Mining loan level data

Deutsche Bank Markets Research

Europe

Special Report

Credit Securitisation

Date 4 September 2013

Spanish RMBS: Mining loan level data

________________________________________________________________________________________________________________

Deutsche Bank AG/London

DISCLOSURES AND ANALYST CERTIFICATIONS ARE LOCATED IN APPENDIX 1. MICA(P) 054/04/2013.

Rachit Prasad

Research Analyst

(+44) 20 754-70328

[email protected]

Conor O'Toole

Research Analyst

(+44) 20 754-59652

[email protected]

Following on from our initial examination of European loan level data from the European DataWarehouse (ED) in our report titled “Irish RMBS: A first look at loan level data”, we move on to analyse another peripheral RMBS market – Spain. We first take the opportunity to review data completeness. With 97 deals amounting to 85% by volume of Spanish RMBS investor placed deals signed up to the European DataWarehouse, coverage is good, albeit there are issues with some mandatory fields being collected but not yet displayed. While a majority of lenders collect a significant (90%+) amount of the 69 mandatory fields defined by the ECB RMBS taxonomy, the incidences of mandatory fields which have been collected by the lender but nevertheless are for now not presented in loan level data is high. Next we take a closer look at broad loan trends across Spanish investor placed deals. The recent observable deterioration in credit metrics leads us to examine what loan level characteristics - to include geography, LTV (current), employment status, loan-to-income ratios, and property occupancy type - are correlated to underperformance. We discuss each of these characteristics in turn (LTV and geography unsurprisingly turn out to be key drivers) and present these at a deal level for our sample universe. Finally, overlaying loan level analysis onto bond attachment points and pricing, we identify relative value opportunities amongst 30 senior Spanish RMBS bonds. We calculate the liquidation loss of loans where the debt service to indexed income ratio is greater than 30%. We use this metric as an indicator of potential future losses in deals since a dynamic data series of loan level characteristics is not yet available, while it captures LTV, geography, and borrower leverage. Comparing the credit enhancement of bonds to this liquidation loss metric leads us to recommend the following key bond switches which emphasize better bond safety while picking up spread Switch from BFTH 13 A2 to BCJAF 6 A2 to pickup 15 bps in spread

Switch from SABA 1 A2 to HIPO HIPO-5 A to pickup 15 bps in spread

Switch from BCJAM 4 A2 to RHIPO 8 A2A to pickup 10 bps in spread

Risks: Worse than expected credit performance in deals from a higher unemployment rate or a persistent deep property downcycle are risks to our recommendations. Spreads and yields are contingent on realised prepayment and default rates. Bond specific illiquidity and technical factors as well as deal specific risks such as those associated with sponsors and counterparties are additional risks to bear in mind.

Table Of Contents

Introduction…………..………..……….......2

Spanish loan level data Completeness....2

Broad trends from Spanish loan

level data…………………………….………4

What does this mean at a deal level?....12

Spanish Senior RMBS Relative Value…14

Summary and recommendations……...16

4 September 2013

Special Report: Spanish RMBS: Mining loan level data


Introduction

Following on from our initial examination of European loan level data from the European DataWarehouse (ED) in our report entitled “Irish RMBS: A first look at loan level data”, we move on to analyse another peripheral RMBS market – Spain.

We first take the opportunity to review data completeness. With 97 deals amounting to 85% by issuance of Spanish RMBS investor placed deals signed up to the European DataWarehouse, coverage is good, albeit there are issues with some mandatory fields being collected but not yet displayed. While a majority of lenders collect a significant (90%+) amount of the 69 mandatory fields defined by the ECB RMBS taxonomy, the incidences of mandatory fields which have been collected by the lender but nevertheless are for now not presented in loan level data currently is high.

Next we take a closer look at broad loan trends across Spanish investor placed deals. The recent observable deterioration in credit metrics leads us to examine what loan level characteristics - to include geography, LTV (current), employment status, loan-to-income ratios, and property occupancy type - are correlated to underperformance. We discuss each of these characteristics in turn (LTV and geography unsurprisingly turn out to be key drivers) and present these at a deal level for our sample universe.

Finally, overlaying loan level analysis onto bond attachment points and pricing, we identify relative value opportunities amongst 30 senior Spanish RMBS bonds. We calculate the liquidation loss of loans where the debt service to indexed income ratio is greater than 30%. We use this metric as an indicator of potential future losses in deals since a dynamic data series of loan level characteristics is not yet available, while it captures LTV, geography, and borrower leverage.

Comparing the credit enhancement of bonds to this liquidation loss metric leads us to recommend the following key bond switches which emphasize better bond safety while picking up spread -

Switch from BFTH 13 A2 to BCJAF 6 A2 to pickup 15 bps in spread

Switch from SABA 1 A2 to HIPO HIPO-5 A to pickup 15 bps in spread

Switch from BCJAM 4 A2 to RHIPO 8 A2A to pickup 10 bps in spread

Risks: Worse than expected credit performance in deals from higher unemployment or a persistent deep property downcycle are risks to our relative value recommendations. Spreads and yields are contingent on prepayment and default rates. Bond specific illiquidity and technical factors as well as deal specific risks such as those associated with sponsors and counterparties are additional risks to bear in mind.

Spanish loan level data completeness

At present some 97 of the 132 investor placed Spanish RMBS deals are available on European DataWarehouse equating to EUR 49.5 billion or 85% of all investor placed Spanish RMBS outstanding (Figure 1) (Please see Appendix A for a breakdown of the programmes in the European DataWarehouse). In common with other European RMBS sectors where loan level data is available highest priority gripes include different dates between loan level data and investor reports, along with key fields not being populated. To recap the RMBS taxonomy/template consists of 191 fields, 69 of which are mandatory.

4 September 2013



Figure 1: European DataWarehouse – Sector snapshot

Investor placed deal count at

ED/total deals in sector

Investor placed outstanding with

loan level data (EUR Bn)*

Share of investor placed

outstanding (EUR Bn)

Stock outstanding (investor placed in

EUR Bn)

Dutch RMBS 74/97 66.2 86% 77

Greek RMBS 0/10 0.0 0% 3

Irish RMBS 6/9 7.0 75% 9

Italian RMBS 32/56 13.6 81% 17

Portuguese RMBS 18/23 8.5 82% 10

Spanish RMBS 97/132 49.5 85% 59

UK prime RMBS* 9/29 42.6 42% 102

Spanish SME CLOS 33/47 5.6 78% 7

Total 269/403 193.0 68% 284 Source: Deutsche Bank,* UK Prime RMBS is in EUR billion

While on the face of it data completeness of these 97 deals appears quite good or at least has the potential to be, when one digs deeper some key fields that are necessary for analysis are still missing at present. A recap on ECB data population guidelines follows.

Where originators cannot provide data in accordance with the template, predefined “no-data” (ND) options are available which clarifies the reason for non-availability of data. These ND options are detailed in Figure 2.

Figure 2: Data Scoring based on ECB guideline

“No Data” option Explanation

ND 1 Data not collected as not required by the underwriting criteria

ND 2 Data collected at application but not loaded in the reporting system at completion

ND 3 Data collected at application but loaded in a separate system from the reporting one

ND 4 Data collected but will only be available from YYYY-MM

ND 5 Not relevant at the present time

ND 6 Not applicable for the jurisdiction

ND 7 Only for CMBS loans with a value less than EUR 500 000, i.e. the value of the whole commercial loan balance at origination

Source: ECB

The ECB next lays out a scoring matrix1 which measures the prevalence of ND (no-data) entries for mandatory fields. This scoring matrix along with the Spanish RMBS specific results are displayed in Figure 3. As the table highlights, the prevalence of ND 1 entries (which measures the incidences of mandatory fields not required by the lenders underwriting criteria) are coded by letter (A to D) while the aggregate prevalence of ND 2, ND 3, ND 4 (which measures the incidences of where mandatory fields are collected by lenders but not yet loaded into the ECB template) are coded by number (1 to 4), which collectively form the matrix.2 Almost all lenders collect a significant (90%+) amount of the 69 mandatory fields defined by the ECB RMBS taxonomy. Indeed some 77% of deals are classified as “A” meaning they collect 100% of

1 http://www.ecb.int/paym/coll/loanlevel/implementation/html/index.en.html

2 More specifically ND 1 - Data not collected as not required by the lenders underwriting criteria, ND 2 and

ND 3 Data collected at application but either not loaded at completion or loaded in a separate system from

the reporting one, and ND 4 - data collected but will only be available from a specified date in the future.

http://www.ecb.int/paym/coll/loanlevel/implementation/html/index.en.html

4 September 2013



required mandatory fields while a further 22% are classified as “B” meaning less than 10% of required mandatory fields are not collected.

Figure 3: Loan level completeness Scoring Matrix – Mandatory fields

ND 1 fields

0 ≤ 10% ≤ 30% > 30%

ND 2,ND 3 or ND 4

0 A1 (21) B1(7) C1(1) D1

≤ 20% A2 (45) B2(15) C2 D2

≤ 40% A3 (8) B3 C3 D3

> 40% A4 B4 C4 D4 Source: ECB, ED, numbers in brackets represent number of Spanish RMBS deals in category. The above includes 88 investor placed deals and 9 retained deals.

However the incidences of mandatory fields which have been collected by the lender but nevertheless are for now not presented in loan level data - categories A2 and A3 – at 53 deals, is high. What this means is when one looks at “Key Fields”, at present many of these are simply not disclosed. We define key fields as those which we believe are best designed to conduct loan level analysis. The 12 key fields designated include LTV, income, property valuation, origination date, arrears, mortgage rate, loan term, original and current loan balance, employment status, property occupancy, and geography. (Please see Appendix C for an extract of an example loan tape). For example, a cursory examination of data completeness shows that in many deals, data as simple as arrears balance and original income levels (used to arrive at debt to income ratios) are currently collected by the underwriter but not reflected at present in loan level data.

Amongst Spanish RMBS shelves, Bankinter’s BFTH and Santander’s HMSF programmes display the best source of data – they score A1 in the ECB mandatory field matrix, while Banco Mare Nostrum’s AYTCH, and the multi-originator AYTH M2, Banco Popular’s IMPAS 4, Banco Sabadell’s TDAC deals and UCI’s (currently part of BNP Santander/Banco Santander) UCI 9 are amongst the worst which score B2 or C1.

