the development of voluntary cash flow statements in germany and ...

182 sbr 52 (2/2000)

Christian Leuz*

THE DEVELOPMENT OF VOLUNTARY CASH FLOW

STATEMENTS IN GERMANY AND THE INFLUENCE OF

INTERNATIONAL REPORTING STANDARDS**

ABSTRACT

This paper studies the incentives of German firms to voluntarily disclose cash flow state-ments. Although cash flow statements have not been mandatory in Germany untilrecently, an increasing number of firms have voluntarily provided cash flow statements.These firms are likely to be influenced by recommendations of the German accountingprofession, IAS 7, and the respective standards of other countries. This paper studies thisinfluence by looking at the adoption pattern of the cash flow statement over time, and itsformat. It uses milestones in the evolution of German professional recommendations andrespective international standards to chart the development of voluntary cash flow state-ments. The paper analyzes the cross-sectional determinants of voluntary (international)cash flow statements using probit regressions and factor analysis. The results support theidea that capital-market forces drive the disclosure of cash flow statements that conformwith international reporting practice.

1 INTRODUCTION

Analyzing a firm’s incentives to voluntarily disclose financial information is amatter of considerable interest in both analytical and empirical accounting re-search. Although many financial disclosures are mandatory, it is important that weunderstand the incentives to provide information in the absence of regulation. Inparticular, the analysis of private disclosure benefits and costs might providevaluable insights to standard setters who are contemplating mandatory disclosures.

This paper studies the incentives of German firms to voluntarily disclose cash flowstatements over time. Cash flow statements are considered an important elementof the annual report and are mandated under many GAAP regimes. However, untilrecently, the disclosure was not compulsory in Germany. Nevertheless, an increas-ing number of German firms did voluntarily provide cash flow statements. These

Schmalenbach Business Review ◆ Vol. 52 ◆ April 2000 ◆ pp. 182–207

* Dr. Christian Leuz, wissenschaftlicher Assistent, Institute of International Accounting, Johann Wolf-gang Goethe-Universität, 60054 Frankfurt am Main, Germany, Fax: +496979823457,Email: [email protected].

** Helpful comments by Prof. Ulrich Rendtel and seminar participants at the London School of Econo-mics and the JW Goethe-Universität/Frankfurt are gratefully acknowledged. I would also like tothank the Schmalenbach Gesellschaft für Betriebswirtschaft for its generous financial support of myEAA congress participation. Data has been kindly provided by Bloomberg, the Deutsche BörseFrankfurt, the Deutsche Vereinigung für Finanzanalyse und Anlageberatung (DVFA) and the Kapi-talmarktdatenbank of the University of Karlsruhe.

Cash Flow Statements

sbr 52 (2/2000) 183

firms were presumably influenced by recommendations of the German auditingprofession, the respective international standards, and the standards of other coun-tries.

The purpose of this paper is to analyze the cross-sectional determinants of volun-tary cash flow statements by large German firms during a period of major changesin disclosure practice. The paper also studies the link between the development ofvoluntary cash flow statement disclosures by German firms, and the evolution ofboth the German professional recommendations and the international reportingstandards. The main hypothesis is that capital-market forces induce firms toprovide cash flow statements, and that international reporting standards andprofessional recommendations influence the form of the statement.

To examine this hypothesis, I look at disclosure practice at three dates. Thesedates represent “milestones” in the evolution of accounting standards for cash flowstatements from the perspective of German firms. In 1992, there were two oppos-ing “standards”, US SFAS 95 and the revised IAS 7 on one side, and the recom-mendation HFA 1978 by the German Institute of Chartered Accountants (IdW) onthe other side. By 1994, the revised IAS 7 was accepted by the International Or-ganization of Securities Commissions (IOSCO) and viewed as equivalent to USSFAS 95 by the Securities and Exchange Commission (SEC). These events lentfurther legitimacy to IAS 7 and made it even more attractive to German firmslooking for a “universal” standard on cash flow statements. By 1996, a Germanfirm preparing a cash flow statement was also able to refer to the newly issuedprofessional recommendation HFA 1995, which is almost identical to the revisedIAS 7.

The paper studies both cross-sectional and time-related aspects. This approachprovides insights into firms’ adoption decisions, as well as documenting the evolu-tion of German accounting practices for the cash flow statement. Recent studieson voluntary disclosures of cash flow statements by German firms are eithermerely descriptive or univariate1. This paper uses multivariate and factor analysisto assess the incremental explanatory power of the variables.

The discussion proceeds as follows: Section 2 describes the institutional setting.Section 3 briefly reviews the findings of previous studies on voluntary disclosuresof cash flow statements in Germany. In Section 4, I characterize firms’ disclosurestrategies. Section 5 describes the selection of the sample, the data, and the adop-tion histories. In Section 6, I derive the hypotheses and analyze the determinantsof voluntary cash flow statements. Section 7 studies the determinants of interna-tional cash flow statements, and Section 8 concludes.

1 See e.g., Haller/Jakoby (1994); Schulte/Müller (1994); Stahn (1996 and 1997); Jakoby et al. (1999).Their findings are reviewed in more detail below. Schneider (1985) provides an extensive study ofvoluntary cash flow information by German firms, but covers disclosure practice only between1972 and 1981.

Chr. Leuz

184 sbr 52 (2/2000)

2 INSTITUTIONAL SETTING

During the time period covered in this study, the fiscal years ending between 1992and 1996, German firms were not required to provide cash flow statements intheir annual reports. The legal accounting rules relevant in this context are §§ 264(2) and 297 (2) HGB. These rules stipulate, among other things, that the annualreport of a corporation must provide a true and fair view of the firm’s financialposition. Based on the legal commentaries, this requirement does not imply a cashflow statement. However, in Germany, the disclosure of a cash flow statement inthe annual report is viewed as sufficient to provide a true and fair view of thefirm’s financial position2.

Before April 1998, the only legal requirement to disclose a cash flow statementcame from § 21 (1) Börsenzulassungsverordnung (BörsZulV). This rule states thatcorporations that register securities for public trading must provide a statement ofsources and uses of funds for the last three years in the firm’s prospectus.However, the rule requires neither a particular format (§ 23 BörsZulV) nor subse-quent disclosure in the annual report.

