87,4 -7,0 -3,4 -6,7 10,4 1.7 -3,9 -3,9 91,92 9,24 90,4 3,3 -1,0 3,5 0,6 1,7 2,1 2,1 54,29 9,24 89,3 -1,2 -1,1 -1,0 0,8 1,7 -0,6 -0,6 48,59 9,24

106,0 15,8 11,0 35,5 125,4 2,0 7,6 7.6 73,97 9,24 243.8714 Sum SARk 1,7 2835,00

mod. SARI -1.7 Without 5.9.2008 supermodil SARk -9.3 wi thout 8.9.2000 and 19.9.2008

CARL R mean return of BEINSUR during the estimation period 6.6

-0,19127 T Is the number of days in the estimation period Modified CA Rt

88 -6.3 irr

0,011364 sup. rmodtflad CARL S standard error of the OLS market3-l regression -33,5

1,651721 square of S standard error of the OLS market model regression

2.7282 sum of (Rmt-Rm squared)

243,3

60 60 58 55 57 63 58 53 57 58 58 60 60 60 60 59 59 59 56 56 62 65 65 59 57 55 57 58 63 63 60 65 66 64 64 62 GO 60 60 60 66 67 72 68 72 79 78

75 75 74 74 74 74 74 74 77 77 80 79 85 82 82

82 83 86 89 B0 85 89 89

115

140 130

Appendix D: Formulas, Variables, Results and Data of the event study method

Zurich

23.04.2008 307,8 0,0 0,0 0,0 0,0 1,6 0,0 24.04 200 304 9 -0,9 0,8 -0,9 1,0 1,6 -0,5

: : 25.04,2008 313,2

: 2,6 -1,4 3,0 1,4 1,6 1,8 28.04.2008 315,8 0,8 1,5 0,8 2,8 1,6 0,5 29.04.2008 314,8 -0,3 2,6 -0,6 7,8 1,7 -0,4 30.04.2008 314,8 0,0 -0,3 0,2 0,0 1,6 0,1 02.05.2008 324,8 3,1 -1,7 3,5 2,3 1,6 2,1 05.05.2008 324,1 -0,2 0,2 -0,1 0,1 1,6 0,0 06.05.2008 318,5 -1,8 -0,6 -1,5 0,2 1,6 -0,9 07.05.2008 319,0 0,2 -1,1 0,5 0,9 1,6 0,3 08.05,2008 317,2 -0,6 0,0 -0,4 0,0 1,6 -0,2 09.05.2008 311,7 -1,8 -0,2 -1,6 0,0 1,6 -1,0 13.05.2008 312,4 0,2 0,2 0,4 0,2 1,6 0,2 14.05.2008 314,3 0,6 -0,1 0,8 0,0 1,6 0,5 15.05.2008 323,4 2,8 0,0 3,0 0,0 1,6 1,8 16.05.2008 321,5 -0,6 -0,2 -0,4 0,0 1,6 -02 19,05.2008 320,6 -0,3 -1,5 0,2 1,7 1,6 0,1 20.05.2008 318,3 -0,7 -2,2 -0,2 4,1 1,7 -O, 1 21.05.2008 314,5 -1,2 -1,1 -O, 8 O, 8 1,6 -0,5 22.05.2008 307,4 -2,3 -0,9 -2,0 0,5 1,6 -1,2 23.05.2008 309,4 0,7 -0,3 0,9 0,0 1,6 0,5 26.05.2008 302,8 -2,2 -0,2 -2,0 0,0 1,6 -1,2 27,05,2008 301,5 -0,4 1,8 -0,6 3,8 1,7 -0,3 28.05.2008 299,6 -0,6 0,7 -0,6 0,7 1,6 -0,3 29.05.2008 302,9 1,1 0,0 1,3 0,0 1,6 0,8 30. 306,6 1,2 -0,7 1,5 0,2 1,6 0,9 02.06.2008 303,2 -1,1 0,3 -1,0 0,2 1,6 -0,6 03 c6.2""8 301,4 -0,6 -0,6 -0,3 0,2 1,6 -0,2 04.06.2008 299,4 -0,7 -0,2 -0,5 0,0 1,6 -0,3 05.06,2008 302,8 1,1 -2,5 1,7 5,3 1,7 1,0 06.06,2008 297,4 -1,8 -0,5 -1,5 0,1 1,6 -0,9 09.06.2008 294,0 -1,2 -0,5 -0,9 0,1 1,6 -0,6 10.06.2008 291,1 -1,0 -1,7 -0,5 2,2 1,6 -0,3 11.06.2008 290,0 1,6 -0,2 3,1 1,6 -0,1 12,06.2008 285,6 -1,8 0,9 -1,8 1,3 1,6 -1,1 13.06.2008 281,4 -2,5 -0,4 -1,3 0,1 1,6 -0,8 16.06.2008 287,0 2,0 0,0 2,1 0,0 1,6 1,3 17,06.2008 287,8 0,3 -1,3 0,7 1,1 1,6 0,4 18.06.2008 281,7 -2,2 -0,9 -1,8 0,4 1,6 -1,1 19.06.2008 277,4 -1,6 -1,6 -1,1 2,0 1,6 -0,7 20.06.2008 274,2 -1,1 -1,3 -0,7 1,3 1,6 -0,5 23.06.2008 269,3 -1,8 -0,4 -1,6 0,0 1,6 -1,0 24.06.2008 268,0 -0,5 1,0 -0,5 1,5 1,6 -0,3 25.06.2008 270,3 0,8 -3,4 1,6 10,1 1,7 1,0 26.06.2008 263,0 -2,8 -1,4 -2,4 1,4 1,6 -1,4 27.06.2008 256,6 -2,5 -0,2 -2,3 0,0 1,6 -1,4 30,06.2008 258,6 0,8 -2,8 1,4 6,8 1,7 0,9 01.07.2008 256,3 -0,9 0,1 -O, 8 0,1 1,6 -0,5 02.07.2008 255,4 1,7 -0,5 3,7 1,7 -0,3 03.07.2008 254,7 -0,3 -2,5 0,3 5,2 1,7 0,2 04.07.2008 254,6 0,0 1,3 -0,1 2,2 1,6 -0,1 07.07,2008 253,3 -0,5 -1,3 -0,1 1,3 1,6 -0,1 08.07.2008 246,9 -2,6 2,4 -2,8 6,7 1,7 -1,7 09.07.2008 254,1 2,8 -1,4 3,2 1,5 1,6 2,0 10.07 . 2008 252,3 -0,7 -3,0 0,0 7,6 1,7 0,0 11.07.2008 267,8 5,8 1,0 5,8 1,3 1,6 3,5 14.07.2008 266,6 -0,4 -3,0 0,3 8,0 1,7 0,2 15.07.2008 257,7 -3,4 0,2 -3,3 0,1 1,6 -2,0 16.07.2008 254,8 -1,1 2,9 . 1,5 9,7 1,7 . 0,9 17.07.2006 265,2 3,9 3,5 3,5 13,6 1,7 2,1 18.07 268,9 1,3 0,2 1,5 0,1 1,6 0,9 21.07.2008 271,1 0,8 -0,2 2,0 0,0 1,6 0,6 22.07.2008 270,1 -0,4 3,6 -0,8 14,6 1,7 -0,5 23.07.2008 282,2 4,3 -1,3 4,7 1,3 1,6 2,8 24.07.2008 286,0 1,3 -4,1 2,2 15,4 1,7 1,3 25.07.2008 274,5 -4,2 -2,0 -3,7 3,4 1,6 -2,2 28.07.2008 270,7 -1,4 0,1 -1,2 0,1 1,6 -0,8 29.07,2008 268,6 -0,5 2,4 -1,0 6,6 1,7 -0,6 30.07.2008 275,6 2,6 0,0 2,7 0,0 1,6 1,7 31.07.2008 276,5 0,3 -0,3 0,5 0,0 1,6 0,3 04.08.2008 274,0 -0,9 -0,6 -0,6 0,2 1,6 -0,4 05.08.2008 278,3 1,5 4,5 1,0 22,2 1,7 0,6 06.08.2008 283,3 1,8 0,4 1,9 0,3 1,6 1,1 07.08.2008 283,1 0,0 0,1 0,1 0,1 1,6 0,1 08.08,2008 282,7 -0,1 1.0 -0,1 - 1,5 1,6 -O, 1 11.08.2008 287,1 1,5 1,4 1,5 2,6 1,6 0,9 12.08.2008 288,2 0,4 -0,3 0,6 0,0 1,6 0,4 13.08.2008 281,5 . 2,4 -3,4 -1,6 10,3 1,7 -1,0 14,08.2008 284,1 0,9 0,5 1,0 0,5 1,6 0,6 15.08.2008 285,6 0,5 0,6 0,6 0,7 1,6 0,4 18.08.2008 282,0 -1,3 -0,6 -1,0 0,2 1,6 -0,6 19.08.2008 275,5 -2,4 -4,2 -1,5 16,0 1,7 -0,9 2(). a8.2008 274,7 -0,3 0,8 -0,2 1,0 1,6 -0,1 21.08.2008 274,1 -0,2 -1,5 0,2 1,8 1,6 0,1 22.08,2008 278,9 1,7 2,8 1,4 9,0 1,7 0,8 25.08.2008 282,2 1,2 0, O 1,4 0,0 1,6 0,8 26.08.2008 279,2 -1,1 0,0 -0,9 0,0 3,6 -0,6 27.08.2008 277,2 -0,7 -0,9 -0,4 0,5 1,6 -0,2 28.08,2008 283,8 2,3 2,7 2,0 8,2 1,7 1,2 29.08 2008 288,5 1,7 0.2 1,8. O. 1 1,6 3,1 01.09.2008 285,0 -1,2 -0,1 -1,1 0,0 1,6 -0,6 02.09.2008 291,7 2,3 2,7 2,0 as 1,7 1,2 03.09,2 008 290,6 -0,4 -1,2 0,0 1,0 1,6 0,0 0409 2008 285,8 -1,7 -2,7 -1,1 6,1 1,7 -0,6 05.09.2008 278,8 -2,5 -2.3 -1,9 4,3 1,7 -1,2 08.09.2008 290,3 4,0 0,0 4,1 0,0 1,6 2,5 09.09.2008 293,5 1,1 5,1 0,4 28,2 1,7 0,2 10.09.2008 288,9 -1,6 -1,3 -1,2 1,1 1,6 -0,7 11.09.2008 282,3 -2,3 -1,4 -1.9 1,5 1,6 -1,2 12.09.2008 794.4 ! 1.4 0,5 0,8 0,5 1,6 0.5

15.09.2008 272,0 -4,6 -6,3 -3,3 37,9 1,6 . 2,0 16.09.2008 265,1 -2,6 -3,4 -1,8 10,4 1,6 -1,1 17.09,2008 278,5 4,8 -1,0 5,1 0,6 1,6 3,1 18.09.2006 277,5 -0,3 -1,3 0,0 0,8 1,6 0.0 19.09.2008 293,9 5,6 11,0 3,9 125,4 1,6 2,4

243,9 Sum SARK

mod. SARI supermodal SARk

2,48 1,68 1,55 0,80 1,12 0,83 1,20 0,44 1,04 0,73 0,65 1,34 1,45 1,05 1,69 1,19 0,87 0,66 0,79 1,30 1,01 0,51 1,06 1,27 0,95 0,88 0,75 0,72 0,72 0,74 0,88 0,76 1,21 1,19 1,88 1,34 1,41 1,03 1,00 1,44 2,19 1,40 1,37 1,05 1,63 1,64 1,21 1,13 1,15 1,25 0,52 0,65 1,37 1,26 0,86 3,75 1,27 1,86 1,91 1,90 1,88 1,10 1,23 1,70 1,36 2,17 1,13 1,06 0,86 0,65 0,76 0,93 0,86 0,66 0,81 0,88 1,16 1,53 0,88 0,97 0,86 1,31 0,89 0,77 1,15 0,66 0,65 0,87 0,98

0,34

67 63 67 68 60 60 65 66 70 70 67 67 66 64 64 67 70 80 86 86 89 82 79

81 81

82 87 82 81 82 83 80 78 78 76 76 76 80 79 82

77,5 77,5 79,5

81 79

1,2 1,14 0,34 73

O, 0 1,01 0,34 75 -0,6 1,39 0,34 76

-1,2 1,22 034 79 2,5 1,76 0,34 73 0,2 1,10 0,34 76

-0,7 1,09 0,34 75

-1,2 1,08 0,34 75 0,5 0.96 0.34 78

-2,0 1,88 0,34- 103 -1,1 2,37 0,34 113

3,1 2,85 0,34 110 0,0 1,90 0,34 105 2,4 2,91 0.34 85 2,4 104,68

-0,1 Without 5.9.2008 -2,5 Without 8.9.2008 and 19.9.2008

CARt R mean return of BEINSUR during the estimation period 9,3

-0,19127 T Is the number of days In the estimation period modlfl. d CARt

88 -0,5 1! T 0.011364 sup-modified CARt

S standard error of the OLS market , 4169

regression -8,9 1,629483 square of S standard error of the OLS market model regression