Additionally, we find that arrears data reported in loan tapes do not necessarily match with those reported in investor reports. We looked for a combination of loan fields which included “account status”,”number of months in arrears”, “arrears balance” to identify loans in arrears, but find that arrears from loan tape can be materially different from investor reports. On average we find that 90d+ arrears (including default) is 3.8% using loan level data while it is 5.7% from investor reports. Possible reasons for the difference include different sample sets (97 deals in loan data set versus 134 overall), timing differences (loan tapes are dated a few months prior to recent investor reports), potential difficulties in identifying loans in arrears or slippages in loan level data in recording loans in arrears.

Broad trends from Spanish loan level data

In the context of ever increasing delinquencies in the Spanish RMBS sector where sector average credit risky loans3 have increased by 1.45 ppts y-o-y (Figure 4) with some deals jumping by more than 3 ppts since January alone (Figure 5), the recent availability of loan level data is timely. Indeed loan level data from the ED Warehouse should facilitate a more rigorous analysis of the drivers of mortgage credit underperformance.

3 Credit risky loans are defined as 90d+ arrears plus defaults-in-hand.

4 September 2013



Figure 4: Credit risky loans rising Figure 5: Top increases in credit risky loans in Spanish

RMBS

0%

1%

2%

3%

4%

5%

6%

7%

Jan

-04

May-0

4

Sep

-04

Jan

-05

May-0

5

Sep

-05

Jan

-06

May-0

6

Sep

-06

Jan

-07

May-0

7

Sep

-07

Jan

-08

May-0

8

Sep

-08

Jan

-09

May-0

9

Sep

-09

Jan

-10

May-1

0

Sep

-10

Jan

-11

May-1

1

Sep

-11

Jan

-12

May-1

2

Sep

-12

Jan

-13

May-1

3

% o

f curr

en

t bala

nce

90d+ Arrears

Defaults in hand

Aggregate 90d+ arrears and defaults in hand

0%

1%

2%

3%

4%

5%

6%

7%

0%

5%

10%

15%

20%

25%

30%

35%

40%

45%

Ch

an

ge

in c

red

it ris

ky lo

an

s (

%)

90

d+

arr

ears

+ d

efa

ults in

han

ds (cre

dit ris

ky

loan

s %

)

Jul-13 (LHS) Jan-13 (LHS) Change: July - Jan (RHS)

Source: Deutsche Bank

Source: Deutsche Bank, ABSnet

We examine 5 key loan level characteristics to include geography, LTV (current), employment status, loan-to-income ratios, and property occupancy type. In our study, we focus in turn on individual factors as a determinant of mortgage credit performance. Indeed, in reality a combination of a large number of factors will matter for a borrower default. Our effort here is to only provide a flavour for where and how credit performance may be affected. To get an idea of how the key fields in a loan tape would actually look, please see Appendix C for an extract from the BFTH 4 loan tape.

We summarise the data set and discuss each of these characteristics in turn below.

Data set and Arrears: The dataset we use includes only investor placed RMBS deals which comprises 97 deals amounting to EUR 51.3 billion of mortgages equating to c. 85% of outstanding Spanish mortgages in investor placed RMBS transactions. Therefore the data provides us with a large and meaningful sample of loans from which to draw conclusions.

At the outset, we highlight that the latest loan tapes are dated December 2012 – April 2013 depending on deal, therefore they reflect the state of the RMBS pools c.3-6 months ago. They will not relate to latest investor reports released by trustees on a monthly basis. Timely and more frequent reporting of loan level data remains an area for improvement.

We find our sample loan universe has some EUR 3.8 billion (7.5%) (See Footnote[5]) of loans in arrears (including those classified as being in default in RMBS pools) of which c EUR 1.9 billion (or 3.8% of total) are in arrears (again including those classified as being in default) for more than 3 months. Indeed, 3m+ arrears balance of 3.8%4 ties in broadly with what Bank of Spain reports

4 We identify loans as being in arrears if

-If account status is available, then loan is non-performing if Account status is 2 (Arrears) or 3 (Default or

Foreclosure)

-If account status is not available, if arrears balance > 0, or if Number of months in Arrears > 0

4 September 2013



as doubtful mortgage loans 4.16 %.5 Figure 6 provides a snapshot of the arrears buckets as a percentage of the total pool.

Figure 6: Snapshot of Spanish RMBS loan level data

0%

1%

2%

3%

4%

5%

6%

7%

8%Tota

l

arr

ears

> 3

m

arr

ears

> 6

m

arr

ears

> 1

2m

arr

ears

> 1

8m

arr

ears

> 2

4m

arr

ears

Arr

ears

as

% o

f to

tal poo

l

Source: European DataWarehouse. 92.6% of loans in the universe are classified as performing

Arrears by geography/region Mortgage credit performance by region has always been of interest to RMBS investors, but was difficult to analyse due to the paucity of data before loan level data became available.

Using geographical information available for 72 deals, we classify the regional codes in loan tapes into broad geographic classification used by the Ministry of Housing (MDV) for its regional house price indices. We then compute the relative levels of distress in Spain by region which we present in a map format in Figure 7. The map is colour coded to differentiate varying level of arrears as a percentage of the regional pool. The grey areas represent the regions that are relatively more distressed, followed by the darker blue colours. The least affected regions are represented as lighter shades of blue.

We find that arrears percentages as a percentage of regional pool are highest in Murcia (18.3% total arrears as a ratio of regional pool), and Catalunya (11.2% total arrears as a ratio of regional pool). Comunidad Valenciana (10.8%), Balears (10.3%) and Andalucia (9.2%) follow closely with total arrears close to 10% in these regions (Figure 7).

While Murcia has the largest share of regional arrears at 18.3%, only 3.3% of the mortgage pool is from the region, and about 8.4% of Spanish arrears is attributable to Murcia. Indeed, Catalunya and Valenciana contribute to 23.4% and 22.5% of national arrears and remain more significant from a RMBS credit perspective, since a higher proportion of the pool (c. 30% of the universe) is from these two regions.

Note that total arrears here refer to all loans which are overdue in payment and includes those loans classified as being in default (12m+ or 18m+ arrears depending on deal, or classified as default by the sponsor).

5 This 3.8% 90d+ arrears (including defaults) figure is lower than 5.7% shown in Figure 4 sourced from

investor reports. Possible reasons for differences include different sample universes, timing differences,

potential difficulties in identifying loans in arrears or slippages in loan level data in recording loans in

arrears.

4 September 2013



Figure 7: Total arrears (including defaults) by region using loan level data

Source: Loan level data from European DataWarehouse. Note that about 18% of mortgages do not have geography information, and another 5.9% are classified as “Extra Regional”- loans to regions outside Spain- Not to scale

Arrears by LTV Loan level data also allows us to examine arrears performance by vintage, and confirms what the market has long known (Figure 8) - that mortgages from 2005 onwards have had a higher propensity to default (8-10% of loans from these years are in arrears) than loans from earlier years. With average seasoning in RMBS deals being c.21 months at the time of issue, mortgages originated in 2005 appeared in deals from late 2006 and 2007 deals where we see, broadly speaking, inferior credit performance.

4 September 2013



Figure 8: Arrears percentage by year of mortgage

origination

Figure 9: Total arrears by Indexed LTV bucket

0%

1%

2%

3%

4%

5%

6%

7%

8%

9%

10%

199719981999200020012002200320042005200620072008

Arr

ears

as %

of

po

ol i

n t

he

vin

tag

e

bu

cke

t

Vintage

0%

2%

4%

6%

8%

10%

12%

14%

16%

18%

20%

Indexed LTV bucket

Arrears as % of pool bucket

Arrears as % of total arrears

Source: European DataWarehouse


The high proportion of arrears in mortgages from 2005-2008 (Figure 8) coincided with a buoyant Spanish property market. Since current property prices are substantially lower than during those years, we can hypothesize that lower borrower equity (compared to mortgages from earlier years) may be driving default rates. In order to confirm this, we adjust property prices at a loan level recorded at the time of origination to current levels using regional house price indices from the Spanish Ministry of Housing (MDV)6. We then arrive at indexed LTV as the ratio of current outstanding loan amount to the indexed property value.

We then classify arrears by indexed LTV bucket (Figure 9). About 14% of the universe of Spanish loans (including arrears and defaulted loans) is in negative equity, with c 17.4% of arrears in negative equity. Arrears likelihood (i.e. arrears as a percent of pool balance in the bucket) increases from c.5% to c.10% as indexed LTV increases from 30% to 90%. We find that some 9.3% of the negative equity loans (to be sure where indexed LTV > 100%) are in arrears indicating broadly that higher LTV loans are indeed seeing higher default rates.

But does the negative equity ratio help address the difference in regional arrears? Indeed negative equity is correlated with regional arrears, as Figure 10 shows. Excluding Murcia, Madrid and Castilla La Mancha which are outliers, some 31% of the variance in regional arrears is explained by negative equity.

However, this simultaneously raises an associated question- some 70% of the regional variance remains unexplained by negative equity. In regions such as Murcia, Catalunya, Balears and Valenciana where we see large arrears as a ratio of the regional pool, only a small percentage of arrears are from the negative equity bucket – In Murcia, arrears where the loan is in negative equity constitutes 1.3 ppts of the regional pool, out of the 18.3% of the regional pool in arrears (See Figure 11).

6 Using regional house price indices from MDV takes into account regional differences and provide more

granularity in estimating negative equity.

Sector average arrears = 7.5%

4 September 2013



Figure 10: Regional arrears versus percentage of pool in

negative equity

Figure 11: Arrears in negative equity loans by region

Murcia (Región

de)

CataluñaBalears (Illes) Comunidad

Valenciana

AndalucíaRioja (La)Aragón

GaliciaCastilla-La

Mancha

Navarra (Com.