Filling the gap left by the regulator, the German Institute of Chartered Accountants(IdW) issued a recommendation on cash flow statements (HFA 1978). The purposeof HFA 1978 was to standardize German practice and to provide guidelines forauditing voluntarily disclosed statements. The professional opinion closely followedAPB 19, the prevailing U.S. accounting standard on cash flow statements at thetime3. HFA 1978 suggested a sources-and-uses format and three alternative fundsdefinitions, of which “net working capital” was the recommended definition.

In 1990, as a consequence of the implementation of the fourth and seventh EUdirectives into German accounting law in 1985, HFA 1978 was slightly modified.However, this adaptation involved only minor changes. For example, the recom-mendations for the format of the statement and the funds definition were notaltered, even though they were no longer in line with international practice at thattime.

This discrepancy led the Schmalenbach Gesellschaft für Betriebswirtschaft (SG), aleading organization of German practitioners and academics, to initiate a workinggroup on cash flow statements. Its recommendation, which was published in Fall19934, closely followed the revised IAS 7 and the US SFAS 95. The group suggest-ed that cash flows be defined as changes in cash and cash equivalents only, andthat firms classify them by operating, investment, and financing activities.

An event that might also have been important for German accounting practice wasthe listing of Daimler Benz at the New York Stock Exchange in October 1993. This

2 See Adler/Düring/Schmaltz (1997), Rz. 70–71; Förschle et al. (1999), p. 1379, Rz. 20; Gebhardt(1999a), C620, p. 5.

3 See Serfling/Marx (1991), p. 345.4 First reports of this working group were published in 1990. See Buchmann/Chmielewicz (1990)

and Haller/Jakoby (1994, p. 646).


sbr 52 (2/2000) 185

made Daimler Benz the first German corporation to prepare a cash flow statementaccording to SFAS 95. It also set a prime example for other companies, which mayhave fuelled the development towards a new format for the cash flow statementand a narrow funds definition5. Another important event was the acceptance ofthe revised IAS 7 by IOSCO in October 1993, and its subsequent acceptance bythe SEC. This lent further legitimacy to IAS 7 and made it even more attractive toGerman firms that were looking for an accepted standard on cash flow statements.

Finally, the IdW and the SG released a joint exposure draft late in 1994 and a finaljoint recommendation in 1995. This new professional opinion, HFA 1995, replacesHFA 1978 and closely follows international standards on cash flow statements.Thus, HFA 1995 harmonizes German with international practice6. HFA 1995suggests a classification of cash flows by operating, investment, and financingactivities, and a narrow funds definition based on cash and cash equivalents7.

In April 1998, new legislation was introduced. The Gesetz zur Kontrolle undTransparenz im Unternehmensbereich (KonTraG) required that all exchange-listedcorporations provide cash flow statements in their consolidated financial state-ments for fiscal years beginning after 12/31/1998. In October 1999, the Germanstandard DRS 2, which provides detailed guidelines for cash flow statements, waspassed by the DRSC, the newly founded German standard-setting body8.

3 VOLUNTARY CASH FLOW STATEMENTS PRIOR TO 1992

Previous studies have examined the development of voluntary disclosures of cashflow statements over time. Table 1 provides a summary of their findings forGerman industrial firms with publicly traded stock9.

Table 1 shows that voluntary disclosures of cash flow statements increase overtime. Based on these studies, German disclosure practice prior to 1992, the start-ing point of my study, can be described as follows: The majority of large Germanfirms reported some form of “cash flow statement”. However, only a few firmsprovided cash flow statements with a separate funds change at the bottom line.That is, most firms still disclosed simple statements on changes in assets and liabil-ities, rather than modern cash flow statements. This is surprising, given that APB19 and HFA 1978 asked for the funds change to be separated out. Moreover, manyGerman firms still preferred the “old” sources-and-uses format for their cash flowstatements. In other words, the disclosure practice of many German firms prior to1992 was not in line with international standards and practice.

5 For this conjecture, see Haller/Jakoby (1994) and Jakoby et al. (1999).6 See IdW/SG (1995), p. 210.7 For an explicit comparison, see Mansch et al. (1995) and Jakoby et al. (1999).8 See also v. Wysocki (1999) and Gebhardt (1999b).9 For comparability reasons, table 1 reports results for exchange-listed industrial firms only. Note

that the studies may comprise more firms, in particular financials. For an overview covering alsostudies on non-listed firms see Halle/Jakoby (1994). See also Stahn (1996 and 1997).

Chr. Leuz

186 sbr 52 (2/2000)

4 CHARACTERIZATION OF DISCLOSURE AND ADOPTION STRATEGIES

As international standards and German professional recommendations on the cashflow statement exhibited an increasing degree of harmonization, German firmsmay have experienced increased pressure to conform with international disclosurepractice. Consequently, the percentage of firms voluntarily disclosing a cash flowstatement is likely to increase between 1992 and 1996. More importantly, the state-ments provided by these firms should be more and more in line with internationalpractice10. Although voluntary disclosure could be a continuation of the trenddocumented by previous studies (see Table 1) and hence unrelated to interna-tional and professional standard setting, the form of the statements are likely to beassociated with recent developments described in Section 2.

10 A recent study by Jakoby et al. (1999) confirms this conjecture for the DAX 30 firms.

Table 1: Development of voluntary cash flow statements disclosed by German firms


sbr 52 (2/2000) 187

However, this conjecture cannot be directly tested, because many firms do notindicate in their annual report which standard they followed in preparing the cashflow statement. Therefore, I use the format of the cash flow statement and thefunds definition as distinguishing features and as a measure of the influence ofinternational standards11. For instance, a firm that provides a cash flow statementin 1994, using the “new” operating, investing, and financing format and a narrowfunds definition (cash and cash equivalents) for the first time, is likely to havefollowed international standards. But if the firm uses the “old” sources-and-usesformat and a wide funds definition (as recommended by HFA 1978), the firm willprobably not have followed the international standards. Similarly, firms thatswitched from the old to the new format were probably influenced by the devel-opments described in Section 2.