2,6552 sum of (Rmt-Rm squared)

243.3

Appendix D: MU I hRO Formulas, p

share rice On c

R

hange

P_

change In Ab

AR

normal

R_ - SD

moan stand it SAR a

ard " AR Sum SAR Volu mon of Mini mum SY C Ds

Variables ; Date t he A'12 In 1 % sh!! ro R rn etu Return deviat ion 3 shar es In Volu me Is ( b p) , 25.04.2008 23,2 0.0 0,0 0,0 0,0 .2 0,0 1,3 0.50 43

Results and

28.04,2008 123,7 0,4 0,8 -0,1 1,0 1,2 -0,1 1,1 40 29.04.2008 122,5 -1,0 -1,4 0,1 3,4 1,2 0,1 1,3 42

Data of the 30.04.2008 123,5 0,7 1.5 -0,3 2.8 1,2 -0,2 1,4 43 02.05.2008 126,4 2,3 2,6 0,5 7,8 1,3 0,4 2,3 39

event study 05.05.2008 126,8 0,3 -0,3 0,6 0,0 1,2 0,5 0,6 39 06.05.2008 124,7 -1,7 -1,7 -0,4 2,3 1,2 -0,3 1,9 42

method 07.05.2008 124,8 0,1 0,2 0,0 0,1 1,2 0.0 1,1 43 08.05.2008 121,7 -2,6 -0,6 -2,1 0,2 1,2 -1,7 3,3 48 09.05.2008 121,1 -0,4 -1,1 0,5 0,9 1,2 0,4 1,5 47 12.05.2008 122,3 0,9 0,0 1,0 0,0 1,2 0,8 0,6 46 13.05.2008 122,2 -0,1 -0,2 0,1 O, 0 1,2 0,1 1,2 44 14.05.2008 122,4 0,2 0,2 0,1 0,2 1,2 0,1 1,3 44 15.05.2008 121,1 -1,0 -0,1 -0,9 0,0 1,2 -0,7 1,4 43 16.05.2008 121,9 0,6 0,0 0,6 0,0 1,2 0,5 1,7 41 19,05.2008 120.8 -0,9 "0,2 -0,7 0,0 1,2 -0,6 1.3 37 20.05.2008 120,2 -0,5 -3,5 0,7 1,7 1,2 0,5 1,2 43 21.05.2008 119,0 .1 C) -2,2 0,7 4,1 1,3 0.6 1,6 45 22.05.2008 118,3 -0,6 -1,1 0,2 0,8 1,2 0,2 1,0 50 23.05.2008 118,8 0,4 -0,9 1,2 0,5 1,2 0,9 1,5 54 26.05.2008 117,9 -0,8 -0,3 -0,5 0,0 1.2 -0.4 0,5 54 27.05.2008 117,9 0,0 -0,2 0,2 0,0 1,2 0,1 1.1 55 28.05.2008 119,5 1,4 1,8 0,1 3,8 1,3 0,1 1,4 49 29.05.2008 119,9 0,3 0,7 -0,1 0,7 1,2 -0,1 1,3 47 30.05.2008 120,7 0,6 0,0 0,7 0,0 1,2 0,5 1,1 45 02.06.2008 119,4 -1,0 -0,7 "0,5 0,2 1,2 -0,4 0,9 48 03.06.2008 119.1 -0,2 0,3 -0,4 0,2 1,2 -0,3 1,1 48 04.06.2008 118,8 -0,2 -0,6 0,3 0,2 3,2 0,2 1,4 50 05.06.2008 119,8 0.8 -0,2 1,0 0,0 1,2 0,8 1,6 50 06.06.2008 119,6 -0,2 -2,5 1,7 5,3 1,3 1,3 2,1. 52 09.06.2008 118,9 -0,6 -0,5 -0,1 0,1 3,2 -0,1 1,2 55 10.06.2008 118,2 . 0,6 -0,5 -0,2 0,1 1,2 -0,2 1,5 56 11.0s. 20013 118,8 0,5 -1,7 1,8 2,2 1,2 1,5 2,2 52 12.06.2008 119,9 0,9 1,6 -0,2 3,1 1,3 -O, 1 2,3 51 13.06.2008 120,9 0,8 0,9 0,2 1,3 1,2 0,2 1,6 48 16.06.2008 121,7 0.6 -0,4 1,0 0,1 1,2 0,8 1,4 47 17.06.2008 121,7 0'0 0,0 0,1 0,0 1.2 0,1 1,1 45 18.06.2008 120,6 .0 ,9 -1,3 0,1 1,1 1,2 0,1 1,3 48 19.06.2008 120,3 -0,2 -0,9 0,5 0,4 1,2 0,4 1,5 50 20.06.2008 119,1 -1,1 -1,6 0,2 2,0 1,2 0,1 4,2 55 23.06.2008 117,3 -1,5 -1,3 -0,5 1,3 1,2 -0,4 2,0 54 24.06.2008 115,5 "1,5 -0,4 -1,1 0,0 1,2 -0,9 2,3 57 25.06.2008 115,9 0,3 1,0 -0,4 1,5 1,2 -0,3 1,4 52 26.06.2008 114,6 "1,1 "3,4 1.4 10,1 1,3 1,1 1,7 57 27.06.2008 112,0 "2.4 -1,4 -1,3 1,4 1,2 -1,0 2,3 60 30.06.2008 110,8 -1,0 -0,2 . 0,8 0,0 1,2 -0,6 2,1 58 01.07.2008 109,7 -1, D -2,8 1,1 6,8 1,3 0.8 1,7 60 02.07.2008 109.2 -0,4 0,1 -0,4 0,1 1,2 -0,3 2,2 61 03.07.2008 108,6 -0,6 1,7 -1,8 3,7 1,3 -1,4 2,1 61 04.07.2008 108, G 0,0 -2,5 1,9 5,2 1,3 1.5 0,9 61 07.07.2008 109,0 0,4 1,3 -0,5 2,2 1,2 . 0.4 1.0 60 08.07.2008 108,3 -0,6 -1,3 0,4 1,3 1,2 0,3 2.1 63 09.07.2008 110,2 1,7 2,4 0,0 6,7 1,3 0,0 1,8 59 10.07.2008 109.5 -0,6 -1,4 0,5 1,5 1.2 0,4 1,5 59 11.07.2008 109,0 -0,5 -3,0 1,7 7,6 1,3 1,3 2,2 58 14.07.2008 109,3 0,3 1,0 -0,4 1,3 1,2 -0.3 1,2 55 15.07.2008 106,7 -2,4 -3,0 -0,2 8,0 1.3 -0,1 3,0 58 16.07.2008 106,1 -0,5 0,2 -0,6 0,1 1,2 -0,5 2,5 56 17.07.2008 108,3 2,0 2,9 0,0 9,7 1,3 0,0 2,0 53 18.07.2008 110,8 2,2 3,5 -0,3 13,6 1,3 -0,2 3,1 52 21.07.2008 112,6 1,6 0,2 1,5 0,1 1,2 1,2 1,6 49 22.07.2008 112,7 0,3 -0,2 0,3 0,0 1,2 0,2 1.3 49 23.07.2008 115,7 2,6 3,6 0,0 14,6 1,3 0,0 1,9 45 24.07.2008 116,4 0,6 -1,3 1,6 1,3 1,2 1,3 1,5 45 25.07.2008 105,6 -10,2 -4,1 -7,1 15,4 1,3 -5,6 11,7 48 28.07.2008 105,3 "0,2 -2,0 1,3 3,4 1,3 1,0 2.6 47 29.07.2008 104,9 -0,4 0,1 -0,4 0,1 1,2 -0,4 2,1 47 30.07.2008 105,7 0,7 2,4 -1,0 6,6 1,3 -0,8 1,6 46 31.07.2008 106,6 0,9 0,0 1,0 0,0 1,2 0,8 1,8 48 01.08.2008 107,1 0,4 -0,3 0,7 0,0 1,2 0,6 1,5 49 04.08.2008 106,4 -0,7 -0,6 -0,2 0,2 1,2 -0.2 1,1 49 05.08.2008 110,1 3,3 4,5 0,1 22,2 1,3 0,1 2,3 47 06.08.2008 111,9 1,7 0,4 1,4 0,3 1.2 1,2 2.4 46 07.08.2008 111.1 -0,7 0,1 -0,7 O, 1 1,2 -0,6 1,8 48 08.08.2008 109,3 -1,6 1,0 -2,3 1,5 1,2 1,8 50 11.08.2008 109,9 0,5 1,4 -0,4 2,6 1,2 -0,4 1,0 47 12.08,2008 109,6 -0,2 -0,3 0,1 0,0 1,2 O, 0 1,1 45 13.08.2008 108,5 -1,0 -3,4 1,5 10,3 1,3 1,2 1,4 45 14.08.2008 107,7 -0,8 0,5 -1,2 0,5 1,2 -0,9 1,2 45 15.08.2008 106,7 . 0,9 0,6 -1,3 0,7 1,2 -1,1 1,7 45 18.08.2008 106,0 -0,6 -0,6 -0,1 0,2 1,2 -0,1 O, 8 45 19.08 104,1 -1,8 -4,2 1,3 16,0 1,3 1,0 1,7 50 20.08.2008 103,9 -0,2 0,8 -0,8 1,0 1,2 -0,6 0,9 48 21.08,2008 103,1 -O, 6 -1,5 0,4 1,8 3,2 0,3 1,1 49 22.08.2008 103,9 0,8 2,8 "1,2 9,0 1,3 -1,0 1,1 44,5 25.08.2008 104,4 0,5 0,0 0,6 0,0 1,2 0,5 0,5 46,5 26.08.2008 104,3 O, D O, O 0,0 0,0 1,2 0,0 0,8 46,5 27.08.2008 103,8 "0,5 "0,9 0,2 0,5 1.2 0,2 1,0 47 28.08.2008 105,1 1,2 2,7 -0,7 8.2 1,3 "0,6 1.4 46 29 08.2008 106.2 1,1 0,2 1,1 0,1 1,2 0.9 1.3 45 0109 2008 105,5 -0,7 -0,1 "O, 5 O, O 1,2 -0.4 -0,4 0,7 0,50 45 02.09.2008 108,1 2,4 2,7 0,5 8,5 1.3 0,4 0.4 1,5 0,50 46 03.09.2008 107,4 -0,7 -1,2 0,2 1,0 1,2 0,2 0,2 1.0 0,50 49 04.09.2008 105,3 -2,0 -2,7 0,0 6,1 1,3 0,0 0,0 1,6 0,50 50 05.09.2008 102,9 -2,3 -2,3 -0,6 4,3 1,3 -0,5 -0,5 2,0 0,50 53 08.09.2008 106,5 3,4 0,0 3,5 0,0 1,2 2,8 2,8 2,4 0,50 47 09.09.2008 106,7 0,1 5,1 -3,5 26,2 1,3 -2,7 -2,7 1.6 0,50 50 10.09.2008 106,6 -0,1 -1,3 0,9 1,1 1,2 0,7 0,7 1.8 0,50 49 11.09.2008 104.5 -2,0 -1,4 -1,0 1,5 1.2 -0,8 -0,8 1.6 0,50 49 12.09.2008 105,0 0,5 0.5 0,2 0,5 1,2 0,7 n7 1,4 0,50 51

15.09.2008 100,1 -5,0 -6,3 -0,3 37,9 1,3 -0.2 -0,2 5,0 0.50 70 16.09.2008 96,4 -3,8 -3,4 -1,2 10,4 1,3 -1.0 -1.0 5,1 0.50 80 17.09.2008 97,2 O, 8 -1,0 1,5 0,6 1,2 1,2 1.2 4,1 0,50 75 18,09.2008 96,9 -0,2 -1,1 0,6 0,8 1,2 0,5 0,5 3,3 0,50 60 19,09,2008 104.2 7,0 11,0 -1,0 125,4 1,5 -0,7 -0,7 6,4 0,50 58

243,9 Sum SARk -0,2 152,4 mod. SARI -3,0 Without 5.9.2008

supermodil SARk -2.4 Without 8.9.2008 and 19.9.2008 CARt

R mean return of BEINSUR during the estimation period -0,9 -0,19127

T is the number of days In the estimation period modified CARL 88 -11.3

1/T 0,011364 supermodiflad CARt

S standard error of the OLS markeßriI regressio n -8,5 1,235524

square of S standard er ror of the OLS market model regression 1,5265

sum of (Pmt-Rm squared) 243,3

Appendix D: "Old-mutual Formulas, share

price on c

Mt

hange

R-

change In Ab

AR., normal

R- - mean

So " SAR standard

AR Sum SAR V k

olumen of Min traded Tr

im -- 6YC adum Spre

DS ads

Variables 2008

Data 23 04 0 0 0 0 0 0

deviation 2

, . . 1 , , 0,0 1,7 0,0 2,10 7.61 120 Results and 24.04,2008 122,6 -0,8 0,8 -0,6 1,0 2,7 -0,4 21,33 120