Foral de)

Canarias

Madrid

(Comunidad de)

Asturias

(Principado de )Castilla y León

CantabriaExtremadura

País VascoCeuta y Melilla

0%

2%

4%

6%

8%

10%

12%

14%

16%

18%

20%

0% 5% 10% 15% 20% 25% 30% 35% 40%

Arr

ears

as p

erc

enta

ge o

f re

gio

nal

pool

% of pool in negative equity

Outliers

Outlier

0%

2%

4%

6%

8%

10%

12%

14%

16%

18%

20%

Arr

ears

as %

of to

tal p

ool

Arrears in negative equity loans (%) Arrears in positive equity loans (%)



A side note on negative equity and arrears liquidation loss While outside the direct scope of examining credit drivers, we take a short detour to use loan level data to, in effect, mark-to-market loans in arrears. This was previously not possible to do with aggregate investor reporting data and adds another dimension to credit analysis.

If lenders were to liquidate all arrears7 in Spanish pools, the loss across the universe would be just 43 bps (“arrears liquidation loss”). But this average hides the significant variation across deals – with the highest arrears liquidation loss in later vintage deals such as HIPO HIPO 10, TDAC 9 and CAJAM 2007-3 (Figure 12).

7 By liquidation loss, we mean the difference between the liquidation value of the property and the

mortgage. Liquidation value is the cash realised if the lender were to repossess the property backing the

mortgage and auction/sell it in the market to realise cash. We assume a 15% liquidation discount i.e, that

the property is sold for 15% lesser than indexed property value, to account for transaction/legal costs,

house specific features and potential firesale discounts.

4 September 2013



Figure 12: Arrears liquidation loss by deal (highest 30)

0.0%

0.5%

1.0%

1.5%

2.0%

2.5%

3.0%H

IPO

HIP

O-1

0

GR

AN

A 2

007

-1

TD

AC

9

CA

JA

M 2

007-3

KU

TX

H 2

TD

AC

7

SH

IPO

3

CA

JA

M 2

006-2

HIP

O H

IPO

-9

CA

JA

M 2

006-1

IMP

AS

4

BC

JA

F 1

1

TD

AC

6

BC

JA

F 1

0

TD

AC

8

TD

A 2

9

BB

VA

R 2

007-3

BB

K 2

006-2

HIP

O H

IPO

-8

IMC

AJ 4

SH

IPO

2

TD

AI 5

BC

JA

F 9

RH

IPO

9

BC

JA

M 4

IMC

AJ 3

TD

AI 4

TD

A 2

6

BB

VA

R 2

007-1

PE

NE

D 1

Arr

ears

liquid

ation lo

ss a

s %

of to

tal p

ool

Source: Deutsche Bank, European DataWarehouse. We add a further 15% liquidation loss realized during the foreclosure procedure and to account for property specific illiquidity factors.

The c.43 bps average arrears liquidation loss applies to current price declines– indeed further house price deterioration would result in higher losses where arrears are liquidated. If Spanish prices decline another 20% to bring the total house price decline from the peak to c.57% (marginally worse than Ireland where peak to trough prices stand at c.50%), the amount of arrears in negative equity would double, to constitute 40% of total arrears (not accounting for incremental arrears formation) and 3.0% of the total pool.

Arrears and unemployment The employment status of a borrower is defined as a static field which means lenders are not required to update this field on an ongoing basis under ECB taxonomy guidelines.8 Curiously, in some deals we do find some exposure to unemployed borrowers – this is either because the lender has updated the new employment status of the borrower anyway, or that the initial loan was granted to an unemployed borrower. While the latter may appear surprising to readers, our understanding from interactions with lenders is that this can occur if the amount of the loan is low relative to the borrower’s net worth, or when monthly debt service can be met with investment income from accounts held at the lender.

The employment status for only EUR 2.9 billion (6.2% of reported mortgages) is reported as “No Data” with most of the mortgages classified into the available categories– some 62% of outstanding mortgages in RMBS pools were given to fully employed borrowers or where the loan was otherwise guaranteed. 10.9% of loans are against self employed borrowers of which 12.2% are currently in arrears (see Figure 13).

With youth unemployment in excess of 50%, loans to students (16.4% arrears ratio) have seen the highest level of stress, although we must add that student loans constitute just 0.4% of the total mortgage universe. The other most affected segments are loans where the borrower is unemployed (11.3% arrears

8 However, in some cases, a lender’s branches may update the status when it finds out that the situation

has changed, or when the risk of a borrower is reassessed as directed by the lender’s systems.

Nevertheless, it is likely that a number of lenders may not report it on an ongoing basis either because

they do not track employment status or they track it in a different system and do not report it in the loan

tapes.

Average arrears liquidation loss = 43 bps

4 September 2013



ratio), or self-employed (12.2% arrears ratio). As expected the best performance is from borrowers classified as “protected life-time employment” (which are Government employees/public servants) where 4.8% of mortgages are in arrears (Figure 13).

Another way to glean the effect of unemployment is to look at the effect of regional unemployment rates. Plotting arrears percentages versus regional unemployment rates, we find that variation in the unemployment rate captures some 24% of the variance in regional arrears after taking out outliers (Figure 14).

In summary therefore, both the prevalence of negative equity and variation in unemployment rates significantly explain the variation in regional arrears.

Figure 13: Arrears versus employment status Figure 14: Regional arrears versus regional

unemployment rate

Stu

de

nt

Un

em

plo

ye

d

Se

lf-e

mp

loye

d

Em

plo

ye

d w

ith

part

ial su

pp

ort

Pe

nsio

ne

r

Em

plo

ye

d o

r

full

loan

gu

ara

nte

ed

Pro

tecte

d

life

-tim

e

em

plo

ym

en

t

No

em

plo

ym

en

t,

bo

rro

we

r is

leg

al e

ntity

0%

2%

4%

6%

8%

10%

12%

14%

16%

18%

0%

2%

4%

6%

8%

10%

12%

14%

Arr

ears

as a

% o

f e

mp

loym

en

t b

ucke

t (%

)

Bu

cke

t as a

% o

f to

tal p

oo

l

Pool as a % of total pool (LHS) Arrears as a % of employment bucket (RHS)

64.3%

Murcia (Región

de)

Cataluña

Balears (Illes)

Comunidad

Valenciana

Andalucía

Rioja (La)

AragónGalicia

Castilla-La

Mancha

Navarra (Com.

Foral de)Canarias

Madrid

(Comunidad de)Asturias

(Principado de )

Castilla y LeónCantabriaExtremadura

País VascoCeuta y Melilla

0%

2%

4%

6%

8%

10%

12%

14%

16%

18%

20%

10 15 20 25 30 35 40

Arr

ears

as p

erc

enta

ge o

f re

gio

nal

pool

Regional unemployment rate %

Outlier

Outlier



Arrears and Loan-to-Income (LTI) ratio – borrower leverage We now turn to consider the impact of borrower leverage as measured by the Loan-to-Income (LTI) ratio9. At the outset, we are hesitant to draw strong conclusions using income-driven leverage ratios because income values shown in loan tapes are “static” fields – i.e, unless lenders choose to report current incomes voluntarily, these are values recorded at the time of origination of the loan. Although we do index individual income using national wage indices in all our income ratios (See Appendix B for data on the historical evolution of average nominal wages in Spain) to capture the average change in income since origination, we would still not capture loans where the borrower is unemployed or has suffered an income shock.

That said, about 4.8 ppts of the 18.3% (Figure 15) arrears from Murcia are from loans where the leverage is higher – i.e. where the LTI ratio is greater than 4. This explains a higher proportion of regional arrears than those in negative equity which constituted just 1.3 ppts. Similarly, Catalunya and Andalucia have 4.8 ppts and 3.1 ppts of their pools in arrears which have a loan to income ratio greater than 4.

9 We define Loan to Income as the ratio of current loan balance to indexed income. We index the incomes

using historical data from OECD on average nominal wages in Spain, as we describe in Appendix II.

4 September 2013



Figure 15: Arrears where loan to income is high Figure 16: Property Occupancy status in Spanish

mortgages

0%

2%

4%

6%

8%

10%

12%

14%

16%

18%

20%

Arr

ears

as %

of to

tal p

ool

Arrears with LTI > 4 as % of regional pool

Arrears with LTI<4 as % of regional pool

0%

10%

20%

30%

40%

50%

60%

70%

80%

Ow

ner-

occupie

d

Part

ially

ow

ner-

occupie

d (A

pro

pert

y w

hic

h is

part

ly rente

d)

Non-o

wner-

occupie

d/b

uy

-to-let

Holiday/s

econd

hom

e Oth

er

No D

ata

Pool as a % of total pool Arrears as a % of pool

Source: Deutsche Bank, European DataWarehouse

Source: Deutsche Bank, European DataWarehouse

In aggregate, this suggests that unemployment and income data are important determinants of mortgage performance, albeit a lot of information is lost because current income or employment status may not be accurately reflected by loan tapes.

Arrears and type of property Property occupancy data is present only in EUR 16.4 billion of mortgages (nearly 32% of the sample set of 97 deals). From the data we infer that 15% of loans towards holiday or second homes are in arrears and remain the worst affected sector, but only 3.5% of loans have been allocated to this sector. Owner occupied loans show 11% of arrears and form 28% of the pool (Figure 16). It is to be noted that the high percentage of arrears (60%) in loans where property occupancy data is not present strongly suggests adverse selection – i.e, where lenders have not recorded (or recorded but not reported) property occupancy status, there is a higher likelihood of arrears. This remains another area where reporting can improve.

Additionally, given the popularity of the Spanish property market with foreign buyers, we were hoping that residency status data may shed further light. Although the ECB template for Spanish RMBS reporting provides for residency to be reported, this is not one of the mandatory fields and we find that only 30% of deals (28% by volume) report residency status. We therefore do not delve further into borrower residency status.

What does this mean at a deal level?

Now that we know what loan characteristics have been prominent in explaining arrears performance, we can define risk factors for each deal. We calculate the following risk factors for each deal in our sample universe below. a) the level of exposure to the worst 5 affected regions – Catalunya, Murcia, Andalucia, Valenciana and Balears (Illes). b) the level of exposure to the worst 3 employment types – students, unemployed and self-employed borrowers, c) the proportion of loans in the pool in negative equity, d) the proportion of borrowers with a LTI ratio > 4.