For this study, I characterize firms according to the three dates and their disclosurestrategies as follows: At the outset, in 1992, five Anglo-American countries and theIASC had completed their standards on the cash flow statement, requiring, amongother things, the operating, investing, and financing format as well as a narrowfunds definition12. In contrast, the German professional recommendation HFA1978 (revised 1990) still suggested the “old” sources-and-uses format and a widefunds definition. Thus, German firms that were preparing a cash flow statementaccording to international standards in 1992 can be characterized as trendsetters.

By 1994, a movement towards international cash flow statements was evident andgaining momentum. The pressure to conform with international practice wasincreasing, largely due to events that took place in Fall 1993 (see Section 2). Thus,I refer to firms that decided to publish a cash flow statement according to interna-tional standards, or switched to such a statement in 1994, as early adopters.

By 1996, German professional recommendations and the international standardswere once again in agreement. The harmonization process for the cash flow state-ment was completed. Firms could rely on a German recommendation. Thus, Iview firms that published an international cash flow statement for the first time in1996 as late followers.

Finally, firms that did not publish a cash flow statement in 1996 were required topublish a cash flow statement at the latest in their fiscal year beginning after12/31/1998.

11 Note that the differences between the recommendation HFA 1978 and the new standards SFAS 95or revised IAS 7 are not just a matter of form. Complying with the latter has implications in termsof the information provided to both capital markets and competitors. The new standards mandatethe disclosure of additional cash flow information that is not provided elsewhere in the annualreport. However, since cash flow statements are voluntary in Germany, it is possible that a firmonly switches the format, without providing additional cash flow information. For this reason I alsouse variables accounting for additional cash flow information provided. See Section 5 for details.

12 See Wallace et al. (1997).

Chr. Leuz

188 sbr 52 (2/2000)

5 DEVELOPMENT OF VOLUNTARY CASH FLOW STATEMENT DISCLOSURES FROM 1992 TO

1996

5.1 SAMPLE SELECTION AND DATA

I choose my sample firms based on the DAX 100 stock index and “Die Großen500” list published by Die Welt in 1996. The Welt 500 list comprises the largest 553German non-financial firms ranked by total revenues13. From the Welt 500 list, Ieliminate all firms that are not listed at a German exchange prior to October 30,1992, because previous studies suggest that the disclosure of cash flow statementsdiffers significantly between listed and non-listed firms14.

I also eliminate from both sources all firms that are either a subsidiary of a foreignfirm, or a subsidiary of a German parent included in the sample because thesubsidiary’s decision to disclose cash flow statements might not be “independent”or voluntary15. For instance, the parent might determine the subsidiary’s disclosurepolicy via group-wide accounting and consolidation procedures16. Or the disclo-sure policy of subsidiaries with a foreign parent might be heavily influenced byforeign disclosure standards and the parent’s disclosure policy. In addition, I elimi-nate two outliers, which I identified as influential observations that distorted someof the regressions17. Thus, my final sample contains 103 non-financial firms.

I create all dependent variables based on the firms’ annual reports. However, dueto the voluntary nature of the disclosure, it is not obvious precisely what consti-tutes a cash flow statement. Thus, I create several variables to check the robust-ness of my results with respect to the classification used.

The first variable, CFS1, is binary. It indicates whether a cash flow statement ofany format is published in the annual report. Because several previous studiesconsider that a separate funds change at the bottom line is the distinguishingfeature of a cash flow statement, my second variable, CFS2, indicates whether thefirm discloses such a cash flow statement. This definition is purely based onformat. Based on the idea of “substance over form”, my third variable, CFS3, clas-sifies cash flow statements according to the fact that a funds change is disclosed(not necessarily at the bottom-line) and that there is at least one line item, which

13 The smallest firm had in 1996 a total revenue of about one billion DM. The list was then double-checked against the Worldscope database. This check revealed five firms with total revenues ofapproximately one billion or more that were missing from Welt 500 list. They were added to thesample.

14 See e.g., Busse von Colbe (1990); Haller/Jakoby (1994).15 The criterion was a stake in the firm’s outstanding capital equal to or greater than 50%. A similar

argument can be made for associated firms, i.e. the case where a sample firm holds a stakebetween 20% and 50% in another sample firm’s capital. However, this case is not present in thefinal sample.

16 See Görges/Schulte (1994) for evidence supporting this conjecture.17 One firm was in severe financial distress during the sample period and the other is a promotional

organization set up by independent retailers.

is typically not provided elsewhere in the annual report and hence help in deter-mining the firm’s cash flows18.

However, this binary classification includes in CFS3 = 0 firms that do not publish acash flow statement and also firms that publish a cash flow statement but that donot meet the additional information criterion. To avoid this clustering and toaccount for qualitative differences in disclosures, I define an ordinal variable withthree levels, indicating a cash flow statement that satisfies the information criterion(CFS4 = 2), any other cash flow statement (CFS4 = 1) and lack thereof (CFS4 = 0).

I also create dummy variables for the form of the cash flow statement provided.The variable FORMAT indicates whether the cash flow statement is based on the“new” operating, investing, financing format (= 1) or the “old” sources-and-usesformat. This variable can be used for cash flow statements of any form (CFS1 = 1)and those that separate out a funds change (CFS2 = 1), respectively, FORMAT1 andFORMAT2. The variable FUNDS indicates whether the funds definition used isnarrow (= 1), i.e., includes only cash and cash equivalents as defined by IAS 7, orwide (= 0). The latter variable requires that cash flow statements separate out afunds change, i.e., CFS2 = 1.

Finally, format and funds dummies can be combined to distinguish between inter-national and more traditional cash flow statements. The ordinal variable CFS5 indi-cates a cash flow statement according to international standards (= 2), a non-inter-national statement with a funds change as separate line item (= 1), and lack of anysuch cash flow statement (= 0)19. The last variable is similar to CFS5, but attemptsto account for additional information content of the cash flow statement by usingCFS3. A classification as “international” requires that CFS3 = 1, FORMAT = 1, andFUNDS = 1. Thus, CFS6 indicates an international statement (= 2), a non-interna-tional statement (= 1) and lack thereof (= 0).

5.2 DESCRIPTIVE STATISTICS AND ADOPTION HISTORIES

Tables 2a, b. and c provide the frequencies of voluntary cash flow statements atthe three dates.