25.04,2008 126,9 3,4 -1,4 3,6 1,4 1,7 2,2 14,02 123 Data of the 28.04.2008 128,8 1.4 1,5 1,6 2,8 1.7 1,0 12,97 118

29.04.2008 125,7 "2,4 2,6 -2,2 7,8 1,7 -1,3 14,92 120

event study 30.04.2008 127,2 1,2 -0,3 1,4 0,0 1,7 0,8 24,27 124 01.05.2008 126,8 -0,3 -1,7 -O, 1 2,3 1,7 -0,1 4,11 113

method 02.05.2008 133,8 5,2 0,2 5,4 0,1 1,7 3,2 15,62 108 06,05.2008 131,7 -1,6 -0,6 -1,4 0,2 1,7 -0,8 21,08 120 07.05.2008 126,3 -4,2 -1,1 -4,0 0,9 1,7 -2,4 21,37 122 08.05,2008 123,8 "2,1 0,0 -1,9 0,0 1,7 -1,1 16,79 122 09 05.2008 120,6 -2,6 -0,2 -2,4 0,0 1,7 -1,5 19,96 123 12.05.2008 120,6 0,0 0,2 0,2 0,2 1,7 0,1 14,33 123 13.05 . 2006 119,4 -1,0 -0,1 -0,8 0,0 1,7 -0,5 19,31 119 14.05.2008 119,4 0,0 0,0 0,2 0,0 1,7 0,1 32,59 119 15.05,2008 120,4 0,8 -0,2 1,0 0,0 1,7 0,6 20,98 117 16.05.2008 123,1 2,2 -1,5 2,3 1,7 1,7 1,4 19,86 117 19.05.2008 121,9 -0,9 -2,2 "0,7 4,1 1,7 -0,4 12,27 117 20 05.2008 119,5 "2,1 -1,1 -1,9 0,8 1,7 -1,1 16,87 117 21.05.2008 114,5 -4,4 -0,9 -4,2 0,5 1,7 -2,5 18,64 119 22.05.2008 111,9 -2,3 -0,3 -2,1 0,0 1,7 -1,3 18,30 126 23.05.2008 113,3 1,2 -0,2 1,4 0,0 1,7 0,9 15,73 131 27.05 112,3 -0,9 1,8 -0,7 3,8 1,7 -0,4 22,85 133 28.05.2008 114,0 1,5 0,7 1,7 0,7 1,7 1,0 27,83 125 29.05,2008 114,2 0,2 0,0 0,4 0,0 1,7 0,2 22,29 119 30.05.2008 117,8 3,0 "0,7 3,2 0,2 1,7 1,9 27,00 119 02.06.2008 115,0 -2,4 0,3 -2,2 0,2 1,7 -1,3 13,66 119 03.06,2008 115,7 0,6 -0,6 0,8 0,2 1,7 0,5 20,10 119 04,06.2008 117,1 1,2 -0,2 1,4 0,0 1,7 0,8 26,49 125 05.06.2008 117,4 0,2 -2,5 0,4 5,3 1,7 0,3 18,76 125 06,06.2008 114,5 -2,5 -0,5 -2,3 0,1 1,7 -1,4 14,21 130 09.06.2008 112,1 -2,1 -0,5 -1,9 0,1 1,7 -1,2 18,05 133 10.06.2008 110,1 -1,8 -1,7 -1,6 2,2 1,7 -1,0 18,76 135 11.06.2008 108,8 -1,2 1,6 -1,1 3,1 1,7 -0,6 20,78 132 12,06.2008 110,5 1,5 0,9 1,7 1,3 1,7 1,0 23,28 132 13.06.2008 109,2 -1,2 -0,4 -1,0 0,1 1,7 -0,6 18,58 129 16.06.2008 110,1 0,9 O, 0 1,1 0,0 1,7 O, 0 13,72 125 17.06.2008 110,8 0,6 -1,3 0,8 1,1 1,7 0,5 19,71 125 18.06.2008 108,2 -2,4 -0,9 -2,2 0,4 1,7 -1,3 18,02 125 19.06.2008 105,9 -2,2 -1,6 -2,0 2,0 1,7 -1,2 23,14 120 20 06.2008 104,8 -1,1 -1,3 -0,9 1,3 1,7 -0,5 16,25 127 23.06.2008 101,0 -3,7 -0,4 -3,5 0,0 1.7 -2,1 23,23 127 24.06.2008 98,0 -3,1 1,0 -2,9 1,5 1,7 -1,7 29,11 132 25.06.2008 95,4 -2,7 -3,4 -2,5 10,1 1,7 -1,5 33,01 132 26.06.2008 93,8 -1,7 "1,4 -1,5 1,4 1,7 -0,9 25,02 137 27.06 91,9 -2,1 . 0,2 -1,9 0,0 1,7 -1,1 30,37 140 30,06.2008 92,0 0,1 -2,9 0,3 6,8 1,7 0,2 26,26 135 01.07.2008 88,9 -3,5 0,1 -3,3 0,1 1,7 -2,0 27,38 137 02.07.2008 91,9 3,3 1,7 3, S 3,7 1,7 2,1 47,07 136 03.07.2008 91,8 "O, 1 -2,5 0,1 5,2 1,7 0,0 36,90 136 04.07.2008 89,4 -2,7 1,3 -2,5 2,2 1,7 -1,5 28,05 136 07.07.2008 90,4 1,1 -1,3 1,3 1,3 1,7 0,8 22,59 136 08.07.2008 88,1 "2,5 2,4 -2,4 6,7 1,7 -1,4 31,96 142 09.07.2008 91,4 3,5 -1,4 3,7 1,5 1,7 2,2 22.61 137 10,07.2008 90,6 -0,9 -3,0 -0,7 7,6 1,7 -0,4 22,55 137 11.07.2008 88,3 -2,6 1,0 -2,4 1,3 1,7 "1,5 35,59 140 14.07.2008 69,4 1,3 -3,0 1,5 8,0 1,7 0,9 20,27 135 15.07.2008 86,4 -3,5 0,2 -3,3 0,1 1,7 -2,0 29,06 140 16,07.2008 86,1 "0,4 2,9 -0,2 9,7 1,7 -0,1 41,35 143 17.07.2008 94,2 8,6 3,5 8,8 13,6 1,7 5,1 47,10 138 18.07.2008 98,4 4.3 0,2 4,5 0,1 1,7 2,7 47,69 138 21.07.2008 99,5 1,0 -0,2 1,2 0,0 1,7 0,7 29,07 138 22.07.2008 94,5 -5,3 3,6 -5,1 14,6 1,7 -3,0 31,57 138 23.07.2008 101,2 6.7 -1,3 6,8 1,3 1,7 4,1 38,67 131 24.07.2008 102,6 1,4 -4,1 1,6 15,4 1,7 0,9 28,62 131 25,07.2008 98,4 -4,3 "2,0 -4,1 3,4 1,7 -2,5 35,22 131 28.07.2008 95,7 -2,8 0,1 -2,6 0,1 1,7 -1,6 20,81 131 29.07 94,3 -1,5 2,4 -1,3 6,6 1,7 -0,8 28,03 131 30.07.2008 99,2 5,0 0,0 5,1 0,0 1,7 3,1 37,36 131 31.07.2008 97,8 . 1,4 -0,3 -1,3 0,0 1,7 -0,8 40,60 131 01.08.2008 99,9 2,1 -0,6 2,2 0,2 1,7 1,3 35,37 135 04.08.2008 99,3 "0,6 4,5 -0,4 22,2 1,7 -0,2 18,78 135 05,08.2008 104,1 4,6 0,4 4,8 0,3 1,7 2,9 24,27 130 06.08.2008 96,5 -7,8 0,1 -7,6 0,1 1,7 -4,6 61,50 135 07,08,2008 95,3 -1,3 1.0 -1,1 1,5 1,7 -0,7 43,71 135 08.08.2008 95,3 0,0 1,4 0,2 2,6 1,7 0,1 42,78 135 11,08,2008 100,0 4,7 -0,3 4,9 0,0 1,7 2,9 35,15 135 12.08.2008 99,7 -0,3 -3,4 -0,1 10,3 1,7 0,0 38,45 135 13.08.2008 97,4 -2,4 0,5 -2,2 0,5 1,7 -1,3 20,24 135 14.08.2008 96.2 -1,3 0,6 -1,1 0,7 1,7 -0,7 48,06 135 15.08.2008 96,0 -0,1 . 0,6 0,1 0,2 1,7 0,0 42,35 135 18.08.2008 96,0 0,0 -4,2 0,2 16,0 1,7 0,1 23,64 135 19.08.2008 92,4 -4,0 0,8 -3,8 1,0 1,7 -2,3 30,11 135 20.08,2008 93,8 1,5 -1,5 1,7 1,8 1,7 1,0 40,56 135 21.08.2008 93,4 -0,4 2,8 -0,2 9,0 1,7 -0,1 19,49 135 22.08.2008 96,5 3,2 0,0 3,4 0,0 1,7 2,1 21,63 133 2608.2008 94,6 -2,0 0,0 -1,8 0,0 1,7 -1,1 23,36 135 27.08,2008 95,1 0,5 -0,9 0,7 0,5 1,7 0,4 20,12 135 28,08.2008 98.4 3,3 2,7 3,5 8,2 1,7 2,1 25,59 134 29 08 2008 98.1 -0,2 0.2 -0,1 0.1 1,7 0,0 20 80 133 01.09.2008 98,5 0,4 -0,1 0,6 0,0 1,7 0,3 0,3 13,25 7,61 133 02.09.2008 101,1 2,6 2,7 2,8 as 1,7 1,6 1,6 26,62 7,61 123 03.09.2008 98,5 -2,6 -1,2 -2,4 1,0 1,7 -1,5 -1,5 19,57 7,61 125 04,09.2008 98,2 -0,3 -2,7 -0,1 6,1 1,7 -0,1 -0,1 32,93 7.61 130 05.09.2008 94,9 -3,5 -2,3 -3,3 4,3 1,7 -2,0 "2,0 24,74 7,61 133 08,09.2008 99,6 4,8 0,0 4,9 0,0 1,7 3,0 3,0 12,67 7,61 123 09.09.2008 101,5 1,9 5,1 2,0 28,2 1,8 1,2 1,2 26,08 7,61 128 10,09.2008 97,0 -4,7 -1,3 -4,5 1,1 1,7 -2,7 -2,7 44,94 7,61 132 11.09.2008 94,2 -3,0 -1,4 -2,8 1,5 1,7 -1,7 -1,7 25.76 7,61 132 12.09.2008 fi

971 3,0 0,5 32 0,5 17 1,9 1,9 47,56 7,61 142

15.09.2008 89,9 -8,0 -6,3 -7,8 37,9 1,8 "4,4 "4,4 43,70 7,61 167 16.09.2008 85,3 -5,4 "3,4 "5,2 10,4 1,7 "3,1 -3,1 47,73 7,61 205 17.09.2008 84,5 "0,9 -1,0 -0,7 0,6 1,7 -0,4 -0,4 48,57 7,61 205 18.09.2008 77,2 -9,4 -1,1 -9,2 0,8 1,7 -5,5 -5,5 46,09 7,61 205 19.09.2008 85,7 9,9 11,0 10,1 125,4 2,0 4,9 4,9 57,82 7,61 195

243,9 Sum SARK -8,3 2333,96 mod. SARI, -11,3 Without 5.9.2008

supermodd SARK -16,2 without 8,9.2008 and 19.9.2008 CARt

R mean return of BEINSUR during the estimation period -32,2 -0,19127

T is the number of days in the estimation period modifWartss CARt 88 -42,2 irr

0,011364 S standard err or of the OLS marke t i7ddl

regression sup. rmodifl

-58,4 .d CARt

1,655075 square of S standard error of the OLS market model regression

2,7393 sum of (Rml-Rm squared)