4 September 2013



We present the 97 deals in this study in the table below and highlight the four mentioned risk factors as well as the estimated arrears liquidation loss and liquidation loss if all loans (even those not in arrears) that have a debt service to gross indexed income ratio greater than 30% is liquidated. Note that the data we present is influenced by data quality – there could be loans where data is not disclosed relative to regions, property value, income or loan amount which would influence our numbers.

We reiterate that arrears as calculated using loan tape data is different from that calculated using investor report data. Possible reasons for differences include different sample universes, timing differences, potential difficulties in identifying loans in arrears or slippages in loan level data in recording loans in arrears.

Figure 17: Spanish deal universe from the European DataWarehouse – A summary of risk factors

Key credit metrics Risk Factors

Deal Loan tape date

Total arrears (as per loan

tape)

Latest total arrears (IR)

Arrears liquidation loss as % of total

pool

Liquidation loss for DTI > 30% as % of

total pool

Worst 5 regions as % of total pool

Worst 3 employment

status as % of total pool

Negative equity as % of total

pool

Loan to Income> 4

as % of total pool

AYTCH I 10-Jan-13 9.0% 7.2% 0.0% 0.0% 0% 13% 0% 5% AYTCH II 27-Mar-13 3.8% 3.7% 0.0% 0.0% 0% 13% 0% 0% AYTGH II 30-Apr-13 0.6% 2.4% 0.0% 0.0% 32% 11% 0% 0% AYTGH III 30-Apr-13 0.6% 1.8% 0.0% 0.0% 37% 11% 0% 0% AYTGH IV 30-Apr-13 0.3% 1.3% 0.0% 0.0% 38% 10% 0% 0% AYTGH IX 30-Apr-13 1.0% 2.8% 0.1% 0.0% 52% 12% 3% 0% AYTGH VI 30-Apr-13 0.5% 2.5% 0.0% 0.0% 44% 12% 0% 0% AYTGH VII 30-Apr-13 0.7% 2.2% 0.0% 0.0% 52% 10% 1% 0% AYTGH VIII 30-Apr-13 0.7% 2.2% 0.0% 0.0% 47% 11% 1% 0% AYTGH X 30-Apr-13 1.4% 3.9% 0.1% 0.0% 57% 12% 6% 0% AYTH M2 12-Mar-13 10.2% 5.3% 0.0% 0.0% 0% 11% 0% 0% BBK 2005-1 30-Apr-13 6.3% 5.1% 0.1% 0.2% 3% 23% 3% 26% BBK 2006-2 30-Apr-13 6.0% 5.3% 0.5% 1.3% 4% 20% 11% 48% BBVAR 2007-1 12-Mar-13 1.7% 11.3% 0.2% 3.3% 0% 10% 20% 82% BBVAR 2007-2 12-Mar-13 2.3% 12.0% 0.1% 0.6% 0% 17% 1% 72% BBVAR 2007-3 14-Feb-13 3.9% 18.2% 0.7% 9.6% 0% 10% 61% 85% BCJAF 10 28-Feb-13 8.8% 12.9% 0.8% 5.9% 74% 12% 23% 82% BCJAF 11 31-Jan-13 9.2% 14.1% 1.1% 7.8% 75% 13% 35% 85% BCJAF 3 31-Mar-13 1.8% 6.4% 0.0% 0.0% 83% 9% 0% 12% BCJAF 4 31-Mar-13 2.2% 7.5% 0.0% 0.0% 76% 0% 0% 0% BCJAF 5 31-Jan-13 1.8% 5.7% 0.0% 0.0% 73% 1% 0% 1% BCJAF 6 28-Feb-13 1.8% 6.7% 0.0% 0.0% 77% 4% 0% 46% BCJAF 7 28-Feb-13 2.8% 8.3% 0.0% 0.0% 71% 17% 0% 50% BCJAF 8 31-Jan-13 5.7% 11.2% 0.1% 0.6% 71% 15% 2% 66% BCJAF 9 31-Mar-13 8.3% 13.7% 0.3% 2.0% 73% 14% 7% 72% BCJAM 1 28-Feb-13 3.8% 11.2% 0.0% 0.0% 87% 20% 0% 26% BCJAM 2 28-Feb-13 5.3% 12.0% 0.0% 0.0% 83% 30% 0% 36% BCJAM 3 31-Mar-13 6.2% 13.4% 0.1% 0.5% 81% 28% 2% 45% BCJAM 4 31-Jan-13 7.5% 15.7% 0.2% 1.1% 83% 27% 5% 49% BFTH 10 31-Mar-13 0.8% 4.2% 0.0% 0.0% 39% 9% 1% 14% BFTH 11 28-Feb-13 0.8% 3.4% 0.0% 0.0% 42% 9% 0% 18% BFTH 13 31-Jan-13 1.5% 5.7% 0.0% 0.6% 42% 11% 10% 31% BFTH 2 31-Jan-13 0.7% 4.6% 0.0% 0.0% 26% 7% 0% 0% BFTH 3 31-Jan-13 0.5% 3.4% 0.0% 0.0% 35% 8% 0% 0% BFTH 4 28-Feb-13 0.2% 3.4% 0.0% 0.0% 38% 6% 0% 0% BFTH 5 28-Feb-13 0.5% 3.3% 0.0% 0.0% 36% 8% 0% 1% BFTH 6 28-Feb-13 0.6% 3.7% 0.0% 0.0% 43% 7% 0% 1% BFTH 7 31-Mar-13 0.8% 5.4% 0.0% 0.0% 36% 10% 0% 3% BFTH 8 31-Mar-13 1.1% 4.6% 0.0% 0.0% 40% 10% 0% 2% BFTH 9 31-Jan-13 0.9% 4.0% 0.0% 0.0% 45% 9% 0% 9% BVA 1 25-Feb-13 1.0% 6.2% 0.0% 0.0% 0% 9% 0% 29% BVA 2 18-Jan-13 3.4% 9.3% 0.0% 0.0% 0% 12% 0% 50% BVA 3 18-Mar-13 4.0% 7.9% 0.0% 0.1% 0% 15% 0% 61%

4 September 2013



CAJAM 2006-1 28-Feb-13 4.7% 18.6% 1.1% 17.5% 18% 6% 66% 91% CAJAM 2006-2 28-Feb-13 5.2% 21.1% 1.2% 16.5% 19% 7% 63% 90% CAJAM 2007-3 28-Feb-13 6.0% 22.3% 1.4% 17.6% 26% 7% 72% 93% CLAB 2006-1 31-Mar-13 3.8% 3.3% 0.0% 0.0% 0% 22% 1% 16% GRANA 2007-1 28-Feb-13 35.0% 34.0% 1.9% 0.0% 0% 23% 14% 0% HIPO HIPO-10 24-Apr-13 21.2% 23.9% 2.6% 4.2% 82% 17% 26% 51% HIPO HIPO-4 15-Mar-13 3.5% 2.5% 0.0% 0.0% 90% 18% 0% 0% HIPO HIPO-5 15-Apr-13 6.3% 4.0% 0.0% 0.0% 92% 16% 0% 8% HIPO HIPO-6 15-Mar-13 8.0% 8.0% 0.0% 0.0% 93% 13% 0% 15% HIPO HIPO-7 15-Apr-13 12.7% 11.7% 0.0% 0.0% 81% 16% 0% 21% HIPO HIPO-8 15-Mar-13 14.9% 14.1% 0.5% 0.5% 91% 18% 2% 32% HIPO HIPO-9 15-Apr-13 17.2% 17.3% 1.2% 1.7% 83% 19% 8% 44% HMSF X 28-Mar-13 1.9% 1.9% 0.0% 0.0% 45% 4% 0% 7% HMSF XI 28-Mar-13 3.2% 2.3% 0.0% 0.0% 50% 5% 0% 20% IMCAJ 3 28-Feb-13 18.0% 14.4% 0.2% 0.2% 89% 11% 1% 30% IMCAJ 4 28-Feb-13 16.8% 13.7% 0.4% 0.6% 91% 10% 3% 37% IMPAS 2 28-Feb-13 13.7% 13.5% 0.0% 0.0% 43% 36% 0% 4% IMPAS 3 28-Feb-13 36.5% 36.3% 0.0% 0.0% 56% 44% 0% 17% IMPAS 4 28-Feb-13 33.4% 33.5% 1.1% 0.7% 52% 45% 5% 23% KUTXH 1 30-Apr-13 2.2% 3.1% 0.1% 0.4% 17% 15% 6% 42% KUTXH 2 30-Apr-13 8.1% 10.5% 1.4% 3.9% 20% 16% 23% 60% PENED 1 31-Dec-12 8.6% 9.9% 0.2% 0.0% 99% 46% 2% 0% RHIPG I 11-Jan-13 3.8% 13.4% 0.0% 0.1% 0% 37% 0% 24% RHIPO 3 13-Mar-13 0.8% 7.8% 0.0% 0.0% 0% 21% 0% 5% RHIPO 4 13-Feb-13 1.2% 10.5% 0.0% 0.0% 0% 21% 0% 8% RHIPO 5 15-Mar-13 1.4% 8.5% 0.0% 0.0% 0% 22% 0% 15% RHIPO 6 10-Jan-13 2.0% 9.6% 0.0% 0.0% 0% 21% 0% 19% RHIPO 7 08-Mar-13 2.3% 10.3% 0.0% 0.0% 0% 28% 0% 20% RHIPO 8 14-Jan-13 2.9% 12.7% 0.0% 0.1% 0% 25% 0% 27% RHIPO 9 12-Feb-13 5.7% 17.6% 0.2% 0.8% 0% 27% 6% 35% SABA 1 28-Feb-13 10.3% 7.8% 0.0% 0.0% 53% 13% 0% 0% SHIPO 1 30-Apr-13 2.9% 2.3% 0.0% 0.0% 51% 2% 0% 46% SHIPO 2 30-Apr-13 4.7% 4.7% 0.3% 2.3% 44% 3% 14% 62% SHIPO 3 30-Apr-13 7.0% 7.6% 1.2% 6.5% 42% 1% 49% 76% TDA 19 28-Feb-13 19.7% 41.3% 0.0% 0.0% 96% 0% 0% 0% TDA 26 31-Dec-12 11.6% 11.5% 0.2% 0.0% 52% 10% 1% 0% TDA 29 31-Jan-13 12.7% 12.4% 0.7% 0.0% 59% 15% 13% 0% TDAC 1 28-Feb-13 9.1% 8.6% 0.0% 0.0% 95% 14% 0% 0% TDAC 2 31-Mar-13 11.5% 10.7% 0.0% 0.0% 86% 14% 0% 5% TDAC 3 31-Mar-13 12.4% 12.1% 0.0% 0.0% 94% 13% 0% 4% TDAC 4 28-Feb-13 12.9% 13.3% 0.0% 0.0% 90% 13% 0% 0% TDAC 5 31-Mar-13 17.1% 16.9% 0.1% 0.0% 83% 11% 1% 10% TDAC 6 31-Mar-13 24.9% 24.3% 0.8% 0.3% 83% 12% 8% 13% TDAC 7 31-Jan-13 22.1% 21.6% 1.2% 0.0% 82% 11% 17% 0% TDAC 8 31-Jan-13 19.8% 21.4% 0.7% 0.0% 90% 14% 9% 0% TDAC 9 31-Mar-13 26.5% 26.0% 1.8% 4.0% 91% 15% 22% 57% TDAI 1 31-Jan-13 1.4% 9.8% 0.0% 0.0% 0% 17% 0% 13% TDAI 2 31-Jan-13 2.5% 11.2% 0.0% 0.0% 0% 20% 0% 38% TDAI 3 02-Apr-13 3.5% 10.2% 0.0% 0.4% 0% 18% 1% 53% TDAI 4 28-Feb-13 5.0% 16.7% 0.2% 1.6% 0% 20% 5% 59% TDAI 5 28-Feb-13 4.8% 14.7% 0.3% 2.8% 0% 21% 15% 65% TDAP 1 28-Feb-13 10.3% 10.9% 0.0% 0.0% 32% 0% 0% 0% TDCAJ 2 28-Feb-13 15.8% 15.2% 0.0% 0.0% 94% 16% 0% 16% UCI 9 12-Mar-13 4.4% 3.9% 0.0% 0.0% 0% 9% 0% 20% Source: Deutsche Bank, European DataWarehouse