Tables 2 a, b, and c demonstrate that the trend found in previous studies contin-ued in the 90s. For all variables, the number of firms that voluntarily disclose acash flow statement increases over time. Moreover, looking at the history ofchanges for each observation shows that firms move only in one direction. Thus,it is the timing that matters, and which distinguishes firms. Notably, this holds for


sbr 52 (2/2000) 189

18 Examples are gains from selling fixed assets, gains from accounting associates at equity, cash flowsfrom selling fixed assets (as opposed to book values), gross cash flows from new loans and repay-ments, changes in cash and cash equivalents due to currency translation or valuation changes.Note that some of these are non-cash items. However, they help us to deduce the firm’s cash flowin a retrograde fashion.

19 That is, CFS5 = 2 if FORMAT = 1 and FUNDS = 1, CFS5 = 1 if CFS2 = 1 (and either FORMAT orFUNDS equal to 0) and CFS5 = 0 if CFS2 = 0.

Chr. Leuz

190 sbr 52 (2/2000)

Table 2a: Voluntary disclosure of cash flow statements in 1992, 1994 and 1996(binary variables)

CFS1 = 1: Cash flow statement of any kindCFS2 = 1: Cash flow statement with funds change as separate line itemCFS3 = 1: Cash flow statements with funds change that satisfies information criterion (see section 5.1)

Table 2b: Voluntary disclosure of cash flow statements in 1992, 1994 and 1996(binary variables)

FORMAT1 = 1: Statement uses operating, investing, financing format (all cash flow statements)FORMAT2 = 1: Statement uses operating, investing, financing format (statements with funds change

only)FUNDS = 1: Statement uses a narrow funds definition

Table 2c: Voluntary disclosure of cash flow statements in 1992, 1994 and 1996(ordinal variables)

CFS4: Variable indicates a cash flow statement with funds change that satisfies the information crite-rion (=2), a cash flow statement with funds change that fails the criterion (=1) and lack of anysuch statement (=0).

CFS5: Variable indicates a cash flow statement according to international standards (=2), a non-inter-national statement with a funds change (=1) and lack of any such cash flow statement (=0).

CFS6: Variable indicates a cash flow statement according to international standards with additional lineitems (=2), a non-international statement with a funds change (=1) and lack of any such cashflow statement (=0).


sbr 52 (2/2000) 191

all dependent variables. That is, firms neither reverse their decision to disclose acash flow statement nor do they reverse their decision to provide additional infor-mation or to use an international format. Thus, there is a steady trend towards“improved” and “more international” cash flow statements (as defined by mydependent variables).

In 1992, roughly 59% of the sample firms published a cash flow statement of anykind (CFS1). This proportion increased to 90% in 1996. The increase is even three-or eight-fold if we consider the more restrictive definitions for the cash flow state-ment (CFS2 or CFS3).

CFS4 provides insights into qualitative differences. In 1992, the majority of firms(53) did not disclose any line items that are typically not provided elsewhere inthe annual report (CFS4 = 1). Only eight firms provided such items (CFS4 = 2). In1994, the CFS4 = 1 group was still the largest, but the CFS4 = 2 group was alreadysecond. By 1996, the majority of firms (66) published cash flow statements con-taining line items that are expected to provide additional information.

The format variables also show drastic changes. In 1992, more than 90% of thefirms used the “old” sources-and-uses format. By 1996, more than 80% had switch-ed to the “new” operating, investing, and financing format. However, the develop-ment of the funds definition was somewhat different, because 70% of the firmsthat provided a cash flow statement with funds change (CFS2) already used anarrow funds definition. However, note that in 1992, most sample firms did notprovide a cash flow statement with funds change.

The variables CFS5 and CFS6 suggest that in 1992 the majority (23) of firms withCFS2 statements followed HFA 1978, or at least did not apply an internationalstandard. By 1994, about half of the firms still did not provide a cash flow state-ment with funds change (49). Of those that did, only 18 followed HFA 1978, andtwice as many applied an international standard. In 1996, most firms (73)published a cash flow statement in line with international standards (CFS5 = 2),and for the first time these firms represented the largest of all three groups.

Table 3 summarizes the different adoption strategies based on my classification inSection 4. The table exhibits a “classic” adoption pattern. The sample consists offive trendsetters, 31 early adopters, and 37 late followers, based on CFS5. Thedistribution is similar when I use CFS6. However, the comparison of CFS5 andCFS6, shows that there are four firms in 1994 and eight in 1996 whose changestowards an international cash flow statement are more cosmetic in nature, becausethey fail to satisfy the additional-information criterion imposed by CFS6.

Looking at the adoption histories in more detail, we see that the proportion ofearly adopters and late followers is quite different when we compare firmswithout cash flow statements in 1992 to firms with non-international cash flowstatements in 199220. The latter (former) group has a high (low) percentage ofearly adopters and low (high) percentage of late followers. This suggests that thetrend towards international cash flow statement first, and more strongly, influ-

20 Conditional adoption histories are not reported, but available upon request.

Chr. Leuz

192 sbr 52 (2/2000)

enced those firms that were already publishing cash flow statements. Firmswithout cash flow statements in 1992 appear to have been more strongly affectedby the events from 1994 to 1996.

In summary, the evidence suggests that German firms were influenced in theirdisclosure choices by international standards and professional recommendations,but also by the increasing harmonization between standards and recommenda-tions. Moreover, based on the adoption histories, it appears that the new profes-sional recommendation HFA 1995 played an important role for the widespreadadoption of international statements prior to a mandatory disclosure requirement.

Table 4 presents the adoption pattern for international cash flow statements ineach of the six major industry sectors21. The adoption pattern is fairly homoge-neous across industries. All industries except engineering had an “industry leader”in 1992. Each industry exhibits a strong movement towards international cash flowstatements between 1992 and 1996. That is, in 1992, the largest group in eachindustry still comprises firms without any cash flow statement (with fundschange). But by 1996, this group consists of firms with international statements,which have now become the standard in all industries. However, the shiftoccurred slightly earlier in some industries than in others. Although by 1994, inindustries 1 to 3, the largest group comprises firms that provide an internationalcash flow statement, firms in industries 4 to 6, this group still consists of firmswithout a cash flow statement.