243.3

Appendix D: Formulas, Variables, Results and Data of the event study method

23.04.2008 243,7 0,0 0.0 0.0 0.0 1,7 0.0 5,89 2,08 N/A 24.04.2008 246,4 1,1 0,8 1,3 1,0 1,7 0,8 3,95 25.04.2008 252,8 2,5 -1,4 2,7 1,4 1,7 1,6 3,79 28.04.2008 251,1 -0,7 1,5 -0,5 2,8 1,7 -0,3 3,31 29.04.2008 248,2 -1,2 2,6 -1,0 7,8 1,7 -0,6 4,02 30.04.2008 249,6 O, 6 -0,3 O, 8 0.0 1,7 O, 5 6,31 01.05.2008 250.5 0,4 -1,7 0,6 2,3 1,7 0,4 2,05 02.05.2008 262,3 4,5 0,2 4,7 0,1 1,7 2,8 5,43 06.05.2008 260,2 -0,8 -0,6 -0,6 0,2 1,7 -0,4 5,12 07.05.2008 260,9 0,3 -1,1 0,5 0,9 1,7 0,3 6,58 08.05.2008 257,4 -1,4 0,0 -1,2 0.0 1,7 -0,7 4,50 09.05.2008 254,9 -1,0 -0,2 -0,8 0,0 1,7 -0,5 2,86 12.05.2008 261,0 2,3 0,2 2,5 0,2 1,7 1,5 3,81 13.05.2008 255,2 -2.2 . 0,1 -2,1 0,0 1,7 -1,2 4,48 14.05.2008 255,3 0,0 0,0 0,2 0,0 1,7 0,1 3,17 15.05.2008 256,0 0,3 -0,2 0,5 0,0 1,7 0,3 3,53 16.05.2008 258,8 1,1 -1,5 1,3 1,7 1,7 0,8 4,90 19.05.2008 256,6 -0,9 -2,2 -0,6 4,1 1,7 -0,4 2,50 20.05.2008 254,1 -1,0 -1,1 -0,8 0,8 1,7 -0,5 4,17 21.05.2008 242,9 -4,6 -0,9 -4,4 0,5 1,7 -2,6 9,04 22.05.2008 239,0 -1,6 -0,3 -1,4 0,0 1,7 -0,9 6,21 23.05.2008 240,9 0,8 -0,2 1,0 0,0 1,7 0,6 4,99 27.05.2008 238,7 -0,9 1,8 -0,8 3,8 1,7 -0,5 4,72 28.05.2008 243,0 1,8 0,7 1,9 0,7 1,7 1,2 3,12 29.05.2008 241,2 -0,8 0,0 -0,6 0,0 1,7 -0,3 4,58 30.05.2008 251,0 3,9 -0,7 4,1 0,2 1,7 2,5 15,71 02.06.2008 249,1 -0,7 0,3 -0,6 0,2 1,7 -0,3 11,47 03.06.2008 256,0 2,7 -O, 6 2,9 0,2 1,7 1,7 7,34 04.06.2008 266,4 3,9 -0,2 4,1 0,0 1,7 2,5 13,11 05.06.2008 266,2 -0,1 -2,5 0,2 5,3 1,7 0,1 8,99 06.06.2008 251,1 -6,0 -0,5 -5,8 0,1 1,7 -3,5 12,71 09.06.2008 247,5 -1,5 -0,5 -1,3 0,1 1,7 -O, 8 5,35 10.06.2008 244,5 -1,2 -3,7 -1,0 2,2 1,7 -0,6 7,40 11.06.2008 237,9 -2,8 1,6 -2,6 3,1 1,7 -1,6 9,09 12.06.2008 239,7 0,7 0,9 0,9 1,3 1,7 0,5 16,21 13.06.2008 240.5 0,3 -0,4 0,5 0,1 1,7 0,3 4,55 16.06.2008 238,6 -0,8 0,0 -0,6 0,0 1.7 -0,4 10,48 17.06.2008 242,9 1,8 -1,3 2,0 1,1 1,7 1,2 5,09 18.06.2008 239,3 -1 -0,9 -1,3 0,4 1,7 -0,8 7,58 19.06.2008 231,6 -3,3 -1,6 -3,1 2,0 1,7 -1,9 11,11 20.06.2008 225,9 -2,5 -1,3 -2,3 1,3 1,7 -1,4 11,18 23.06.2008 222,9 -1,3 -0,4 -1,1 0,0 1,7 -0,7 4,27 24.06.2008 220,8 -1,0 1,0 -0,8 1,5 1,7 -0,5 10.71 25.06.2008 222,6 0,8 -3,4 1,1 10,1 1,7 0,7 4,00 26.06.2008 216.9 -2,6 -1,4 -2,4 1,4 1.7 -1,4 10,06 27.06.2008 211.1 -2,8 -0,2 -2,6 0,0 1,7 -1,5 7,76 30.06.2008 208,6 -1,2 -2,8 -0,9 6,8 1,7 -0,5 5,07 DI . 07.2008 202,2 -3,2 0,1 -3,0 0,1 1,7 -1,8 9,05 02.07.2008 206,4 2,0 1,7 2.2 3,7 1,7 1,3 9,61 03.07.2008 208,0 as -2,5 1,1 5,2 1,7 0,6 7,81 04.07.2008 203,4 -2,3 1,3 -2,1 2,2 1,7 -1,3 5,61 07.07.2008 206,0 1,2 -1,3 1,5 1,3 1,7 0,9 5,92 08.07.2008 207,1 0,6 2,4 0,7 6,7 1,7 0,4 9,75 09.07.2008 214,6 3,5 -1,4 3,7 1,5 1,7 2,2 10,25 10.07.2008 209,9 -2,2 -3,0 -2,0 7,6 1,7 -1,2 5,89 11.07.2008 202,8 -3,5 1,0 -3,3 1,3 1,7 -2,0 7,72 14.07.2008 206,5 3,8 -3,0 2,1 8,0 1,7 1,2 5,30 15.07.2008 198,2 -4,2 0,2 -4,0 0,1 1,7 -2,4 13,84 16.07.2008 196,1 -1.1 2,9 -1,0 9,7 1,7 -0,6 10,43 17.07.2008 206,3 5,0 3,5 5,1 13,6 1,7 3,0 10,49 18.07.2008 212,8 3,0 0,2 3,2 0,1 1,7 1,9 8,28 21.07.2008 216,7 1,8 -0,2 2,0 0,0 1,7 1,2 6,61 22.07.2008 214,4 -1,1 3,6 -1,0 14,6 1,7 -0,6 6,65 23.07 228,0 6,0 -1,3 6,2 1,3 1,7 3,7 14,23 24.07.2008 231,7 1,6 -4,1 1,9 15,4 1,7 1,1 10,73 25.07.2008 219.8 -5,4 -2,0 -5.1 3,4 1,7 -3,1 12,09 28.07.2008 215,7 .1 ,9 0,1 -1,7 0,1 1,7 -1,0 4,39 29.07.2008 213,8 -0,9 2,4 -0,8 6,6 1,7 -0,5 5,83 30.07.2008 223,3 4,3 0,0 4,4 0,0 1,7 2,7 5,75 31.07.2008 224,9 0,8 -0,3 0,9 0,0 1,7 0,6 7.47 01.08.2008 225,8 0,4 -0,6 0,6 0,2 1,7 0,3 6,38 04.08.2008 223,4 -1,0 4,5 . 1,0 22,2 1,7 . 0,6 3,24 05.08.2008 240,1 7,0 0,4 7,1 0,3 1,7 4,3 8,50 06.08.2008 238,3 -0,8 0.1 -0,6 0,1 1,7 -0,4 12,46 07.08.2008 243,1 1,2 1,0 1,3 1,5 1,7 as 13,58 CIS. 08.2008 248,8 3,1 1,4 3,2 2,6 1,7 1.9 7,14 13.08.2008 250,7 0,7 -0,3 0,9 0,0 1,7 0,6 5,39 12.082008 249,7 -0,4 -3,4 -0,1 10,3 1,7 -0.1 6,76 23.0a. 200a 242,0 -3,1 0,5 -3,0 0,5 1,7 -1,8 7,17 34.08.2008 241,8 -0.1 0,6 0,1 0,7 1,7 0,0 4,84 15.08.2008 242,8 0,4 -0,6 0,6 0,2 1,7 0,4 4,44 18.08.2008 239,9 . 1,2 -4,2 -0,9 16,0 1,7 -0,5 3,18 19.08.2008 228,7 -4,9 0,8 -4,7 1,0 1,7 -2,8 5,78 20.08 . 2008 229,4 0,3 -1,5 0,5 1,8 1,7 0,3 4,70 21.08.2008 231,4 0.8 2,8 0,9 9,0 1,7 0.6 5,83 22.08.2008 241,8 4,3 0.0 4,5 0,0 1,7 2,7 6,85 26.08.2008 241.4 -O, 1 0,0 0,0 0,0 1,7 0,0 6,32 27.0 242,6 0,5 -0,9 0,7 0,5 1,7 0,4 4,79 28.08 251,4 3,5 2,7 3,6 8,2 1.7 2,3 6,22 79 05 2008 251,8 0,2 0.2 0,3 0.1 1,7 0,2 9 53 01.09.2008 252,1 O, 1 -0,1 0,3 0.0 1.7 0,2 0,2 3,65 2,08 02.09.2008 258,8 2,6 2,7 2,7 8,5 1,7 1,6 1,6 6,20 2,08 03.09.2008 252,9 -2,4 -1,2 -2,1 1,0 1.7 -1,3 -1,3 5,54 2,08 04.09.2008 245,8 -2,9 -2,7 . 2,6 6,1 1,7 -1,6 -1,6 8,99 2,08 05.09.2008 234,0 -5,0 -2,3 -4,8 4,3 3,7 -2,8 -2,8 9,26 2,08 08.09.2008 258,3 9,4 0,0 9,6 0,0 1,7 5,8 5,8 5,12 2,08 09.09.2008 255,3 -1,2 5,1 -1.1 28,2 1.8 -0.6 -0.6 10,62 2,08 10.09.2008 244,2 -4,6 -1,3 -4,3 1,1 1,7 -2,6 -2.6 10,11 2,08 31.09.2008 240,3 -1.6 -3,4 -1,4 1,5 1,7 -O, 8 -0,8 5,89 2,08 12.09.2008 239,9 -0,2 0,5 0,0 0,5 17 0,0 0,0 5,10 2,08

15092008 232,4 -3,2 -6,3 -2,8 37,9 1.8 -1,6 10,49 2,08 16.09.2008 230,5 -O, 8 -3,4 -0,5 10,4 1,7 -0.3 -0,3 18,73 2,08 17.09.2008 236,5 2,5 -1,0 2,7 0,6 1,7 1,6 1,6 14,19 2,08 18.09.2008 230,6 -2.6 -1,1 -2,4 0,8 1,7 -1,4 -1,4 10,39 2,08 19.09.2008 260,4 11,4 11,0 11,3 125,4 2,0 5.5 5,5 22,92 2,08

243,9 Sum SARk 1,7 637,91 mod . SARI -4,1 Without 5.9.2008

supermodit SAR k -9,6 Without 6.9.2008 and 19.9.2008 CARL

R mean return of BEINSUR during the estimation period 6,5 -0,19127

T is the number of days in the estimation period modifiziert.. CARt 86 -15.3

1R 0,011364

S standard error of the OLS market37igl regress sup. mtodtfi. d CARt

ion . 34,7 1,653148

square of S standard error of the OLS market mod el regression 2,7329

sum of (Rmt-R m squared) 243.3

Appendix D: ä ""e #t nnöv . -Re Formulas, share, P_ R- AR,, R., - SD,, SAR =ARS uMSAR -

Variables, Date --

P, I. on 'Data th.

h. rig. h.. 9.1. A b. -I -. an st andard I SD., k traded Trade Sp r*ads

Results and 25.04 2008 33,4872 0 0 0 0,0 1,4 O 0,43 0 21 47 28.04.2008 33,7627 0,8 0,8 0,6 1,0 1,4 0,4 0 67

,

Data of the 29.04.2008 33,9913 0,7 -1,4 1,3 1,4 1,4 0,9 ,

0 60 44 4 30.04.2008 34,4694 1,4 1,5 0,9 2,8 1,4 0,7 ,

0 66 6

event $6 .4 02,05.2008 35,7963 3,7 2,6 2,8 7,8 1,4 2,0 ,

1 00 47

05.05,2008 35,7684 -0,1 -0,3 0.2 0,0 1.4 0 1 , 0 66

43

method 06.05.2008 36,0284 0,7 -2,7 1,5 2,3 1.4

, 1 1 , 43

07.05.2008 32,9286 -9,4 0,2 -9,4 0,1 1,4 ,

-6 6 1,60 48

08.05.2008 33,4893 1,7 -0,6 2,0 0,2 1 4 ,

1 4 1,24 47

09.05.2008 33,7938 0,9 -1,1 1,5 0,9 ,

1 4 ,

1 0 1,16 51

12,05.2008 34,0057 0,6 O. 0 0,8 0,0 ,

1 4 ,

0 5 0,87 51

13.05,2008 33,9831 -0,1 -0,2 0,1 0,0 ,

1,4 ,

0 1 0,24 0

50

14.05.2008 34,3273 1,0 0,2 1,0 0,2 1.4 ,

0 7 , 68 48

15.05.2008 34,2112 -0,3 -0,1 -0,2 0,0 1 4 ,

-0 1 0,52 50

16.05.2008 34,7429 1,5 O, 0 1,6 0,0 ,

1 4 ,

1 2 0,32 49

19.05.2008 34,8396 0,3 -0,2 0,5 0,0 ,

1,4 ,

0 3 0.64 0

47

20.05,2008 34,8289 0,0 0,7 1,7 1,4 ,

0,5 , 39 0 52

43

2105200 34,6845 -0,4 -2,2 0,5 4,1 1,4 0,4 , 0 52

47

22,05,2008 34,8069 0,4 -1.1 0,9 0,8 1,4 0,6 . 0 64

51

23,05.2008 34,0657 -2,2 -0,9 -1,7 0,5 1,4 -1,2 ,

0 82 59

26,05.2008 34,1066 O, 3 -0,3 0,3 0,0 1,4 0,2 , 0 20

66

27.05.2008 33,9759 -0,4 -0,2 -0,2 0,0 1.4 -0,1 ,

0 80 6G

28.05.2008 34,3112 1,0 1,8 0,4 3,8 1,4 0,3 , 0 62

68 6 29.05.2008 34,8196 1,5 0,7 1,3 0,7 1,4 0,9 ,

0 70 0

30.05.2008 35,4338 1,7 0,0 1,9 0,0 1,4 1,3 , 2 66

57 02.06.2008 35,5975 0,5 -0,7 0,8 0,2 1,4 0,6 ,

0 70 55

03.06.2008 35,3337 -0,7 0,3 -0,7 0,2 1,4 -0,5 ,

0 43 58

04. OG, 2008 34,8125 -1,5 -0,6 -1,1 0,2 1,4 -0,8 .

0 79 60

05.06.2008 35,0679 0,7 -0,2 0,9 0,0 1,4 0,7 , 0 74

62

06 06.2008 35,1922 0,4 -2,5 1,4 5,3 1,4 1,0 , 0 49

63 66 09.06.2008 34,9321 -0,7 -0,5 -0,4 0,1 1,4 -0,3

, 0 54 68 10.06.2008 33,9431 -2,9 -0,5 -2,6 0,1 1.4 -1,8

, 0 89 6 11 06.200a 33,8741 -0,2 -1,7 0,5 2,2 1,4 0,4 , 0 80

8

12,06,2008 33,954 0,2 1,6 -0,2 3,1 1,4 -0,2 ,

0 70 64

13.06.2008 34,6737 2,1 0,9 1,8 1,3 1,4 1,3 , 42 0

63

16.0"i . 20011 34,4776 -0,6 -0,4 -0,3 0,1 1,4 -0,2

, 0,21

59 58 17.06,2008 34,7556 0,8 0,0 0,9 0,0 1,4 0,7 0 36 56 18.06.2008 34,1682 -1,7 -1,3 -1,1 1,1 1,4 -0,8