Spanish Senior RMBS Relative Value

For relative value in Spanish RMBS, we selected 30 recently traded senior Spanish bonds where loan level data is available in the European DataWarehouse. Please see Appendix D for details about individual bonds and selected credit metrics of the deals used in this section.

4 September 2013



We classify bonds trading under 300 bps as Tier 1 – these constitute what are perceived to be among the safest bonds in the Spanish universe, or those with a potential tender optionality. Tier 2 bonds are defined as offering a spread under 375 bps but over 300 bps – these are perceived to be of inferior quality or from weaker originators, and Tier 3 bonds are those that trade over 375 bps in the current environment and trade wide due to credit impairment, or shelf or name–specific investor risk aversion. Note that our classification of tiers is not on any “scientific” basis, but based on what we think subjectively as appropriate – there could well be bonds (and we highlight a few) which should belong to the tighter spread sector, therefore offering value.

Following this tiering, we note that Spanish 3rd tier senior bonds offer a pick-up of anywhere between 50-225 bps over top tier senior bonds (Figure 18).

We find that the market prices bonds broadly in line with the ratio of credit enhancement of the bond to the total arrears in the pool (Figure 18). As the credit enhancement increases relative to total arrears, the spread is found to decrease. However, this ignores the fact that although two deals can have the same total arrears percentage, they can differ significantly in the ultimate degree of loss. Loans in arrears in one deal could be from loans where there is substantial equity in the property, hence resulting in lower losses. Additionally, this does not take into account prospective arrears formation i.e. one of the deals could have a higher chance of deteriorating credit performance, while the other may not.

In order to overcome these shortcomings, in each deal, we liquidated all loans with a debt service to gross indexed income (DTI) ratio > 30%. While the measure has some shortcomings in the form of using indexed income and not capturing loan specific income shock and unemployment, it still has information value related to aggregate borrower leverage. At an aggregate level, one can expect that a pool originated at a higher DTI will, over the course of the deal life and all else being equal, end up with higher arrears. The liquidation value10 accounts for regional differences in house prices as well as the individual loan value and gives a more accurate picture of the loss. This should act as a better factor for pool credit quality than just total current arrears percentage since both borrower leverage and house price impact is accounted for, in addition to being a metric for prospective arrears. Note that this has been made possible by loan level data, something that is not possible to calculate using aggregate data from investor reports.

Figure 19 shows the same bonds, but this time the X-axis is set at our new bond safety metric – the ratio of credit enhancement (CE) to liquidation loss of the loan subpool where DTI > 30%11. Instead of plotting the absolute metric on the X-axis, we scaled this number to 100- the deals with the highest CE to the above liquidation loss metric are at 100 (higher bond safety), and the lowest CE to liquidation loss metric are close to 012(lower bond safety).

10 For liquidation value, we assume a that the property is sold at a discount of 15% to indexed property

value. This is meant to account for firesale risks, property specific features and legal/transaction costs. 11

We have chosen a 30% threshold as it is broadly accepted as a general thumbrule for borrowers to get

into distress –any higher threshold above 30% such as 35% and 40% may also be chosen. There is not a

proven threshold level – too high a threshold and one is likely to miss a number of borrowers who may get

into distress, and too low a threshold, and one is likely to capture borrowers who are easily able to afford

the mortgage. 12

This removes the difficulties associated with very small liquidation loss percentages which can result in

the ratio being very high. It would make chart interpretation difficult. The scaled information captures the

relevant data adequately for relative value purposes.

4 September 2013



In the charts below, we show WAL of the bond within brackets after the name of the bond in the label.

Figure 18: Spread versus (CE to total arrears ratio) Figure 19: Spread versus (CE to DTI > 30% Liquidation

loss ratio)

AYTGH IX A2

AYTGH IV ABFTH 13 A2

BVA 3 A2

BCJAF 4 A

BCJAF 10 A2

BCJAM 4 A2

CAJAM 2006-1 A2

SABA 1 A2

TDAC 8 A

TDAC 9 A2

UCI 9 A

HIPO HIPO-8 A2

BFTH 10 A2

BCJAF 6 A2TDAC 4 A

HIPO HIPO-7 A2

TDAI 4 A2

AYTGH II ABFTH 3 A

BFTH 4 ABFTH 6 ABFTH 7 A

BFTH 8 A

SHIPO 1 ASHIPO 2 ATDAC 3 A

RHIPO 8 A2A

225

275

325

375

425

475

525

0 10 20 30 40 50

Sp

read

(b

ps)

Credit enhancement % / Total arrears as % of pool

AYTGH IX A2 (8.0)

AYTGH IV A (6.2)BFTH 13 A2 (8.0)

BVA 3 A2 (6.2)

BCJAF 4 A (5.0)

BCJAF 10 A2 (5.0)

BCJAM 4 A2 (8.0)

CAJAM 2006-1 A2

(7.0)

SABA 1 A2 (5.3)

TDAC 8 A (6.4)

TDAC 9 A2 (6.7)

UCI 9 A (6.9)

HIPO HIPO-8 A2 (5.6)

AYTGH IV A (6.2)

BFTH 10 A2 (7.6)

BCJAF 6 A2 (4.6)TDAC 4 A (4.7)

HIPO HIPO-7 A2 (6.3)

TDAI 4 A2 (8.5)

AYTGH II A (6.0)

BFTH 3 A (4.0)

BFTH 4 A (5.0)BFTH 6 A (6.5)BFTH 7 A (6.5)BFTH 8 A (6.7)

SHIPO 1 A (4.5)

SHIPO 2 A (7.0)

TDAC 3 A (3.5)

HIPO HIPO-5 A (6.0)RHIPO 8 A2A (6.0)

225

275

325

375

425

475

525

0 20 40 60 80 100 120

Sp

rea

d (

bp

s)

Credit enhancement / Liquidation loss of DTI > 30% loans scaled to 100

Source: Deutsche Bank, Indicative spread levels, European DataWarehouse. Total Arrears include defaults-in-hand and reflect total arrears from loan tapes

Source: Deutsche Bank, Indicative spread levels, European DataWarehouse. Note that deals where the liquidation loss is zero are assigned an X-axis value of 100..

This method throws up interesting relative value areas in the Spanish senior RMBS sector to look further into, some of which are highlighted by the red ovals. We look for bond switches where a bond A that is trading tighter has a lower ratio of credit enhancement to the liquidation loss for loans with DTI > 30% relative to another bond B. In such a scenario, an investor could switch to bond B from bond A, pickup spread, and not compromise on bond safety.

As a double check, we nevertheless also look at the ratio of credit enhancement to total arrears. This is to avoid the pitfalls of basing our recommendations on a single liquidation loss metric. We make our recommendations in the following section.

Summary and recommendations

Mining loan level data allows for a new layer of credit analysis which can add substantial value not discernible using aggregate data from investor reports. Using loan-level data, we compare credit quality for 97 deals across the Spanish RMBS universe and identify the extent of stressed mortgages based on various risk factors – geography, LTV (current indexed), employment status, loan-to-income ratios, and property occupancy type.

We find that the regions on the Mediterranean coast – Andalucia, Murcia, Valenciana and Catalunya are the worst hit regions with regional arrears percentages close to or well over 10%. Unemployment rates and regional house price declines remain prominent explanatory factors for regional variation in arrears performance.

We identify relative value based on the ratio of credit enhancement relative to liquidation loss of loans with DTI > 30%, and secondary spreads. As a double check, we also look to see that we do not compromise on bond safety as

Tier 1

Tier 2

Tier 3 Tier 3

Tier 2

Tier 1

4 September 2013



measured by the ratio of credit enhancement to total arrears13. We also show whether the deals have been previously tendered – if they were tendered, the bonds are very likely to be consolidated on the sponsor’s balance sheet making future tenders a possibility. If bonds have not been tendered previously, unless explicitly mentioned in the sponsor’s financial reports, it is hard to be sure whether the bonds are consolidated.