6 DETERMINANTS OF VOLUNTARY CASH FLOW STATEMENTS

6.1 HYPOTHESES AND MODEL SPECIFICATION

The following hypotheses on the cross-sectional determinants of voluntary cashflow statements are not based on a particular model, but on a survey of the extant

21 Firms are first classified based on the Composite DAX industry classification and then furtheraggregated into the six major sectors shown in Table 4. Details are available upon request.

Table 3: Adoption strategies


sbr 52 (2/2000) 193

analytical and empirical literature with a focus on prior research pertaining to cashflow statements22:

A firm’s auditor is generally expected to have some influence on the firm’s disclo-sure policy. In particular, the so-called “Big Six” auditors are likely to encourageinternationally accepted accounting and disclosure standards as part of theircompetitive strategy. Thus, I hypothesize that firms with a Big Six auditor aremore likely to publish voluntary, and in particular international, cash flow state-ments. (However, I note that in their 1996 study, Bauer/Schader do not find asignificant association in their analysis of voluntary cash flow statements byAustrian firms.) To test this hypothesis, I use a dummy (BSIX), which is equal toone if the firms has a Big Six auditor in the respective year.

A listing at a foreign exchange presumably increases shareholdings of foreigninvestors and hence the demand for information23. Moreover, I expect Germanfirms to face competitive pressures in the foreign capital market if most domesticfirms provide cash flow statements. Because cash flow statements are common inmany countries, I predict that a listing at a foreign stock exchange is positivelyassociated with voluntary cash flow statements. Several prior studies report inter-national listing effects24. Thus, I use a dummy variable accounting for a firm’sforeign listing status (FORLIST).

22 See also Lang/Lundholm (1993) for this approach.23 See e.g., Meek/Gray (1989). A foreign exchange (e.g., NYSE) may also demand cash flow statements

and other disclosures as part of its listing requirements. However, my sample contains only one suchfirm, Daimler Benz, that has to provide a cash flow statement due to its listing choice. Dropping thisfirm has virtually no effect on the results. In addition, there are seven firms in the sample that haveadopted financial statements according to IAS or US GAAP by 1996. For these firms, cash flow state-ments are also no longer voluntary. Eliminating these firms does not materially affect my results.

24 See e.g., Cooke (1989); Meek et al. (1995).

Table 4: Adoption of international cash flow statements by industry (based on CFS5)

Chr. Leuz

194 sbr 52 (2/2000)

Multinational firms face different demands for disclosures than firms that generatemost of its business domestically. Therefore, the former are more likely to providecash flow statements voluntarily25.

Schneider (1985) reports a significant positive (univariate) association with cashflow statement disclosures. However, Leuz (1999), using a multiple regression,finds that the percentage of sales generated outside Germany does not have asignificant association once industry effects are accounted for. For this reason, thevariable is not included in the model26.

Diamond (1985) shows that firms have an incentive to provide information volun-tarily, thus preempting costly private information acquisition. Since at least someinvestors base their trading on privately acquired information, trading volume is aproxy for information cost savings and, more generally, capital-market benefitsthat are generated by voluntary disclosures27. Thus, I hypothesize that tradingvolume is positively associated with the disclosure of cash flow statements28. Toavoid size-related collinearities with other variables, I use share turnover. Shareturnover is a scaled measure of trading volume, i.e., annual trading volume in allmarket segments at the main exchange divided by the total number of all sharesoutstanding. Due to the skewness of share turnover (LN_TV), I use its naturallogarithm.

Jensen/Meckling (1976) posit that agency costs are likely to increase as the level ofoutside equity rises and the concentration of ownership decreases. Leftwich et al.(1981) argue that firms can reduce these agency costs by providing additionalinformation. Therefore, voluntary disclosures are likely to increase with the firm’sfree float. Moreover, information cost savings generated by voluntary disclosuresdecrease with the number of shareholders and the concentration of ownership29.In addition, large shareholders probably rely less on annual report disclosures,because they often have access to other information channels (e.g., the board ofdirectors). For these reasons, I hypothesize that voluntary cash flow statementsand the dispersion of ownership are positively associated. Many prior studiessupport this hypothesis30. I measure the dispersion of ownership by the free float(FFLOAT), which I define as the percentage of voting shares widely held andknown to be available for free trading.

Many studies hypothesize that voluntary disclosures are positively associated withleverage, because debt-related agency problems are likely to increase with

25 See e.g., Gray/Radebaugh (1984).26 A factor analysis (Section 6.3) shows that the variable clusters with the foreign listing dummy and

has a positive factor loading as one would expected. Thus, collinearities may be responsible forthe insignificance. Note that including the variable in the model has no material effect on the othervariables.

27 See also Scott (1994). Note that cash flow statements may also provide cost savings even if theyprovide no additional information. See Gebhardt (1984).

28 Note, however, that the causal relationship is not obvious. Voluntary disclosures may reduce infor-mation asymmetry and hence increase liquidity-motivated trading volume.

29 See also Diamond (1985).30 See e.g., Schneider (1985); Scott (1994).


sbr 52 (2/2000) 195

leverage31. Again, the idea is that voluntary disclosures enhance monitoring.However, in Germany, public debt agreements are rare. For bank debt agree-ments, there are other means than the annual report to effectively communicateinformation for monitoring purposes. But to equity investors (and other users),cash flow statements provide – at least in principle – valuable information aboutthe firm’s cash-generating ability and financial situation. This information is partic-ularly useful for highly leveraged firms. That is, this information becomes moreimportant as leverage increases. On the other hand, leverage can be inverselyrelated to information cost savings, because a higher leverage implies, ceterisparibus, less outside equity. The latter suggests a negative association. Thus, therelation between leverage and voluntary cash flow statements is not obvious apriori. Not surprisingly, many studies report insignificant results for leverage32. Imeasure leverage (LEVER) as ratio of total liabilities (including provisions) to totalassets.