, 0 48 59 39.06.2008 34,0789 -0,3 -0,9 0,2 0,4 1,4 0,1 , 0 36

20.06.2008 34,1834 0,3 -1,6 1,0 2,0 1,4 0,7 ,

0 82 62

23.06.2008 33,2044 -2,9 -1,3 -2,3 1,3 1,4 -1,6 ,

0 47 68

24.06.2008 33,1157 -0,3 -0,4 0,0 0,0 1,4 0,0 , 0 93

67

25.06.2008 32,5472 -1.7 1.0 -2,0 IN 1,4 -1 4 ,

0 85 73

26.06.2008 31,8316 -2,2 . 3,4 -0,9 10,1 1,4 ,

-0,6 ,

0 91 67

27.06.2008 30,8743 -3,1 -1,4 -2,5 1,4 1,4 -1,7 ,

0,66 69 72 30.06.2008 31,1238 0,8 -0,2 1,0 0,0 1,4 0,7 0 54 69 01.07.2008 30,6645 -1,5 -2,8 -0,3 6,8 1,4 -0.2

, 0,62 73 02.07.2008 32,1022 4,5 0,1 4,6 0,1 1,4 3,2 1.16 73 03.07.2008 32,1795 0,2 1,7 -0,3 3,7 1,4 -0,2 1,14 73 04.07.2008 31,8296 -1,1 -2,5 0,0 5,2 1,4 0.0 0,35 73 07.07.2008 32,0047 0,5 1,3 0,2 2,2 1,4 0,1 0,49 72 08.07.2008 31,3968 -1,9 -1,3 -1,3 1,3 1,4 -0,9 0,49 77 09.07.2008 32,2998 2,8 2,4 2,0 6,7 1,4 1,4 0,39 71 10.07.2008 31,8003 -1,6 -1,4 -0,9 1,5 1,4 -0,6 0,49 72 11.07,2008 31,2535 -1,7 -3,0 -0,5 7,6 1,4 -0,4 0,48 73 14 07.2008 30,8363 -1,4 1,0 -1,6 1,3 1.4 -1.1 0,61 69 15.07.2008 29,8821 -3,2 -3,0 -1,9 8,0 1,4 -1,3 0,74 72 16.07.2008 29,5379 0,2 -1,1 0,1 1,4 -0,8 0,73 70 17.07.2008 30,1589 2,1 2,9 1,1 9,7 1.4 0,7 0,78 64 18,07.2008 30,9091 2,4 3,5 1,2 13,6 1,4 0,8 0,60 63 21.07.2008 31,4598 1,8 0,2 1,8 0,1 1,4 1,3 0,51 60 22.07.2008 31,3137 -0,5 -0,2 -0,3 0,0 1,4 -0,2 0,78 61 23.07.2008 32,5041 3,7 3,6 2,4 14,6 1.5 1,7 0,94 57 24.07.2008 32,4326 -0,2 -1,3 0,4 1.3 1.4 0,3 0 59 57 25.07.2008 28,8549 -12,4 -4,1 -10,7 15,4 1,5 -7,4

, 4,07 59 28.07.2008 29,7168 2,9 -2,0 3,8 3,4 1,4 2,7 0,79 58 29.07.2008 29,766 0,2 0,1 0,2 0,1 1,4 0,2 0,58 58 30.07,2008 30,0472 0,9 2,4 0,1 G, 6 1,4 0,1 0,60 59 31.07.2008 30,6855 2,1 0,0 2,2 0,0 1,4 1,6 0,67 61 01,08.2008 30,2977 -1,3 -0,3 -1,1 0,0 1,4 -0,7 0 36 64 04,08.2008 29,9397 -1,2 -0,6 -0,8 0,2 1.4 -0,6

, 0 45 65 05.08.2008 30,8306 2,9 4,5 1,3 22,2 1,5 0,9

, 47 0 61 06.08.2008 31,2377 1,3 0,4 1,3 0,3 1,4 0,9

, 0 57 62 07.08.2008 30,4268 -2,7 0,1 -2,6 0,1 1,4 -1,8

, 0 79 63 08.08.2008 29,7133 -2,4 1,0 -2,7 1,5 1,4 -1,9

, 0 82 64 11.08.2008 29,7539 0,1 1,4 -0.3 2,6 1,4 -0,2

, 0 98 61 12.08.2008 29,9498 0,7 -0,3 0,9 0,0 1,4 0,6

, 0 55 59 13.08.2008 28,8485 . 3,8 -3,4 -2,4 10,3 1,4 -1,7

, 1 08 60 14.08.2008 28,7722 -0,3 0,5 -0,3 0,5 1.4 -0,2

, 57 0 58 15.08.2008 28,9092 0,5 0.6 0,4 0,7 1.4 0,2 ,

0 54 58 18,08.2008 28,6435 . 0,9 -0,6 -0,6 0,2 1.4 -0,4 ,

0 40 58 19.08.2008 28,3545 -1,0 -4,2 0,7 16,0 1,5 0 5 , 0 67 64 20.08.2008 28,2095 -0,5 0,8 -0,7 1,0 1,4 .

-0,5 .

0 49 63 21.08.2008 28,1558 -0,2 -1,5 O, s 1,8 1,4 0,4 ,

0 80 65 22.08.2008 29,1079 3,3 2,8 2,3 9,0 1,4 1 6 , 0 63 61 5 25.08.2008 29,3663 0,9 0,0 1,0 0,0 1,4

, 0 7 ,

0 38 ,

1 26.08.2008 28,9235 . 1,5 0,0 -1,4 0,0 1,4 ,

-1 0 , 6 ,5 27.08.2008 28,4478 -1,7 -0,9 -1,2 0,5 1,4

, -0 9

0,42 G3,5

28.08.2008 28,9804 1,8 2,7 0,9 8,2 1,4 ,

0 7 0,39 65

29 O8 2008 29,3695 1,3 0.2 1,4 0.1 1,4 , 0 1

0.39 63

01.09.2008 29,0026 -1,3 -0,1 -1,1 0,0 1,4 ,

-0 8 -0 8 0 45 62

02.09.2008 29,839 2,8 2,7 1,9 8,5 1,4 , 1 3

, 1 3

0,32 0,21 62

03.09.2008 29,5599 -0,9 -1,2 . 0,4 1,0 1,4 ,

-0 3 ,

-0 3 0,66 0,21 59

04,09.2008 28,842 -2,5 -2,7 -1,4 6,1 1,4 ,

-1 0 ,

-1 0 0,69 0,21 62

05.09,2008 27,9496 -3,2 -2,3 -2,2 4,3 1,4 , .1 6

, -1 6

0,80 0,21 66

08.09.2008 29,1284 4,0 0,0 4,2 0,0 1,4 , 3 0 ,

3 0 0,63 0.21 69

09.09,2008 29,1511 0,1 5,1 -1,7 28,2 1,5 ,

-1 2 , -1 2

0,62 0,21 61

10.09.2008 28,1422 -3,6 -1,3 -3,0 1,1 1 4 ,

-2 1 ,

2 0,61 0,21 64

11,09.2008 27,5379 -2,2 -1.4 -1.5 1.5 .

1,4 .

-1 1 - ,1 -1 1

0,96 0,21 63 17 no 70n8 77.2f+ -12 0 5 .1 3 0 5 1 4

, , 0,68 0,21 63 wr- , , , , -0,9 -0,9 0,60 021 68

1509.2008 25,9626 -4,8 -6,3 -2,3 37,9 1,5 -1,5 -1 5

ýr .. r.

K

16.09.2008 24,946 -4,1 -3,4 -2,7 10,4 1,4 -1,8 ,

-1 8 1,08 0,21 91

17 09.2008 24,4763 -1,9 -1,0 -1,4 0,6 1,4 -1,0 ,

-1 0 1,40 0

0,21 101

18,09.2008 23,9431 -2,2 -1,1 -1,7 0,8 1,4 -1,2 ,

-1 2 , 99 0

0,21 95 19 09.2008 25,7083 6,9 11,0 2,8 125,4 1,7 1.6

, 16 , 91

1 0,21 85

243,9 Sum SARk 8,5 9 C'

63,07 0,21 75

mod. SARI -11,5 Without 5.9.2008 supermoditSARk

CA -13,1 Without 8.9.2008 and 19,9.2008

Rt R mean return of BEINSUR during the estimation period -33,0

-0.19127 T Is the number of days In the estimation period modlflzl. rtes CARt

88 -42.9 1/1 0.011364

S standard error of the O supormodifl. d CARt

LS marke +7,3I regression -47 2 1,401592 ,

square of S sta ndard error of the OLS market model regression 1,9645

sum of (Rmt-Rm squared) 243.3

25.04.2008 128.99 0 28.04.2008 130,07 0,83 29.04.2008 128,31 -1,37 30.04.2008 130,25 1,49 02.05.2008 133,73 2,60 05.05.2008 133,28 -0,34 06.05.2008 131,03 -1,72 07.05.2008 131,27 0,18 08.05.2008 130,45 -0,63 09.05.2008 128,98 -1,14 12.05.2008 128,93 -0,04 13.05.2008 128,65 -0,22 14.05.2008 128.96 0,24 15.05.2008 125,84 -0,09 16.05.2008 128,87 0,02 19.05.2008 128,63 -0,19 20.05.2008 126.75 -1,48 21.05.2008 124,01 -2,21 22.05.2008 122,71 -1,06 23.05.2008 121,58 -0,93 26.05.2008 121,23 -0,29 27.05.2008 120,98 -0,21 28.05.2008 123,16 1.77

29.05.2008 123,97 0,65 30.05.2008 123,97 0,00 02.06.2008 123,12 -0,69 03.06.2008 123,43 0,25 04.06.2008 122,66 -0,63 05.06.2008 122,42 -0,20 06.06.2008 119,45 -2,49 09.06.2008 118,83 -0,52 10.06.2008 118,2 -0,53 11.06.2008 116,27 -1.66 12.06.2008 118.14 1,58 13.06.2008 119,27 0,95 16.06.2008 118,74 -0,45 17.06.2008 118,74 0,00 18.06.2008 117,27 -1,25 19.06.2008 116,28 -0.85 20.06.2008 114.46 -1,59 23.06.2008 112,98 -1,31 24.06.2008 112,56 -0,37 25.06.2008 113,74 1,04 26.06.2008 110,04 -3,36 27.06.2008 108,56 -1,36 30.06.2008 108,29 -0,25 01.07.2008 105,35 -2,79 02.07.2008 105,42 0,07 03.07.2008 107,27 1.72 04.07.2008 104.69 -2,46 07.07.2008 106,05 1,28 08.07.2008 104,65 -1,34 09.07.2008 107,23 2,41 10.07.2008 105,72 -1,43 11.07.2008 102,69 -2.95 14.07.2008 103,68 0,95 15.07.2008 100,64 -3,02 16.07.2008 100,81 0,17 17.07.2008 103,85 2.93 18.07.2008 107,62 3,50 21.07.2008 107,81 0,18 22.07.2008 107,62 -0,18 23.07.2008 111,68 3,64 24.07.2008 110,22 -1,32 25.07.2008 105,86 -4,12 28.07.2008 103,76 -2,02 29.07.2008 103,89 0,13 30.07.2008 106,43 2,39 31.07.2008 106,38 -0,05 01.08.2008 106,08 -0,28 04.08.2008 105.44 -0.61 05.08.2008 110,43 4,52 06.08.2008 110,85 0,38 07.08.2008 110,92 0,06 08.08.2008 112,06 1.02 11.08.2008 113,68 1,43 12.08.2008 113,3 -0,34 13.08.2008 109,57 -3,40 14.08.2008 110,16 0,54 15.08.2008 110,87 0,64 18.08.2008 110,18 -0,63 19.08.2008 105,75 -4.19 20.08.2008 106,63 0,83 21.08.2008 105,03 -1,52 22.08.2008 108,06 2,80 25.08.2008 108,01 -0,05 26.08.2008 108,04 0,03 27.08.2008 107,05 -0,92 28.08.2008 109,98 2,66 29.08.2008 110,15 0,15 01.09.2008 110,08 -0,06 02.09.2008 113.16 2.72 03.09.2008 111,83 -1,19 04.09.2008 108,93 -2,66 05.09.2008 106.51 -2,27 08.09.2008 106,51 0,00 09.09.2008 112,26 5,12 10.09.2008 110,87 -1.25 11.09.2008 109,34 -1,40 12-09.2008 109.91 0.52