We recommend the following four bond switches among senior Spanish RMBS:– Spread levels are indicative only, and bid-offer basis may reduce the spread pickup available in the switch. Although spread pickup looks marginal in the switches, we think that the switches offer value in terms of moving to a bond with greater safety.

a) Switch from BFTH 13 A2 to BCJAF 6 A2 to pickup 15 bps

Rationale – Switch to BCJAF 6 A2 to pickup c15bps in spread and at the same time reduce WAL. Credit enhancement of the BCJAF 6 bond at 22% is well above that of BFTH 13 A2, with liquidation loss of the former being nearly 0% as noted below.

Figure 20: BFTH 13 A2 to BCJAF 6 A2

Bond Tier Tendered? DM (bp) WAL CE of bond Liquidation Loss for DTI > 30%

Total Arrears as % of total

pool*

CE/ Liquidation

Loss

CE/Total arrears

Vintage Action

BFTH 13 A2 Tier 1 No 300 8.4 11.0% -0.6% 5.7% 19.5 1.9 2006 Switch from

BCJAF 6 A2 Tier 2 No 315 4.3 22.4% 0.0% 6.7% 12,267.1 3.3 2003 Switch to Source: Deutsche Bank, Indicative Spreads and WAL, European Data Warehouse * Total arrears includes loans in defaults. The higher number from investor report and loan tape is considered here to be conservative

b) Switch from SABA 1 A2 to HIPO HIPO-5 A to pickup 15 bps

Rationale – Switching to HIPO HIPO 5 A would provide a spread pickup of 15 bps while moving to a cleaner pool (total arrears at 6.3% lower than SABA 1 where it is 10.3%). Credit enhancement of the former bond is much higher at c.20% compared to the 5.8% credit enhancement of SABA 1 A2.

Figure 21: SABA 1 A2 to HIPO HIPO-5 A



pool*

CE/ Liquidation

Loss

CE/Total arrears

Vintage Action

SABA 1 A2 Tier 2 Yes 310 5.4 5.8% 0.0% 10.3% NA 0.6 2004 Switch from

HIPO HIPO-5 A

Tier 2 Yes 325 6.0 19.9% 0.0% 6.3% NA 3.1 2002 Switch to

Source: Deutsche Bank, Indicative Spreads and WAL, European Data Warehouse * Total arrears includes loans in defaults. The higher number from investor report and loan tape is considered here to be conservative

c) Switch from BCJAM 4 A2 to RHIPO 8 A2A to pickup 10 bps

Rationale – Pickup 10 bps in spread terms by moving to RHIPO 8 A2A which is also shorter in WAL by about 2 years. While the credit enhancement of RHIPO 8 A2A is lower by about 1.3ppts, total arrears at 12.7% in RHIPO 8 is lower than the 15.7% total arrears in BCJAM 4 A2. Liquidation loss (for DTI>30%) in RHIPO 8 also stands lower at close to 0.1% compared to a 1.1% estimated liquidation loss for BCJAM 4 A2.

13In order to be conservative, where total arrears using the loan tape is different from total arrears using

the investor report, we have used the higher of the two numbers in calculating the CE/Total arrears ratio.

4 September 2013



Figure 22: BCJAM 4 A2 to RHIPO 8 A2A



pool*

CE/ Liquidation

Loss

CE/Total arrears

Vintage Action

BCJAM 4 A2

Tier 2 Yes 325 8.0 13.1% -1.1% 15.7% 12.1 0.8 2007 Switch from

RHIPO 8 A2A

Tier 2 Yes 335 6.0 11.8% -0.1% 12.7% 150.6 0.9 2006 Switch to

Source: Deutsche Bank, Indicative Spreads and WAL, European Data Warehouse * Total arrears includes loans in defaults. The higher number from investor report and loan tape is considered here to be conservative

d) Switch from BVA 3 A2 to HIPO HIPO-5 A

Rationale-While there is just 9 bps pickup in spread from the switch, HIPO HIPO-5 has a lower estimated liquidation loss and the senior bond has a higher attachment point (nearly 2x) compared to that of BVA 3. CE/liquidation loss as well as CE/total arrears ratio of BVA 3 A2 stands lower than HIPO HIPO-5 A making the switch attractive.

HIPO HIPO-5 has also been tendered before by the sponsor (Catalunya Banc) while BVA 3 has not been tendered previously (although we note that the deal is consolidated on sponsor’s books). While admittedly not certain, it opens up the possibility that the Hipocat 5 deal has a higher tender optionality given repeat tenders from sponsors that have been coming through in the sector.

Figure 23: BVA 3 A2 to HIPO HIPO-5 A

Bond Tier Tendered? DM (bp) WAL CE of bond Liquidation Loss for

DTI > 30%

Total Arrears as % of

total pool*

CE/ Liquidation

Loss

CE/Total arrears

Vintage Action

BVA 3 A2** Tier 2 No 316 6.7 10.1% -0.1% 7.9% 182.4 1.3 2006 Switch from

HIPO HIPO-5 A Tier 2 Yes 325 6.0 19.9% 0.0% 6.3% NA 3.1 2002 Switch to Source: Deutsche Bank. * Total arrears includes loans in defaults. The higher number from investor report and loan tape is considered here to be conservative **While BVA 3 has not been tendered before, we note that the deal is consolidated on the books of CaixaBank. Indicative Spreads and WAL, European Data Warehouse

Risks: Worse than expected credit performance in deals from higher unemployment rate or a persistent deep property downcycle are risks to our recommendations. Spreads and yields are contingent on prepayment and default rates. Bond specific illiquidity and technical factors as well as deal specific risks such as those associated with sponsors and counterparties are additional risks to bear in mind.

Note that spread levels in the relative value tables above are indicative, and are sensitive to prepayment rate, default rate, and loss severity assumptions. A 2.5% - 4.5% CPR is assumed to derive DMs for the bonds using the Bloomberg cash flow model. The above indicated spread levels are also driven by when (and if) pro-rata triggers are toggled which depends on credit performance, and spread levels may change materially depending on trigger assumptions.

4 September 2013



APPENDIX A

The following table shows the number of deals in each programme for which the European DataWarehouse has loan level data and current sponsors of Spanish RMBS programmes taking into account the consolidation in the Spanish banking system.

Figure 24: European DataWarehouse coverage by Programme

BBG Ticker # placed deals Original Sponsor Investor placed deals Covered by European DataWarehouse

New Sponsor (due to banking sector consolidation)

AYT 3 Vital Kutxa 0 Kutxa Bank

AYTCH 2 Caja Murcia 2 Banco Mare Nostrum

AYTGH 8 Barclays 8 Barclays

AYTH 6 Banca Civica ,Caja Granada ,Spanish savings banks ,CajaRioja ,Caja Vital ,ElMonte ,Banco

Gallego ,Credifimo ,Caixa de Manlleu ,Banesto

1 Caixabank ,Banco Mare Nostrum ,Spanish savings

banks ,Bankia ,Kutxa Bank ,Banco

Sabadell ,BBVA ,Banco Santander

BANES 1 Banesto 0 Banco Santander

BBVAR 3 BBVA 3 BBVA

BCJAF 9 Bancaja 9 Bankia

BCJAM 4 Bancaja 4 Bankia

BFTH 11 Bankinter 11 Bankinter

BVA 3 Banco de Valencia 3 Caixabank

CAJAM 3 Caja Madrid 3 Bankia

CLAB 1 Caja Laboral 1 Caja Laboral Popular

GCPAS 5 1 Banco Pastor 0 Banco Popular

HIPO HIPO 8 Diada 7 Catalunya Banc

HMSF 2 Santander 2 Banco Santander

IMCAJ 3 Cajamar 2 Cajamar Caja Rural

IMPAS 3 Banco Pastor 2 Banco Popular/bancopopular-e.com

IMTEM 1 Unnim 0 BBVA

KUTXHA 2 Kutxa 2 Kutxa Bank

PENED 2 Caixa Penedes 1 Banco Sabadell

RHIPG 1 Spanish savings banks 1 Spanish savings banks

RHIPO 7 Caja Rural del Sur SCC ,Caja de Ahorros y Monte de Piedad de Ronda ,Cadiz ,Almeria ,Unicaja ,Caja

Rural del Sur SCC ,Spanish savings banks

7 Caja Rural del Sur SCC,Unicaja Banco,Spanish savings banks

SABA 1 Banco Sabadell 1 Banco Sabadell

SHIPO 3 Santander 3 Banco Santander

TDA 13 Kutxa ,Caja de Ahorros Provincial San Fernando de Sevilla y Jerez,Banco

Guipuzcoano,Unnim,Cajamar,SA Nostra ,Caja Caminos ,Caja Ingenieros ,Barclays,Banco

Gallego,Credifimo,Banca Civica,Diada,Banca March,Caja de Ahorros de Castil la

Mancha,Bankpime,Credifimo,Caja Granada,Vital Kutxa

3 Kutxa Bank ,CaixaBank,Banco Sabadell,BBVA,Cajamar Caja

Rural,Banco Mare Nostrum,Banco Caminos,Caja Ingenieros,Barclays,Catalunya Banc,Banca March,Liberbank

TDAC 9 Caja Mediterraneo 9 Banco Sabadell

TDAI 5 Ibercaja 5 Ibercaja/Banco Grupo Cajatres (In process)

TDAP 1 Banco Pastor 1 Banco Popular

TDCAJ 1 Cajamar 1 Cajamar Caja Rural

UCI 9 UCI 1 Banco Santander/BNP Paribas Source: Deutsche Bank, European DataWarehouse, deal documents. Consolidation details taken from deal documentation,annual reports of Spanish banks, rating agency data, FROB website, Bank of Spain data.

4 September 2013



APPENDIX B

We present below the annual average wages in Spain from OECD Data. The data represents nominal wages until 2011. In order to complete the indexation to 2013, we use the Ministerio de Hacienda’s real wages index in 2012 and 2013 and convert the figures to nominal percentage changes using the Spanish GDP deflator from Eurostat.