In principle, more profitable firms are expected to disclose information volunta-rily, thus distinguishing themselves from less profitable firms. But because thisinformation might be useful to competitors, the capital-market benefits could beoutweighed by the proprietary costs, in particular for “moderately” profitablefirms33. Thus, extant models show that the relation between voluntary disclosuresand profitability is complex and depends, for instance, on the type ofcompetition34. Moreover, Harris (1998) and Leuz (1999) argue that for long-rundisclosure policy choices, where the firm commits to the disclosure regardless ofparticular future realizations, the association with profitability is likely to be nega-tive if the proprietary costs of the disclosure are substantial. But cash flow state-ments might not reveal much proprietary information, since ample informationabout the firm’s profitability is already publicly available. Previous studies producemixed results35. They generally find a negative association, although it is often notsignificant. Thus, I follow the extant literature and predict a negative association,but note that it may be weak for voluntary cash flow statements. I use the salesmargin (PROF) as proxy, which I define as the ratio of operating income to totalrevenues.

The firm’s capital intensity is a proxy for its financing needs36. Prior researchshows that voluntary disclosures and security offerings are positively associated37.Thus, I hypothesize that, ceteris paribus, capital-intensive firms are more likely toprovide voluntary cash flow statements. I measure capital intensity (LTA) as theratio of long-term assets over total assets.

Finally, there are many reasons why firm size might be positively associated withvoluntary disclosures. First, the costs of producing and disseminating information

31 See e.g., Leftwich (1981).32 See e.g., Schneider (1985); Wagenhofer (1990b).33 See e.g., Verrecchia (1983); Wagenhofer (1990a).34 See e.g., Verrecchia (1990); Ewert/Wagenhofer (1992); Feltham et al. (1992).35 See e.g., Schneider (1985); Wagenhofer (1990b); Harris (1998).36 Capital intensity may also be a proxy for entry barriers. But as cash flow statements are not very

proprietary in nature, this interpretation of the proxy is likely to be less relevant in this context.37 See e.g., Lang/Lundholm (1993 and 1997).

Chr. Leuz

196 sbr 52 (2/2000)

are likely to be decreasing per unit of firm size. Second, the larger the firm, themore investors and analysts are likely to be interested in the firm. Hence, firm sizeis also a proxy for potential cost savings in private information acquisition. Third, Iexpect that larger firms will have more foreign investors, and therefore will beunder more pressure to conform with international accounting practice. A positivesize effect is well-documented in the literature38. Thus, I hypothesize that largerfirms are more likely to provide voluntary cash flow statements. Followingprevious studies, I measure firm size (LN_TA) as the natural logarithm of totalassets.

I have obtained proxies from publicly or commercially available sources39. Pair-wise correlations (not reported) show that several of the variables are significantlycorrelated. For instance, the variables related to capital-market aspects (i.e., freefloat, foreign listing, and trading volume) are correlated. Furthermore, size iscorrelated with Big Six auditor, free float, and especially with foreign listing.Because pairwise correlations can be misleading for dichotomous variables, I alsocompute the variance inflation. I find that it is well below two for all variablesexcept firm size and foreign listing. In addition, I perform collinearity diagnosticsaccording to Belsley et al. (1980). Although the tests fail the suggested criteria indi-cating severe collinearity40, they do confirm a potential problem between firm sizeand foreign listing, particularly in 1992 and 1994.

Therefore, I drop firm size from the model. Direct and indirect costs associatedwith the production and dissemination of cash flow statements are unlikely to begreat enough that size-related economies of scale in information production reallymatter. Thus, size is primarily a proxy for cost savings in private informationacquisition. However, these cost savings are already, and more precisely, capturedby other variables, notably free float and share turnover41.

In summary, I estimate a multiple regression for each t using the following modelto examine the incremental explanatory power of the variables:

Disclosure Variableit = Interceptit + β1 BSIXit + β2 FORLSTit + β3 LN_TVit+ β4 FFLOATit + β5 LEVERit +β6 PROFit + β7 LTAit + εit

for i = 1, .., 103 and t = {1992, 1994, 1996}.

6.2 RESULTS

Table 5 presents binary probit regressions for cash flow statements, showing fundschanges as separate line item (CFS2) and all three years. The results for CFS1 and

38 See e.g., Schneider (1985); Lang/Lundholm (1993); Bauer/Schader (1996).39 More details, descriptive statistics or the data itself are available from the author upon request.40 That is, I do not find condition indices above 10 and two or more variables with variance propor-

tions above 0.5 for the same eigenvalue in any year. The condition numbers, i.e., the highestcondition index, is below 4 in the intercept-adjusted diagnostics.

41 Including firm size generally renders the foreign listing dummy insignificant in 1992 and 1994regressions, but does not alter any of my conclusions.


sbr 52 (2/2000) 197

CFS3 are not reported, but are similar to those in Table 5. The predicted signsappear in the first column. Each subsequent column provides estimated coeffi-cients and z-statistics (in parentheses). I estimate the regressions with quasi-maximum likelihood procedures such that the standard errors are robust tomisspecifications of the underlying distribution, as suggested by White (1982).

The model has significant explanatory power in all years. That is, the likelihoodratio (LR) statistic, which tests for the explanatory power of the full model, is highlysignificant and the within-sample classification rate is superior to the naive classifi-

Table 5: Probit regressions for cash flow statements with separate funds change

* significant with p < 0.10 (two-sided), ** significant with p < 0.05, *** significant with p < 0.01

cation (based on the largest cell). All coefficients for which I have specific hypoth-eses have the predicted signs. Exceptions are capital intensity in 1992, free float in1994, and Big Six auditor in the last two regressions, but none of these signs is sig-

Chr. Leuz

198 sbr 52 (2/2000)

nificant. In 1992, only foreign listing reaches conventional significance levels. In1994, foreign listing and trading volume are significant. In addition to these varia-bles, free float and capital intensity turn out to be significant in 1996. The high sig-nifcance level of foreign listing in this year obtains because all firms with a foreignlisting disclose cash flow statements, which makes the variable a perfect predictorin 25 cases and leaves 78 observations to be explained by the other variables42.

Table 6 presents ordered probit regressions based on CFS4, which accounts forqualitative differences across cash flow statements. Again, the predicted signs arepresented in the first column. Each subsequent column provides estimated coeffi-cients and z-statistics (in parentheses). I estimate the regressions with quasi-maximum likelihood procedures such that the standard errors are robust tomisspecifications of the underlying distribution as suggested by White (1982).