15.09.2008 103,35 -6,35 16.09.2008 99,94 -3,41 17.09.2008 98,99 -0,96 18.09.2008 97.91 -1,10 19.09.2008 110,02 11.01

Appendix D: Formulas, Variables, Results and Data of the event study method

374

Appendix E: Document Modified Equity-ratio Analysis

for measuring

Research period: 1.1.2008 to 31.12.2008

the modified equity-ratio

Determination of the modified equity-ratio as of January Ist, 2008

Equity stated in the financial statements as of 31.12.2007

excluding reserves for currency translation

excluding own shares for share based payment schemes and buy back programms

excluding insurance related changes of equity (pension plans) and unspecified reserves

excluding dividends paid and distributions to shareholders

excluding issues of new share capital

excluding for tax effects

excluding for transactions among subsiduaries

excluding shadow accounting

excluding deeply subordinated debt

Modified equity as of 31.12.2007

Total assets as of 31.12.2007

Modified equity-ratio as of 31.12.2007

B. Modified equity as of 1.1.2008 (identical to modified equity as of 31.12.2007)

excluding reserves for currency translation

excluding own shares for share based payment schemes and buy back programms

excluding insurance related changes of equity (pension plans) and unspecified reserves

excluding dividends paid and distributions to shareholders

excluding issues of new share capital

excluding for tax effects

excluding for transactions among subsiduaries

excluding shadow accounting

excluding deeply subordinated debt

Not excluded:

unrealized gains and losses from fair market value measurement of Financial instruments

net income of the period 2008

Modified equity as of 31.12.2008

Total assets as of 31.12.2008

Modified equity-ratio as of 31.12.2008

C. Change of modified equity-ratio due to the efQjýf the financial crisis in percent

Modified equity as of 31.12.2008 - modified equity as of 31.12.2007 / modified equity as of 31.12.2008

Appendix F: Clustering listed European insurance groups 2007

Analysis 2007 data

N

N Z

u A

C

C C O t

C

U m L 0 C A

M A

0

7 P O

C

O cl C A

C U

-10

-20

-30

-40

ý* Reihe l ,ý

376

Risk Disclosure Performance in group financial statements 2007

Appendix G: Clustering European insurance groups 2008

Analysis of data 2008

30

u 20

C

CC

L 10

ö ü

0o m C a

-10

ö, -20 c V

ý(

-4I!

377

Risk Disclosure Performance in the group financial statements 2008

Appendix H: Guidelines for structured interviews with experts

Guideline for a structured interview

Thank you for participating in this interview. Its focus is to explain, question and or verify the statistical results of the research performed by Ansgar Völker about the risk disclosure of international insurance groups and the capital market reaction. This interview is structured using questions about the three variables and their relationship that are analyzed in this research project.

1. Negative effects on equity ratios as a result of the financial crisis 2007-2009 2. Risk disclosure in group financial statements as of 2007 and 2008 3. Reaction of the capital market in a moment of financial distress

Data from following international insurance groups have been used: Allianz, Aegon, Aviva, AXA, CNP Assurance, Generali, Hannover Rück, ING, Legal&General, MunichRe, Old Mutual, Standard Life, Zurich

The interview can proceed in English or German, person to person, by phone or in writing. The English manuscript of the interview will be sent to you in draft form. Only content that is agreed upon will be published. You can opt for anonymity in the research paper.

Effect on equity ratio: " Do you think that the insurance industry has been hit hard by the financial crisis in

comparison to other industries?

" Which one of the thirteen insurance groups analyzed has been affected most negatively by the crisis?

" Are you aware of confounding negative effects that lowered the equity ratio of the analyzed insurance groups and that were not based on impairments or lower market valuation of financial instruments?

" Can you name international insurance groups that have received financial aid from governments during the financial crisis?

Risk disclosure behaviour: " Did you get a clear picture about how the insurance companies have been affected by

the crisis by reading their group financial statements? Where you able to predict the effect the crisis had on an insurance group based on those annual information? What information about the insurance group that proved important later on did you learn from the group financial statements first?

" Are you supportive of the theory that timely and extended disclosure by international insurance groups increases trust that investors have in that group?

" Do you think that managers in the insurance industry can influence the extent of risk disclosure of the groups they lead in the financial statements? What might restrict them?

" Is it your impression that the quality of risk disclosure about troublesome financial instruments (such as: ABS, CDO, RMBS, CMBS) has improved as the crisis proceeded?

Capital market functionality: " Are you convinced about the ability of the capital market to include all information

available in the share price? Do you support the theory of an efficient capital market in setting share prices?

" Do you think that reliable conclusions can be drawn from the research method of measuring the capital market reactions on single event dates (Event study method that measures the deviation of actual share prices from predicted share prices that are

378

calculated based on historic data of an industry share index over a short period of time)? In your opinion can the near insolvency of AIG on September 17`h, 2008 be considered the most threatening situation for international insurance groups during the financial crisis? Do other situations come to your mind? Are you surprised about the research result that the capital market detected those international insurance companies whose equity ratio was reduced most by the financial crisis? Can you explain why the capital market detected them before they disclosed detailed information about impairments and lower market valuation about critical financial instruments in their financial statements 2008?

Overall questions: " According to your judgement: What are the international insurance groups that have

been affected most negatively by the financial crisis - which ones disclosed their risk timely and extensively, which ones didn't?

" What has to be done in order to allow private investors and analysts to be better informed about crisis relevant risk factors of insurance companies?

" Do you think that international insurance groups are as reluctant to disclose "sensitive" information about other topics as they have been reluctant to disclose risk information about "toxic" financial instruments that they hold during this crisis?

" Do you believe the results of this research might affect the ongoing discussion about fair value measurement/full fair value accounting in the insurance industry?

Thank you for your answers. Please state your name, job title, business responsibility or your link to the research topic, or opt for anonymity.

Do not hesitate to contact me via mail a. voelker(o)surrey. ac. uk or phone 01605254139.

For organizational questions or ethical concerns please contact Professor Gilbert at d. ciilbert@surrey. ac. uk or Professor Chen at j. Chen

. surrey. ac. uk at the university of Surrey.

Ansgar Voelker (DBA-Student

379

380

Appendix la: Interview Anonymous (2010)

Structured interview

Anonymous from a leading German investment company. The name and company cannot be disclosed due to internal governance regulations. Interview taken on December 27,2010 by Ansgar Völker.

Effect on equity ratio: " Do you think that the insurance industry has been hit hard by the financial crisis

in comparison to other industries?

The insurance industry got through the financial crisis quite well. Single listed European insurance groups with banking subsidiaries or business in the US faced challenges. However, a key element of the crisis - selling credit default swaps - is not a core business of insurance companies. The share price of listed European insurance groups suffered as a result of the financial crisis and the share price level is still lower than before the crisis.

" Which one of the thirteen insurance groups analyzed has been affected most negatively by the crisis?

AIG and SwissRe suffered most from the financial crisis because they have sold credit default swaps. AEGON sold variable annuities in the US market - these are life insurance products which are backed by US-investments. These investments were only partially hedged against market risk. As a result AEGON had to cover losses that arose from the unhedged part. Allianz owned Dresdner Bank at the beginning of the crisis and suffered losses from their US-exposure.

Are you aware of confounding negative effects that lowered the equity ratio of the analyzed insurance groups and that were not based on impairments or lower market valuation of financial instruments?

Listed European insurance groups wrote down their corporate bonds due to an increase in spreads that lowered their fair values. This put the equity ratio of insurance companies under pressure. Single insurance groups with banking subsidiaries like Allianz and ING made losses in the baking segment that lowered their equity ratio additionally.

" Can you name international insurance groups that have received financial aid from governments during the financial crisis?

AEGON and ING received state subsidies. SwissRe (not included in the group of accounts because it reports according to US-GAAP, not IFRS) received financial support by Berkshire Hathaway an institutional investor that increased its capital, signed a reinsurance contract and bought a convertible bond.

Risk disclosure behaviour: " Did you get a clear picture about how the insurance companies have been

affected by the crisis by reading their group financial statements? 381

The disclosure of financial risk in the annual financial statements of listed European insurance groups is insufficient and non-specific. I would prefer risk information to be quantitative, specified to the situation of the company and more extensive.

" Where you able to predict the effect the crisis had on an insurance group based on those annual information? What information about the insurance group that proved important later on did you learn from the group financial statements first?

Yes, it was possible to evaluate the impact of the crisis on the insurance groups. In order to do this I have used the equity movement tables, the disclosure of investment fair value and of deferred acquisition cost in the notes as additional information for my forecasts.

" Are you supportive of the theory that timely and extended disclosure by international insurance groups increases trust that investors have in that group?

Yes, I believe that timely and extensive risk disclosure increases trust of investors. Right now this disclosure is not sufficient and should be timely and extensive.

" Do you think that managers in the insurance industry can influence the extent of risk disclosure of the groups they lead in the financial statements? What might restrict them?

The management is hesitating to present the risk situation of their company. It prefers to present chances and success stories instead of disclosing risks.

Is it your impression that the quality of risk disclosure about troublesome financial instruments (such as: ABS, CDO, RMBS, CMBS) has improved as the crisis proceeded?

There is a positive development in disclosure about crisis related financial instruments. Disclosure about them has been more detailed and with more quantitative information in 2008 and 2009.

Capital market functionality: " Are you convinced about the ability of the capital market to include all

information available in the share price? Do you support the theory of an efficient capital market in setting share prices?

In general the market is always right. However, in this crisis a few phenomena appeared: Shares of listed insurance groups suffered unduly. The capital market overreacted which can be explained by the complexity of the insurance business model and the interdependencies within the insurance market. As a result the capital market perceived too high a risk - aside from US investments the insurance market continued to perform.

" Do you think that reliable conclusions can be drawn from the research method of measuring the capital market reactions on single event dates (Event study method that measures the deviation of actual share prices from predicted share prices that are calculated based on historic data of an industry share index over a short period of time)?

382

. Idd

This financial crisis has seen some extraordinary share price reactions. Three examples: The share price of AEGON increased after a rights issue has been announced. Such an announcement is typically followed by falling share prices. The announcement of a takeover of AIGA by MetLife has increased its share price, although it tends to fall after such an announcement. The presentation of negative quarterly results of ING has been followed by a share price increase. All these share price reactions are counter-intuitive and would need further research to be explained.

In your opinion can the near insolvency of AIG on September 17th, 2008 be considered the most threatening situation for international insurance groups during the financial crisis? Do other situations come to your mind?

AIG has been the biggest test of the insurance industry within the last decade. It has been an unanticipated one-time disaster. Another test of trust has been the recapitalization of ING by the Dutch government. Owner of saving accounts issued by the bank branch of this insurance group always assumed their savings to be safe. They were shocked by the need to increase INGs share capital. (Explanation: The owners of saving accounts are concentrated in Germany at the ING Diba)

" Are you surprised about the research result that the capital market detected those international insurance companies whose equity ratio was reduced most by the financial crisis? Can you explain why the capital market detected them before they disclosed detailed information about impairments and lower market valuation about critical financial instruments in their financial statements 2008?

I am not surprised that the capital market developed according to the effect the financial crisis had on the European insurance groups. It has been public knowledge that AIG had a credit default swap exposure or that Generali did not have investments in the US. When it became apparent that credit default swaps and portfolio based financial instruments had been the drivers of the financial crisis during 2008, the capital market reacted based on this knowledge. The publication of group financial statements in the first half of 2009 has been a non- event. It rarely happens that data from financial statements surprise the capital market. Investor days and conference calls provide additional information. However, insurance groups are prohibited to disclose market relevant information to a small group of investors only. They must be disclosed to the whole market via ad-hoc announcements.

Overall questions: " According to your judgement: What are the international insurance groups that

have been affected most negatively by the financial crisis - which ones disclosed their risk timely and extensively, which ones did not?