Figure 25: Spanish nominal wage evolution

15000

17000

19000

21000

23000

25000

27000

29000

1995 1997 1999 2001 2003 2005 2007 2009 2011 2013

Curr

ent avera

ge a

nnual w

ag

es

Year % y-o-y2005 3.83%

2006 3.20%

2007 4.48%2008 6.38%

2009 4.99%

2010 2.33%2011 2.06%

2012 -4.82%

2013 -3.75%

Source: Deutsche Bank, Minsterio de Hacienda, OECD Statistics, Eurostat

.

Sp

ecia

l Rep

ort: S

pan

ish R

MB

S: M

inin

g lo

an

level d

ata

4 S

ep

tem

ber 2

01

3

Deu

tsch

e B

an

k A

G/L

on

do

n

Pag

e 2

1

APPENDIX C

We present a table below which is a snapshot of the key fields extracted from the loan tape of a randomly selected Spanish RMBS deal

Figure 26: Snapshot of key fields in loan level data

LoanID OriginalLTV

Primary

Income

Secondary

Income

Valuation

Amount

OriginationDate

Account

Status

Months

In

Arrears

Current

Interest

Rate

Loan

Term

Original

Balance

Current

Balance

Employment

Status

Occupancy Geography

XXXXX1 68 33439 0 105928.38 01-Oct-01 1 (Performing)

0 1.28 240 72,121 19,794

1 (Employed or full loan is guaranteed)

1 (Owner-occupied)

ES511 (Cataluña)


0 2.06 274 144,243 11,313


1 (Owner-occupied)

ES300 (Madrid (Comunidad de))

XXXXX3 26 33662 0 177058.17 01-Jul-01 1 (Performing)

0 1.7 180 46,879 12,838

3 (Protected life-time employment)

1 (Owner-occupied)

ES614 (Andalucía)

XXXXX4 73 32914 0 112389.26 01-Jun-01 1 (Performing)

0 1.81 300 81,738 51,889

2 (Employed with partial support (company subsidy))

1 (Owner-occupied)

ES120 (Asturias (Principado de ))

XXXXX5 52 36194 0 115394.32 01-Jun-01 1 (Performing)

0 1.75 300 60,101 29,028


1 (Owner-occupied)

ES511 (Cataluña)


0 1.49 420 36,061 28,080

2 (Employed with partial support (company subsidy))

1 (Owner-occupied)

ES300 (Madrid (Comunidad de))

XXXXX7 44 35582 0 245813.95 01-Dec-00 1 (Performing)

0 0.99 300 108,182 66,527

9 (Other) 1 (Owner-occupied)

ES111 (Galicia)

XXXXX8 78 97503 0 209044.03 01-Nov-00 1 (Performing)

0 1.05 360 162,273 106,482

9 (Other) 1 (Owner-occupied)

ES511 (Cataluña)

Source: Deutsche Bank, European DataWarehouse. Loan ID and Deal name withheld to avoid license conflicts

4 September 2013



APPENDIX D

Bonds for relative value analysis

In the table below, we show the sample set of 30 bonds that were used in our relative value study. Details about bond credit enhancement as well as deal credit metrics such as total arrears and liquidation loss metrics are also presented.

Figure 27: Bonds for relative value analysis

Bond Loan tape date

Spread WAL Tendered?

Tier Credit Enhance

ment

Liquidation loss

DTI>30%

Total arrears (as per

loan tape)

Latest total arrears (IR)

Arrears liquidatio

n loss

Vintage CE/liquidation loss

DTI>30%

CE/total arrears

AYTGH II A Apr-13 270 6.0 No 1 9.9% 0.0% 0.6% 2.4% 0.0% 2003 NA 4.2 AYTGH IV A Apr-13 280 6.2 No 1 9.5% 0.0% 0.3% 1.3% 0.0% 2004 NA 7.3 AYTGH IX A2 Apr-13 300 8.0 No 2 8.3% 0.0% 1.0% 2.8% -0.1% 2006 NA 2.9 BCJAF 10 A2 Feb-13 475 5.0 Yes 3 13.1% -5.9% 8.8% 12.9% -0.8% 2007 2 1.0 BCJAF 4 A Mar-13 280 5.0 No 1 9.7% 0.0% 2.2% 7.5% 0.0% 2002 NA 1.3 BCJAF 6 A2 Feb-13 315 4.3 No 2 22.4% 0.0% 1.8% 6.7% 0.0% 2003 12,267 3.3 BCJAM 4 A2 Jan-13 325 8.0 Yes 2 13.1% -1.1% 7.5% 15.7% -0.2% 2007 12 0.8 BCJAM 4 A3* Jan-13 657 2.2 Yes 3 13.1% -1.1% 7.5% 15.7% -0.2% 2007 12 0.8 BFTH 10 A2 Mar-13 290 7.6 No 1 10.4% 0.0% 0.8% 4.2% 0.0% 2005 667 2.5 BFTH 13 A2 Jan-13 300 8.4 No 1 11.0% -0.6% 1.5% 5.7% 0.0% 2006 20 1.9 BFTH 3 A Jan-13 260 4.0 No 1 17.7% 0.0% 0.5% 3.4% 0.0% 2001 NA 5.2 BFTH 4 A Feb-13 265 5.0 No 1 12.4% 0.0% 0.2% 3.4% 0.0% 2002 NA 3.7 BFTH 6 A Feb-13 275 6.5 No 1 12.1% 0.0% 0.6% 3.7% 0.0% 2003 NA 3.2 BFTH 7 A Mar-13 285 6.5 No 1 11.5% 0.0% 0.8% 5.4% 0.0% 2004 NA 2.1 BFTH 8 A Mar-13 285 6.7 No 1 11.3% 0.0% 1.1% 4.6% 0.0% 2004 NA 2.5 BVA 3 A2 Mar-13 316 6.7 No 2 10.1% -0.1% 4.0% 7.9% 0.0% 2006 182 1.3 CAJAM 2006-1 A2 Feb-13 335 7.0 Yes 2 30.8% -17.5% 4.7% 18.6% -1.1% 2006 2 1.7 HIPO HIPO-5 A Apr-13 325 6.0 Yes 2 19.9% 0.0% 6.3% 4.0% 0.0% 2002 NA 3.1 HIPO HIPO-7 A2 Apr-13 350 6.3 Yes 2 23.1% 0.0% 12.7% 11.7% 0.0% 2004 590 1.8 HIPO HIPO-8 A2 Mar-13 375 5.6 Yes 3 22.4% -0.5% 14.9% 14.1% -0.5% 2005 43 1.5 RHIPO 8 A2A Jan-13 335 6.0 Yes 2 11.8% -0.1% 2.9% 12.7% 0.0% 2006 151 0.9 SABA 1 A2 Feb-13 310 5.4 Yes 2 5.8% 0.0% 10.3% 7.8% 0.0% 2004 NA 0.6 SHIPO 1 A Apr-13 265 4.5 Yes 1 31.5% 0.0% 2.9% 2.3% 0.0% 2004 1,137 10.7 SHIPO 2 A Apr-13 275 7.0 Yes 1 19.3% -2.3% 4.7% 4.7% -0.3% 2006 8 4.1 TDAC 3 A Mar-13 285 3.5 Yes 1 15.6% 0.0% 12.4% 12.1% 0.0% 2004 NA 1.3 TDAC 4 A Feb-13 316 4.7 Yes 2 12.9% 0.0% 12.9% 13.3% 0.0% 2005 NA 1.0 TDAC 8 A Jan-13 375 6.4 Yes 3 12.6% 0.0% 19.8% 21.4% -0.7% 2007 NA 0.6 TDAC 9 A2 Mar-13 385 6.7 Yes 3 14.7% -4.0% 26.5% 26.0% -1.8% 2007 4 0.6 TDAI 4 A2 Feb-13 368 8.5 Yes 2 10.8% -1.6% 5.0% 16.7% -0.2% 2006 7 0.6 UCI 9 A Mar-13 285 6.9 Yes 1 14.7% 0.0% 4.4% 3.9% 0.0% 2003 NA 3.3 Source: Deutsche Bank, Bloomberg Finance LP.*Bond spread reduces to 223 bps (source: IntexCalc) if senior pro-rata payment trigger is toggled on.Arrears liquidation loss is using loan tape data

We note that there are differences in total arrears as calculated by loan tape and as shown in investor reports, with the latter being on average higher. On average, 90d+ arrears in the sector amounts to 3.8% when looking at loan tapes, which is lower than the 5.7% when sourced from investor reports. Possible reasons for differences include different sample universes, timing differences and potential difficulties in identifying loans in arrears or slippages in loan level data in recording loans in arrears.

In the loan tape, a loan is considered as being in arrears if “account status” is 2 (Arrears) or 3 (Default or Foreclosure).If account status is not available for a loan, it is classified as being in arrears if “arrears balance” is greater than 0, or if “number of months in Arrears” is greater than 0.

Note that the spread levels are purely indicative, and are sensitive to prepayment and default assumptions. A 3 or 4% CPR is assumed to derive the spreads using the Bloomberg cash flow model (CFT page). The above indicated spread levels are also driven by when (and if) pro-rata triggers are toggled which depends on credit performance, therefore spread levels may change materially depending on trigger assumptions.

4 September 2013



Appendix 1

Important Disclosures

Additional information available upon request

For disclosures pertaining to recommendations or estimates made on securities other than the primary subject of this research, please see the most recently published company report or visit our global disclosure look-up page on our website at http://gm.db.com/ger/disclosure/DisclosureDirectory.eqsr

Analyst Certification

The views expressed in this report accurately reflect the personal views of the undersigned lead analyst(s). In addition, the undersigned lead analyst(s) has not and will not receive any compensation for providing a specific recommendation or view in this report. Rachit Prasad/Conor O'Toole

http://gm.db.com/ger/disclosure/DisclosureDirectory.eqsr

4 September 2013



Regulatory Disclosures

1. Important Additional Conflict Disclosures

Aside from within this report, important conflict disclosures can also be found at https://gm.db.com/equities under the "Disclosures Lookup" and "Legal" tabs. Investors are strongly encouraged to review this information before investing.