The model has significant explanatory power in all years, as indicated by the like-lihood ratio statistic. All coefficients for which I have specific hypotheses have thepredicted signs. Exceptions are profitability in 1992 and free float in the first tworegressions. However, all unexpected signs have very low p-values. Again, onlyforeign listing is significant in 1992. However, the Big Six dummy is close toconventional significance levels (p = 0.140). In 1994, foreign listing and tradingvolume are both highly significant. Capital intensity almost attains the 10% signifi-cance level. In 1996, capital intensity, free float, and leverage are significant, as areforeign listing and trading volume, which now have lower z-statistics. Interest-ingly, leverage is only significant for cash flow statements that are likely to revealadditional information, i.e., when I use CFS3 and CFS4 as the dependent variables.This finding is consistent with theory, in that cash flow information is particularlyimportant for highly levered firms43.

Overall, binary and ordered regressions produce similar results. The findings areconsistent with the notion that capital-market benefits (or pressures) drive volun-tary cash flow statements. In 1992, only firms that faced pressures in foreigncapital markets provided cash flow statements. Later, firms with substantial domes-tic capital-market benefits (as measured by the proxies) followed.

6.3 FACTOR ANALYSIS

To gain further insights about the driving forces of voluntary cash flow statements,I perform a common factor analysis. Since all the variables stand for related, butunobservable (capital-market) factors, the question is whether the variables clusterin a way that is consistent with theory. That is, the identified factor pattern mightfacilitate the interpretation of the previous results. Moreover, the factor patternallows the construction of factor scores, which then can be used in the regressionsinstead of the individual variables.

42 Re-estimating the regression for 1996 without foreign listing shows that the estimates of the othercoefficients are reliable.

43 Recall that both CFS3 and CFS4 require that the cash flow statement has additional line items thatare generally not contained elsewhere in the annual report. See also Leuz (1999) for a similarfinding.


sbr 52 (2/2000) 199

Tables 7a, b, and c present the factor patterns identified by principal factoranalysis using adjusted squared multiple correlations (SMC) as prior communalityestimates44. This is a form of factor analysis in which factors are retained based onthe variation proportion criterion45. Due to the subjectivity involved, I refrain fromany rotation methods46. For clarity, all factor loadings smaller than 0.5 are omittedin Tables 7a, b, and c.

The factor pattern is similar across the years. That is, there are always the samethree factors with the same ordering. In all cases, the three factors account foralmost 100% of the variation. Based on the clustering of the variables, the factors

44 I acknowledge that factor analysis can be problematic with dichotomous variables. However, theanalysis is primarily used to identify variable clusters to facilitate the interpretation of previousresults and hence in a more heuristic sense. Kim/Mueller (1978, p. 75) suggest that factor analysiswith dummy variables can be justified in this case.

45 Principal component analysis, which sets all prior communalities to 1 and retains components withan eigenvalue smaller than one, yields a very similar factor pattern.

46 Note, however, that both the orthogonal and the oblique transformation yield essentially the samefactor clustering and interpretation.

Table 6: Ordered Probit regressions for voluntary cash flow statements


Chr. Leuz

200 sbr 52 (2/2000)

can be interpreted as follows: The first factor, which combines foreign listing, freefloat, and the share turnover variables, is related to the firm’s capital-market orien-tation. That is, firms with a foreign listing, widely dispersed ownership, and hightrading volume are likely to be more capital-market oriented and more likely tobenefit from the disclosure of cash flow statements.

The second factor is related to the firm’s financial position, combining financialleverage and profitability. Note that the factor loadings have opposite signs, pre-cisely as theory suggests. That is, high leverage and low profitability are generallyan indication of a weak financial position. Firms with a weak financial position are

Table 7a: Factor pattern in 1992

Table 7b: Factor pattern in 1994

Table 7c: Factor pattern in 1996


sbr 52 (2/2000) 201

likely to be under more pressure to produce cash flow statements, which facilitatethe assessment of the firm’s cash-generating ability.

The last factor is related to the firm’s financing needs. The primary factor loadingis the capital-intensity variable. Firms with a higher proportion of long-term assetsoften have higher long-term financing needs. Theory predicts that firms withlarger financing needs are more forthcoming in their disclosure policy47.

Clearly, the identified factor pattern is consistent with disclosure theory. Further,ordered regressions of CFS4 on factor scores constructed from the factor patternproduce results (not reported) that are consistent with the findings in Section 6.2. Ifind that the factors have the expected positive signs. However, in 1992, only thefirst factor (capital-market orientation) is highly significant. In 1994 and 1996, theother two factors (financial position and financing needs) become highly signifi-cant as well. This pattern, and my finding that the factors’ marginal effects andtheir significance levels increase over time, suggest an increasing importance ofcapital-market related aspects for voluntary disclosures by German firms.

7 DETERMINANTS OF INTERNATIONAL CASH FLOW STATEMENTS

7.1 HYPOTHESES AND UNIVARIATE RESULTS

An important hypothesis of this paper is that German firms’ voluntary cash flowstatements are influenced by both the developments of international reportingstandards and the German professional recommendations. However, not everyfirm is influenced by these developments in the same way. That is, the impact islikely to depend on firm characteristics, which are proxies for the benefits asso-ciated with an early adoption of international cash flow statements. Therefore, inthis section I analyze these cross-sectional differences.

Most of my sample firms adopt an international cash flow statement during theperiod considered. As noted in Section 5.2, it is the timing of the adoption thatdistinguishes firms. However, as international cash flow statements are generallyviewed as more informative48, I expect similar associations between the timing ofthe adoption and the independent variables, as hypothesized in Section 6.1. Thatis, I hypothesize that firms that adopt international cash flow statements earlier aremore likely to have a Big Six auditor, a foreign listing, a higher percentage of salesgenerated abroad, higher trading volume, free float, and capital intensity. Again,the association with leverage is not obvious. Moreover, the timing of voluntarycash flow statements is more likely to be positively associated with profitability.That is, firms are more likely to make the decision when they have “good news”.

47 See also Lang/Lundholm (1997).48 See e.g., Haller/Jakoby (1994); Mansch et al. (1995); Gebhardt (1999). This view is also supported

by the high correlation between cash flow statements with at least one additional line item gener-ally not provided elsewhere in the annual report (CFS3 = 1) and cash flow statements that conformwith the international format and funds definition (CFS5 = 2). As noted in section 5.2, there areonly few “international” cash flow statements that fail the additional line item criterion.