ING has been significantly affected by the financial crisis. However this company has disclosed a lot of information. Its annual financial statements exceeded 300 pages and their quarterly financial statements 260 pages. Allianz made an effort to be transparent about their risk exposure but they could have done more. Standard Life is not covered by our company. MunichRe and Zurich have been fully transparent in their exposure during the financial crisis. AXA has been transparent but hesitated to disclose risk information as the crisis continued.

383

" What has to be done in order to allow private investors and analysts to be better informed about crisis relevant risk factors of insurance companies?

There should be a level-playing field approach. Institutional and private investors shall rely on the same kind of information.

" Do you think that international insurance groups are as reluctant to disclose "sensitive" information about other topics as they have been reluctant to disclose risk information about "toxic" financial instruments that they hold during this crisis?

Listed insurance companies trade at a discount in comparison to other industries because investors perceive their accounting as incomparable and complex. Thus managers hesitate to disclose information that would further unsettle investors.

" Do you believe the results of this research might affect the ongoing discussion about fair value measurement/full fair value accounting in the insurance industry?

Full fair value of insurance groups would be based on estimates due to the lack of markets for insurance liabilities. Updating these estimates would lead to earning fluctuations between quarterly financial statements. This would run counter to the long term business orientation of insurance groups.

384

Appendix Ib: Interview Noack (2010)

Structured interview with Thomas Noack from WestLB, working as a Senior Equity Analysts (Director) on December 22,2010

Effect on equity ratio: " Do you think that the insurance industry has been hit hard by the financial crisis

in comparison to other industries?

Yes the insurance industry has been hit harder by the financial crisis than other industries. Although banks were affected directly and had to balance bigger financial losses, insurance companies as significant asset managers were affected by the financial crisis indirectly and on a long lasting basis. During the crisis federal reserve banks increased liquidity in the market. This market influence lowered interest rates. This causes a decline in the return on assets that insurance companies use to pay for customer claims.

" Which one of the thirteen insurance groups analyzed has been affected most negatively by the crisis?

Among the analyzed companies AEGON and ING have been hit hardest by the financial crisis, and they received governmental subsidies to cover their losses. Both insurance groups were engaged in the US-market and invested in mortgage-based financial instruments there. Aside from the group of analysis AIG and SwissRe suffered most from the crisis, because these groups both invested in mortgage-based financial instruments and sold Credit Default Swaps - although each group on a different scale. (Explanation: Both insurance groups are not analyzed in this research study, because they apply US- accounting standards and do not comply with IFRS 7)

Are you aware of confounding negative effects that lowered the equity ratio of the analyzed insurance groups and that were not based on impairments or lower market valuation of financial instruments?

The equity ration of only AIG and SwissRe has been lowered directly by losses from Credit Default Swaps. The equity ratio of all analyzed European insurance groups have been affected indirectly. Defaults of corporate bonds issuers and higher risk spreads on governmental bonds caused the fair value of these assets to decline resulting in lower equity ratios.

" Can you name international insurance groups that have received financial aid from governments during the financial crisis?

Among the analyzed companies ING and AEGON received governmental subsidies. Additionally SNS (Explanation: A smaller insurance group in the Netherlands and Belgium) received money form the state, whereas SwissRe received money from Warren Buffet, an institutional investor.

Risk disclosure behavior: " Did you get a clear picture about how the insurance companies have been

affected by the crisis by reading their group financial statements?

385

The financial crisis has changed the way European insurance groups communicate with analysts. There is a tendency for explicit information disclosure during the fiscal year. In 2009 and 2010 detailed information about the embedded value, the net asset value and the duration of investment categories have been disclosed by listed insurance groups. Especially the assumption that influence the embedded value and explicit information about asset allocation such as the character and extent of exposure in investments in Greece have been timely communicated to analysts. This is important for us, because our judgement about the value of a company differ when an insurance group is either invested in a Greek insurance subsidiary or is heavily invested in Greek sovereign bonds. The press coverage and resulting public pressure forced listed insurance groups to disclose detailed information to analysts. Just remember when questions about the exposure to structured investment instruments were followed by questions about their corporate bonds exposure and finally their sovereign bonds exposure. This kind of explicit and timely information has not been available during 2007 and 2008.

" Where you able to predict the effect the crisis had on an insurance group based on those annual information? What information about the insurance group that proved important later on did you learn from the group financial statements first?

n. a.

" Are you supportive of the theory that timely and extended disclosure by international insurance groups increases trust that investors have in that group?

Yes, I believe that timely and extended disclosure creates trust in investors. Investors don't like suprise. They can - up to a certain personal limit - cope with bad news as long as they are prepared for them and appreciate a long term relationship. Investors lost trust in SwissRe when it delayed bad news in 2007 so that the management had to go. Insurance groups have to swallow the pill and communicate problems and failures because they will become public knowledge sooner or later.

" Do you think that managers in the insurance industry can influence the extent of risk disclosure of the groups they lead in the financial statements? What might restrict them?

The management has to fulfill all accounting requirements. They can exceed them with discretionary disclosure. However the timing of disclosure has proved to be more important during the crisis. So disclosure of information during the fiscal year and not primarily in the financial statements is how management expresses themselves to investors. The financial crisis showed that managements have learned a lesson.

Is it your impression that the quality of risk disclosure about troublesome financial instruments (such as: ABS, CDO, RMBS, CMBS) has improved as the crisis proceeded?

Yes, the disclosure has improved along the financial crisis. At the beginning structured products have been disclosed as one amount. On enquiry the products have been subdivided into categories like CDO, RMBS, CMBS and so on.

386

Capital market functionality:

" Are you convinced about the ability of the capital market to include all information available in the share price? Do you support the theory of an efficient capital market in setting share prices?

The capital market proved quite efficient during the crisis. Whenever trouble arouse the share price moved correspondingly. The share price of SwissRe for example declined after its lagged risk disclosure.

" Do you think that reliable conclusions can be drawn from the research method of measuring the capital market reactions on single event dates (Event study method that measures the deviation of actual share prices from predicted share prices that are calculated based on historic data of an industry share index over a short period of time)?

n. a.

In your opinion can the near insolvency of AIG on September 17th, 2008 be considered the most threatening situation for international insurance groups during the financial crisis? Do other situations come to your mind?

The near insolvency of AIG had a significant impact on the insurance industry. The stumbling of such an insurance giant arose fear among investors. However an insolvency would have affected not only the insurance industry but governmental and industrial financial contracts that had been "secured" by Credit Default Swaps with AIG as a counterparty. Another event that affected the insurance industry has been the insolvency of Lehman. Although a bank its default led to write-offs in Lehman bonds that many insurance companies were invested in. For German insurance groups an insolvency of the later saved Hypo-Real-Estate Bank would have been significant because the Pfandbriefe issued by it are part of their asset mix.

" Are you surprised about the research result that the capital market detected those international insurance companies whose equity ratio was reduced most by the financial crisis? Can you explain why the capital market detected them before they disclosed detailed information about impairments and lower market valuation about critical financial instruments in their financial statements 2008?

It is no surprise to me, because information about toxic assets have been available to investors before the group financial statements were published. Alarmed by ad-hoc announcements analysts asked the management of all European insurance groups during conference calls and investor days about problems that one insurance group had to disclose. This created transparency in the financial market and share prices reacted correspondingly. This happened with SwissRe: The company made an ad-hoc announcement that their projected profits could not be met. A snap conference call revealed Credit Default Swaps as the troublemaker which has been hidden in off- balance vehicle before. SwissRe management had to explain itself and other insurance groups were asked the same questions as SwissRe during their gatherings with analysts.

387

Overall questions: " According to your judgement: What are the international insurance groups that

have been affected most negatively by the financial crisis - which ones disclosed their risk timely and extensively, which ones didn't?

How companies were affected by the financial crisis depends especially on where they were doing business. European insurance groups with business in the US invested more likely in mortgage based financial instruments. Generali as a European only insurance group praised itself as not being directly affected by the financial crisis. How the companies handle information about such risk differs. Assuming that management knows about the risk portfolio of their insurance group some pursue an active and other a passive way to disclose risk. MunicheRe, HannoverRe, Allianz and Zurich pursue an active way. The national requirement for listed German companies to disclose forecast information might be a reason for that behavior. HannoverRe for example breaks down their goals and quantifies them in a matrix which is disclosed to analysts. AEGON does not sufficiently disclose the assumptions of their embedded value calculation.

" What has to be done in order to allow private investors and analysts to be better informed about crisis relevant risk factors of insurance companies?

The internet provides transparency. Management presentations in conference calls and investors' days are place in the internet. They are accessible to all interested private investors. However, analysts have an advantage because they can place questions on such occasions.

" Do you think that international insurance groups are as reluctant to disclose "sensitive" information about other topics as they have been reluctant to disclose risk information about "toxic" financial instruments that they hold during this crisis?

Some European insurance groups still hesitate to disclose risk information. Most groups have improved their communication on urgent risk topic. Among these groups is a preference to disclose and settle these risk topics by building reserves or writing down assets. Follow-up disclosure on risk topics once settled is not provided by insurance groups. An example of such preference is the asbestos risk of reinsurance companies such as MunichRe.

Do you believe the results of this research might affect the ongoing discussion about fair value measurement/full fair value accounting in the insurance industry?

Risk disclosure as well as fair value accounting can have a significant impact on the share price of listed European insurance groups. The financial crisis set limits in the disclosure of insurance groups regarding share price relevant information. The IASB allowed a recategorization of investments away from fair value when the financial crisis peaked and companies wanted to conceal drops in fair values. Or, MunichRe disclosed - unlike other insurance groups -a negative embedded value for its life

388

segment, because it was of no relevance in the assessment of the overall performance of the group.

389

Appendix Ic: Interview Seel (2010)

Structured interview with Alexander Seel from Deutsche Bank, Financial Institutions Group on January 24,2011

Effect on equity ratio: " Do you think that the insurance industry has been hit hard by the financial crisis

in comparison to other industries?

Yes, the insurance industry has been hit hard by the financial crisis. Insurance groups are asset managers that are sensible to market volatility caused by the crisis. But insurance companies have been associated by the public with banks during the financial crisis, although they differ in their business models.

" Which one of the thirteen insurance groups analyzed has been affected most negatively by the crisis?

Insurance groups with bank business like Allianz or ING have been hit hard. As well as insurance groups with high leverage and a significant proportion of high risk assets in their portfolio suffered from the reduction of their equity.

" Are you aware of confounding negative effects that lowered the equity ratio of the analyzed insurance groups and that were not based on impairments or lower market valuation of financial instruments?

The development of the equity ratio can be used as a proxy for the effect the financial crisis had on European insurance groups. Preferably a proxy should be based on several effects such as the change in IFRS accounting results, regulatory accounting results, the change in their rating status and the change in the disclosed market consistent embedded value. All changes should be measured from the beginning to the end of the financial crisis and then accumulated. Such a modified proxy would consider offsetting effects. Example given: An increased market interest rate would improve the embedded value but would decrease the IFRS accounting results due to lower market values of their fixed interest assets.

" Can you name international insurance groups that have received financial aid from governments during the financial crisis?

European insurance groups with banking activities such as ING or Fortis received state subsidies as well as SNS a smaller insurance group in the Netherlands and Belgium that is not researched. Other than that US financial institutions such as AIG or Heartfort received financial aid.

Risk disclosure behavior: " Did you get a clear picture about how the insurance companies have been

affected by the crisis by reading their group financial statements?

No annual financial statements offer plenty of information. However for an analyst it is necessary to compare this information with other insurance groups or other industries. To benchmark this information is "almost impossible because of the complexity of the business models, assumptions that are made in order to calculate embedded values, different categorization of assets according to AS 39 that lead to different effects in the

390

profit & loss accounts. Additionally national regulatory requirements differ for insurance groups. Example given: Changes in the assets value based on increased interest rates can be deducted from regulatory capital requirements for life insurance companies in some European countries but not in other ones.

" Where you able to predict the effect the crisis had on an insurance group based on this annual information?

The insurance industry is far too complex in order to be able to make precise predictions about the effects such a financial crisis has on the results of single European insurance groups.

" What information about the insurance group that proved important later on did you learn from the group financial statements first?

Information about structured financial instruments have improved during the financial crisis. But the disclosure about sovereign bonds has to improve.

" Are you supportive of the theory that timely and extended disclosure by international insurance groups increases trust that investors have in that group?

Yes, I can support this theory. However, it is not the quantity of disclosed information that matters to analysts but its quality. The quality of the information depends on its comparability especially in benchmarking. Example given: Allianz disclosed plenty of information in a very structured way. As an analyst you cannot compare those data with other insurance groups.

" Do you think that managers in the insurance industry can influence the extent of risk disclosure of the groups they lead in the financial statements? What might restrict them?

The management of insurance groups are restrained by international financial reporting standards that they all have to comply with. They can exceed these requirements by discretionary disclosure. In the' long run the insurance industry influences the setting. of standards by involvement: in lobbying. Example, given: The current standardizing process of IFRS 9.

Is it your impression that the quality of risk disclosure about troublesome financial instruments (such as: ABS, CDO, RMBS, CMBS) has improved as the crisis proceeded?