2. Short-Term Trade Ideas

Deutsche Bank equity research analysts sometimes have shorter-term trade ideas (known as SOLAR ideas) that are consistent or inconsistent with Deutsche Bank's existing longer term ratings. These trade ideas can be found at the SOLAR link at http://gm.db.com.

3. Country-Specific Disclosures

Australia and New Zealand: This research, and any access to it, is intended only for "wholesale clients" within the meaning of the Australian Corporations Act and New Zealand Financial Advisors Act respectively. Brazil: The views expressed above accurately reflect personal views of the authors about the subject company(ies) and its(their) securities, including in relation to Deutsche Bank. The compensation of the equity research analyst(s) is indirectly affected by revenues deriving from the business and financial transactions of Deutsche Bank. In cases where at least one Brazil based analyst (identified by a phone number starting with +55 country code) has taken part in the preparation of this research report, the Brazil based analyst whose name appears first assumes primary responsibility for its content from a Brazilian regulatory perspective and for its compliance with CVM Instruction # 483. EU countries: Disclosures relating to our obligations under MiFiD can be found at http://www.globalmarkets.db.com/riskdisclosures. Japan: Disclosures under the Financial Instruments and Exchange Law: Company name - Deutsche Securities Inc. Registration number - Registered as a financial instruments dealer by the Head of the Kanto Local Finance Bureau (Kinsho) No. 117. Member of associations: JSDA, Type II Financial Instruments Firms Association, The Financial Futures Association of Japan, Japan Investment Advisers Association. This report is not meant to solicit the purchase of specific financial instruments or related services. We may charge commissions and fees for certain categories of investment advice, products and services. Recommended investment strategies, products and services carry the risk of losses to principal and other losses as a result of changes in market and/or economic trends, and/or fluctuations in market value. Before deciding on the purchase of financial products and/or services, customers should carefully read the relevant disclosures, prospectuses and other documentation. "Moody's", "Standard & Poor's", and "Fitch" mentioned in this report are not registered credit rating agencies in Japan unless "Japan" or "Nippon" is specifically designated in the name of the entity. Malaysia: Deutsche Bank AG and/or its affiliate(s) may maintain positions in the securities referred to herein and may from time to time offer those securities for purchase or may have an interest to purchase such securities. Deutsche Bank may engage in transactions in a manner inconsistent with the views discussed herein. Russia: This information, interpretation and opinions submitted herein are not in the context of, and do not constitute, any appraisal or evaluation activity requiring a license in the Russian Federation.

Risks to Fixed Income Positions

Macroeconomic fluctuations often account for most of the risks associated with exposures to instruments that promise to pay fixed or variable interest rates. For an investor that is long fixed rate instruments (thus receiving these cash flows), increases in interest rates naturally lift the discount factors applied to the expected cash flows and thus cause a loss. The longer the maturity of a certain cash flow and the higher the move in the discount factor, the higher will be the loss. Upside surprises in inflation, fiscal funding needs, and FX depreciation rates are among the most common adverse macroeconomic shocks to receivers. But counterparty exposure, issuer creditworthiness, client segmentation, regulation (including changes in assets holding limits for different types of investors), changes in tax policies, currency convertibility (which may constrain currency conversion, repatriation of profits and/or the liquidation of positions), and settlement issues related to local clearing houses are also important risk factors to be considered. The sensitivity of fixed income instruments to macroeconomic shocks may be mitigated by indexing the contracted cash flows to inflation, to FX depreciation, or to specified interest rates - these are common in emerging markets. It is important to note that the index fixings may -- by construction -- lag or mis-measure the actual move in the underlying variables they are intended to track. The choice of the proper fixing (or metric) is particularly important in swaps markets, where floating coupon rates (i.e., coupons indexed to a typically short-dated interest rate reference index) are exchanged for fixed coupons. It is also important to acknowledge that funding in a currency that differs from the currency in which the coupons to be received are denominated carries FX risk. Naturally, options on swaps (swaptions) also bear the risks typical to options in addition to the risks related to rates movements.

http://gm.db.com/

http://www.globalmarkets.db.com/riskdisclosures

David Folkerts-Landau Global Head of Research

Marcel Cassard Global Head

CB&S Research

Ralf Hoffmann & Bernhard Speyer Co-Heads

DB Research

Guy Ashton Chief Operating Officer

Research

Richard Smith Associate Director Equity Research

Asia-Pacific

Fergus Lynch Regional Head

Germany

Andreas Neubauer Regional Head

North America

Steve Pollard Regional Head

International Locations

Deutsche Bank AG

Deutsche Bank Place

Level 16

Corner of Hunter & Phillip Streets

Sydney, NSW 2000

Australia

Tel: (61) 2 8258 1234

Deutsche Bank AG

Große Gallusstraße 10-14

60272 Frankfurt am Main

Germany

Tel: (49) 69 910 00

Deutsche Bank AG

Filiale Hongkong

International Commerce Centre,

1 Austin Road West,Kowloon,

Hong Kong

Tel: (852) 2203 8888

Deutsche Securities Inc.

2-11-1 Nagatacho

Sanno Park Tower

Chiyoda-ku, Tokyo 100-6171

Japan

Tel: (81) 3 5156 6770

Deutsche Bank AG London

1 Great Winchester Street

London EC2N 2EQ

United Kingdom

Tel: (44) 20 7545 8000

Deutsche Bank Securities Inc.

60 Wall Street

New York, NY 10005

United States of America

Tel: (1) 212 250 2500

Global Disclaimer

The information and opinions in this report were prepared by Deutsche Bank AG or one of its affiliates (collectively "Deutsche Bank"). The information herein is believed to be reliable and has been obtained from public sources believed to be reliable. Deutsche Bank makes no representation as to the accuracy or completeness of such information.

Deutsche Bank may engage in securities transactions, on a proprietary basis or otherwise, in a manner inconsistent with the view taken in this research report. In addition, others within Deutsche Bank, including strategists and sales staff, may take a view that is inconsistent with that taken in this research report.

Opinions, estimates and projections in this report constitute the current judgement of the author as of the date of this report. They do not necessarily reflect the opinions of Deutsche Bank and are subject to change without notice. Deutsche Bank has no obligation to update, modify or amend this report or to otherwise notify a recipient thereof in the event that any opinion, forecast or estimate set forth herein, changes or subsequently becomes inaccurate. Prices and availability of financial instruments are subject to change without notice. This report is provided for informational purposes only. It is not an offer or a solicitation of an offer to buy or sell any financial instruments or to participate in any particular trading strategy. Target prices are inherently imprecise and a product of the analyst judgement. As a result of Deutsche Bank’s March 2010 acquisition of BHF-Bank AG, a security may be covered by more than one analyst within the Deutsche Bank group. Each of these analysts may use differing methodologies to value the security; as a result, the recommendations may differ and the price targets and estimates of each may vary widely. The financial instruments discussed in this report may not be suitable for all investors and investors must make their own informed investment decisions. Stock transactions can lead to losses as a result of price fluctuations and other factors. If a financial instrument is denominated in a currency other than an investor's currency, a change in exchange rates may adversely affect the investment. Past performance is not necessarily indicative of future results. Deutsche Bank may with respect to securities covered by this report, sell to or buy from customers on a principal basis, and consider this report in deciding to trade on a proprietary basis.

Derivative transactions involve numerous risks including, among others, market, counterparty default and illiquidity risk. The appropriateness or otherwise of these products for use by investors is dependent on the investors' own circumstances including their tax position, their regulatory environment and the nature of their other assets and liabilities and as such investors should take expert legal and financial advice before entering into any transaction similar to or inspired by the contents of this publication. Trading in options involves risk and is not suitable for all investors. Prior to buying or selling an option investors must review the "Characteristics and Risks of Standardized Options," at http://www.theocc.com/components/docs/riskstoc.pdf . If you are unable to access the website please contact Deutsche Bank AG at +1 (212) 250-7994, for a copy of this important document.

The risk of loss in futures trading, foreign or domestic, can be substantial. As a result of the high degree of leverage obtainable in futures trading, losses may be incurred that are greater than the amount of funds initially deposited.

Unless governing law provides otherwise, all transactions should be executed through the Deutsche Bank entity in the investor's home jurisdiction. In the U.S. this report is approved and/or distributed by Deutsche Bank Securities Inc., a member of the NYSE, the NASD, NFA and SIPC. In Germany this report is approved and/or communicated by Deutsche Bank AG Frankfurt authorized by the BaFin. In the United Kingdom this report is approved and/or communicated by Deutsche Bank AG London, a member of the London Stock Exchange and regulated by the Financial Services Authority for the conduct of investment business in the UK and authorized by the BaFin. This report is distributed in Hong Kong by Deutsche Bank AG, Hong Kong Branch, in Korea by Deutsche Securities Korea Co. This report is distributed in Singapore by Deutsche Bank AG, Singapore Branch or Deutsche Securities Asia Limited, Singapore Branch, and recipients in Singapore of this report are to contact Deutsche Bank AG, Singapore Branch or Deutsche Securities Asia Limited, Singapore Branch in respect of any matters arising from, or in connection with, this report. Where this report is issued or promulgated in Singapore to a person who is not an accredited investor, expert investor or institutional investor (as defined in the applicable Singapore laws and regulations), Deutsche Bank AG, Singapore Branch or Deutsche Securities Asia Limited, Singapore Branch accepts legal responsibility to such person for the contents of this report. In Japan this report is approved and/or distributed by Deutsche Securities Inc. The information contained in this report does not constitute the provision of investment advice. In Australia, retail clients should obtain a copy of a Product Disclosure Statement (PDS) relating to any financial product referred to in this report and consider the PDS before making any decision about whether to acquire the product. Deutsche Bank AG Johannesburg is incorporated in the Federal Republic of Germany (Branch Register Number in South Africa: 1998/003298/10). Additional information relative to securities, other financial products or issuers discussed in this report is available upon request. This report may not be reproduced, distributed or published by any person for any purpose without Deutsche Bank's prior written consent. Please cite source when quoting.

Copyright © 2013 Deutsche Bank AG

Mining loan level data

Documents

Transcript of Mining loan level data