The quality of disclosure about structured financial instruments has improved during the crisis.

Capital market functionality: " Are you convinced about the ability of the capital market to include all

information available in the share price? Do you support the theory of an efficient capital market in setting share prices?

Capital market theories always assume efficiency and comparable information at a level playing field of all participants. These assumptions do not meet reality. Internal studies by DeutscheBank tried to compare new business margins among European

391

insurance groups. We found that every country has special regulations and every company makes differing assumptions. These factors restrict the efficiency of the capital market.

" Do you think that reliable conclusions can be drawn from the research method of measuring the capital market reactions on single event dates (Event study method that measures the deviation of actual share prices -from predicted share prices that are calculated based on historic data of an industry share index over a short period of time)?

Not applicable

" In your opinion can the near insolvency of AIG on September 17th, 2008 be considered the most threatening situation for international insurance groups during the financial crisis? Do other situations come to your mind?

I don't think about certain days but periods. The period between autum 2008 and spring 2009 has been characterized by extreme volatility. As negative information emerged from insurance groups investors found it hard to build up trust in these companies. A similar period has been the sovereign risk crisis around spring 2010. Although the information has been around longer and could not be considered by investors as unanticipated events they arouse similar fear of investors as the afore mentioned period.

" Are you surprised about the research result that the capital market detected those international insurance companies whose equity ratio was reduced most by the financial crisis? Can you explain why the capital market detected them before they disclosed detailed information about impairments and lower market valuation about critical financial instruments in their financial statements 2008?

No, I am not surprised that the capital market sorted out those European insurance groups that were affected negatively by the financial crisis although relevant information had not been disclosed in their annual financial statements so far. Investor calls, workshop, investor information ant other opportunities of the management to explain their current business results improve the understanding of these companies by analysts. Due to these activities the capital market gains an understanding about how the financial crisis affects European insurance groups.

Overall questions: " According to your judgement: What are the international insurance groups that

have been affected most negatively by the financial crisis - which ones disclosed their risk timely and extensively, which ones didn't?

In general those European insurance groups with life-insurance activities have risks that are harder to explain and to disclose to investors. In addition to that insurance groups with banking activities such as ING or Allianz have experienced a significant decrease of their equity ratio. MunichRe on the opposite had their main activity in reinsurance that has not been hit hard by the crisis and a more conservative asset allocation. After the crisis 2003 they have converted their business in a more risk sensitive company.

392

" What has to be done in order to allow private investors and analysts to be better informed about crisis relevant risk factors of insurance companies?

Private investors enjoy the transparency of European insurance groups the same way as institutional investors. However only combined with experience this information can be used beneficially in their financial decision making.

" Do you think that international insurance groups are as reluctant to disclose "sensitive" information about other topics as they have been reluctant to disclose risk information about "toxic" financial instruments that they hold during this crisis?

n. a.

" Do you believe the results of this research might affect the ongoing discussion about fair value measurement/full fair value accounting in the insurance industry?

Although I agree that additional and comparable disclosure would be beneficial for investors I doubt that the disclosure of fair value based measurement of both sides of the balance sheet (full fair value) would be practical and provide decision usefulness to investors. They would not understand the high volatility of earnings caused by the measurement process. The disclosed information would not be comparable by investors due to the different assumptions made that are necessary to estimate fair values when real market based fair values do not exist.

393

Appendix Id: Interview Anonymous (2011a)

Structured Interview with Anonymous from a major Investment Bank on January 28,2011

Effect on equity ratio: " Do you think that the insurance industry has been hit hard by the financial crisis

in comparison to other industries? Yes the financial sector with banks and insurance companies have been hit hard by the crisis.

" Which one of the thirteen insurance groups analyzed has been affected most negatively by the crisis?

I don't want to mention single companies.

" Are you aware of confounding negative effects that lowered the equity ratio of the analyzed insurance groups and that were not based on impairments or lower market valuation of financial instruments?

It is possible to use equity ratios as proxies for the impact the financial crisis had on insurance groups. However, a better proxy would be to measure the write ups on financial instruments after the crisis and market values increased. With such a proxy you could measure to what extend European insurance groups have written down and written up their mortgage based financial instruments. In addition to the effect of the crisis you could research to what extend these groups accounted true and fair. Those with the biggest write ups have shown the true value of the financial instruments at each point of time, whereas those groups without or limited write ups did not mark their assets down during the financial crisis.

" Can you name international insurance groups that have received financial aid from governments during the financial crisis?

n. a

Risk disclosure behaviour: " Did you get a clear picture about how the insurance companies have been

affected by the crisis by reading their group financial statements? Where you able to predict the effect the crisis had on an insurance group based on this annual information?

It has been possible to predict the effect of the crisis on insurance groups. Gathering all available information about an insurance group over a long period of time analysts can predict the effects of certain developments. But the group financial statements are not the only source of information used by analysts.

What information about the insurance group that proved important later on did you learn from the group financial statements first?

n. a

" Are you supportive of the theory that timely and extended disclosure by international insurance groups increases trust that investors have in that group?

I support the theory that timely and extended disclosure increases trust by investors. It is necessary that this trust remains in crisis'situation because the business model of insurance companies is so complex. When investors do not receive timely and

394

comprehensible information in a crisis situation they lose trust and assume the worst. This can lead to overreactions that we have experienced during the crisis.

" Do you think that managers in the insurance industry can influence the extent of risk disclosure of the groups they lead in the financial statements? What might restrict them?

n. a

" Is it your impression that the quality of risk disclosure about troublesome financial instruments (such as: ABS, CDO, RMBS, CMBS) has improved as the crisis proceeded?

Yes, the quality of risk disclosure has improved as the crisis proceeded. Information about troublesome financial instruments has become more structured, more specific to the situation of the company and in a few cases related to the statutory capital requirements and the excess capital of European insurance groups.

Capital market functionality: " Are you convinced about the ability of the capital market to include all

information available in the share price? Do you support the theory of an efficient capital market in setting share prices?

n. a.

Do you think that reliable conclusions can be drawn from the research method of measuring the capital market reactions on single event dates (Event study method that measures the deviation of actual share prices from predicted share prices that are calculated based on historic data of an industry share index over a short period of time)?

n. a.

" In your opinion can the near insolvency of AIG on September 17th, 2008 be considered the most threatening situation for international insurance groups during the financial crisis? Do other situations come to your mind?

At this point of time private and institutional investors were shocked and fearful. This event has affected the whole capital market.

Are you surprised about the research result that the capital market detected those international insurance companies whose equity ratio was reduced most by the financial crisis? Can you explain why the capital market detected them before they disclosed detailed information about impairments and lower market valuation about critical financial instruments in their financial statements 2008?

n. a.

Overall questions: " According to your judgement: What are the international insurance groups that

have been affected most negatively by the financial crisis - which ones disclosed their risk timely and extensively, which ones didn't?

n. a. " What has to be done in order to allow private investors and analysts to be better

informed about crisis relevant risk factors of insurance companies? The transparency of how the industry accounts for their investments should improve. Categories of investment and the related measurements should be fully transparent or

395

abolished in favour of a consistent market value related measurement of both sides of the balance sheet.

" Do you think that international insurance groups are as reluctant to disclose "sensitive" information about other topics as they have been reluctant to disclose risk information about "toxic" financial instruments that they hold during this crisis?

They might hesitate but they are pressed by analyst to disclose them. When investor relation departments of European insurance groups do not comment on specific questions analysts assume the worst and act accordingly.

Do you believe the results of this research might affect the ongoing discussion about fair value measurement/full fair value accounting in the insurance industry?

I believe that European insurance groups should disclose the fair values of their assets and liabilities. This would lead to higher transparency. Although I think that insurance groups would hesitate to measure and disclose in that way. They would do so when they could afford it.

396

Appendix Ie: Interview Anonymous (2011a)

Structured Interview with Anonymous from a major Rating Agency on January 25,2011

Effect on equity ratio: " Do you think that the insurance industry has been hit hard by the financial crisis

in comparison to other industries? In comparison to the banking sector the insurance industry took it well. There has been an impact on their balance sheets but not as intense as in the banking sector. This is because insurance companies have a longer liquidity position than banks. It enabled them to hold on to their investments until after the crisis. Banks instead had to fire sell parts of their investments and realized losses.

" Which one of the thirteen insurance groups analyzed has been affected most negatively by the crisis?

European insurance groups with an exposure to the US-life insurance market or with banking activities have been downgraded as a result of the financial crisis. The life- insurance exposure forced the companies to meet financial guarantees of the policies they have sold that could not be achieved with the actual return of their investments.

" Are you aware of confounding negative effects that lowered the equity ratio of the analyzed insurance groups and that were not based on impairments or lower market valuation of financial instruments?

To use the equity ratios as a proxy for the effects the financial crisis had on European insurance groups makes sense keeping in mind the measurement rules of IAS 39 that they all had to apply. Lower fair values of investments, fire sales of assets at a loss and negative operating results of the banking segments, lowered the economic value of the insurance groups together with their equity ratio.

" Can you name international insurance groups that have received financial aid from governments during the financial crisis?

Fewer insurance groups received financial aid from governments than banks. AEGON ING and SNS (explanation: a Dutch/Belgium insurance group) come to mind as recipients.

Risk disclosure behaviour: " Did you get a clear picture about how, the insurance companies, have been

affected by the crisis by reading their group financial statements? Where you able to predict the effect the crisis had on an insurance group based on this annual information?

Sensitivity information in financial statements that disclose the effect of . different scenarios on key business figures can be useful for rating decisions.

" What information about the insurance group, that proved important later on, did you learn from the group financial statements first?

Some European insurance groups disclosed their capital sensitivity. ' This information enabled investors to estimate how much loss the companies could take. It was useful for companies because it reassured investors about the capital position of these groups.

397

" Are you supportive of the theory that timely and extended disclosure by international insurance groups increases trust that investors have in that group?

Timely and extended disclosure can be considered beneficial to investors. However, this disclosure is not exclusively provided by financial statements but in particular in investor presentations.

" Do you think that managers in the insurance industry can influence the extent of risk disclosure of the groups they lead in the financial statements? What might restrict them?

Managers of European insurance groups face the problem that the information they disclose its hard to get across, because investors do not really; understand the industry. Therefore, it is their challenge to put information out that is understood. -

" Is it your impression that the quality of risk disclosure about troublesome financial instruments (such as: ABS, CDO, RMBS, CMBS) has improved as the crisis proceeded?

Most European insurance groups have improved their transparency during the financial crisis because investors demanded the information. Investors assumed the worst and were assured by positive and negative information.

Capital market functionality: " Are you convinced about the ability of the capital market to include all

information available in the share price? Do you support the theory of an efficient capital market in setting share prices?

These are model assumptions. Investors as key figures of the capital markets often do not completely understand the insurance industry they are invested in. This is caused by the complexity of the business model and the variety of regulations that must be observed.

Do you think that reliable conclusions can be drawn from the research method of measuring the capital market reactions on single event dates (Event study method that measures the deviation of actual share prices from predicted share prices that are calculated based on historic data of an industry share index over a short period of time)?

n. a

" In your opinion can the near insolvency of AIG on September 17th, 2008 be considered the most threatening situation for international insurance groups during the financial crisis? Do other situations come to your mind?

The near insolvency of AIG has been a point of time of overall anxiety in the whole financial industry not just the insurance industry.

" Are you surprised about the research result that the capital market detected those international insurance companies whose equity ratio was reduced most by the financial crisis? Can you explain why the capital market detected them before they disclosed detailed information about impairments and lower market valuation about critical financial instruments in their financial statements 2008?

Annual financial statements came too late to provide investors with the necessary information about the financial crisis. This kind of information has been disclosed during investor presentations, road shows and conference calls. Investors rely more on this kind of sources during crisis situations.

398

Overall questions: " According to your judgement: What are the international insurance groups that

have been affected most negatively by the financial crisis - which ones disclosed their risk timely and extensively, which ones didn't?

European insurance groups with life insurance such as Aegon and ING have been downgraded more often than ones with property and casualty insurance like Aviva, AXA, Generali, HannoverRe, MunichRe and Zurich.

" What has to be done in order to allow private investors and analysts to be better informed about crisis relevant risk factors of insurance companies?

The transparency of the industry should improve in order to make it a more attractive asset class. Accounting rules like IFRS 4 that require measurement based on market values try that. However, it is unclear whether or not this will be achieved.

" Do you think that international insurance groups are as reluctant to disclose "sensitive" information about other topics as they have been reluctant to disclose risk information about "toxic" financial instruments that they hold during this crisis?

European insurance groups are reluctant to disclose this kind of information but they have less of a choice. They have to disclose the right information in the right way so that investors understand it better.

" Do you believe the results of this research might affect the ongoing discussion about fair value measurement/full fair value accounting in the insurance industry?

The financial crisis had an effect on the standard setting process of IFRS 4. It made clear that during a crisis market values do not exist despite pressure on the accountants to market value their assets. However, it would be beneficial having a consistent approach to measure the debit and credit side of the balance sheet.

399