Chr. Leuz

202 sbr 52 (2/2000)

I analyze these hypotheses based on the classification of disclosure strategiesintroduced in Section 4. I use the variable CLASS = {0, 1, 2, 3} to distinguishbetween trendsetters, early adopters, late followers and waiting firms. In thisvariable, three indicates a trendsetter, two an early adopter, one a late follower,and zero a waiting firm (see also Table 3).

Table 8 presents the univariate results for 1996. The predicted signs are presentedin the first column. The (two-sided) p-values in parentheses stem from a nonpara-metric Kruskal-Wallis test, and indicate whether the medians of the four groupsare statistically different49.

Both the medians and means of the four groups exhibit an ordering that is consist-ent with the hypothesized sign. Although the ordering is not always perfect, theexpected trend prevails in all cases. Moreover, the differences in the medians arestatistically significant for share turnover, free float, capital intensity, and firm size.When I use a classification based on CFS5, the results are similar. Thus, the univar-iate results support my hypotheses, but, as always, must be interpreted cautiously.

7.2 MULTIVARIATE RESULTS

In this section, I examine the incremental explanatory power of the variables forthe decision to adopt an international cash flow statement. Table 9 presentsordered probit regressions based on the model specified in Section 6.1, usingCFS6 as dependent variable. As before, the predicted signs appear in the firstcolumn. Each subsequent column provides estimated coefficients and z-statistics(in parentheses). I estimate the regressions with quasi-maximum likelihood proce-dures such that the standard errors are robust to misspecifications of the under-lying distribution as suggested by White (1982).

The likelihood ratio statistic indicates that the model has significant explanatorypower in every year. All coefficients for which I have specific hypotheses have thepredicted signs, except for free float in 1994 and Big Six auditor in the last tworegressions. But again, all these unexpected signs have very low p-values.

Overall, the results are very similar to those reported in Table 6. In 1992, onlyforeign listing is significant. In 1994, foreign listing and trading volume are theonly significant variables, and in 1996, trading volume, free float, and capitalintensity exhibit significant positive associations.

As noted in Section 5.2, firms do not switch back to the old form once they haveadopted an international cash flow statement. However, this fact is not accountedfor in the three regressions in Table 9. To address this issue, I re-estimate theabove regressions, using only those firms that have not adopted an internationalcash flow statement at the previous point in time. Firms that provide such a state-

49 I have created similar classification variables in 1992 and 1994: CLASS92 = {0, 1}, where 1 indicatesa trendsetter; CLASS94 = {0, 1, 2}, where 2 indicates a trendsetter and 1 an early adopter. Usingthese variables and the independent variables in the respective years yields analogous results.


sbr 52 (2/2000) 203

Table 8: Univariate results for the different disclosure strategies (CLASS)

Chr. Leuz

204 sbr 52 (2/2000)

ment for the first time are coded as one; all others as zero. These binary probitregressions (not reported) yield results similar to those reported in Table 9, albeitat lower LR statistics due to the smaller sample size.

8 SUMMARY AND CONCLUSIONS

This paper studies the determinants of voluntary and international cash flow state-ments for a sample of 103 large German firms at three different points in time. Itcovers a period of major changes in the disclosure of cash flow statements byGerman firms: In 1992, the majority of sample firms did not provide cash flow (oreven simple funds flow) statements. Only four years later, this situation hadcompletely changed. In 1996, most sample firms disclosed cash flow statements inline with international reporting practice. This strong trend towards internationalcash flow statements seems to have been influenced by the international standardsfor cash flow statements as well as the German professional recommendation HFA1995.

Table 9: Ordered probit regressions for international cash flow statements (CFS6)



sbr 52 (2/2000) 205

Looking at the adoption history in detail shows that the trend towards interna-tional cash flow statements first, and more strongly, influenced those firms thatwere already publishing cash flow statements. Firms without cash flow statementappear to have been more strongly influenced by the events from 1994 to 1996,e.g., the professional recommendation HFA 1995. Overall, voluntary disclosures ofinternational cash flow statements follow a classic adoption pattern over time: Inthe beginning, there are only a few trendsetters, then the number of firms steadilyincreases. This adoption pattern is fairly homogeneous across industries, althoughthere are some industries where firms have switched slightly earlier.

Across all regressions, the cross-sectional determinants seem to be stable, althoughtheir relative importance changes over time. In 1992, voluntary cash flow state-ments are provided primarily by firms that are listed on foreign exchanges. In1994, foreign listing and trading volume have significant positive associations withvoluntary cash flow statements. In 1996, the explanatory power of the listingstatus decreases and trading volume, free float, and capital intensity exhibit signifi-cant positive associations. These results are in line with the idea that capitalmarket pressures drive voluntary disclosures of cash flow statements, and thatfirms with the largest capital-market pressures (or benefits) adopt these disclosuresfirst. In particular, firms with foreign listings move first, presumably because theyface the greatest pressures, which stem from foreign capital markets. Later, firmswith high trading volume, large free float, and high capital intensity follow. Theyare likely to have relatively large benefits in the domestic capital market.

A subsequent factor analysis confirms the importance of capital-market relatedaspects. The identified factor pattern is stable across the three points in time andconsists of three factors: the firm’s capital-market orientation, its financial position,and its financing needs. Subsequent regressions using factor scores produce corro-borating results.

Finally, the regressions suggest similar determinants for the adoption of interna-tional cash flow statements. I find that firms with foreign listing, higher tradingvolume, free float, and capital intensity are more likely to adopt internationalstatements early. Again, these findings support the idea that capital-market bene-fits drive voluntary disclosures that are in line with international practice, even ifthere is no regulation.

REFERENCES

Adler, Hans/Düring, Walther/Schmaltz, Kurt (1997), Rechnungslegung und Prüfung der Unternehmen,Kommentar, 6th edition, Teilband 2.

Bauer, Anna/Schader, Heimo (1996), Darstellung der Finanzlage im Jahresabschluß – Eine empirischeUntersuchung, in: Journal für Betriebswirtschaft, Vol. 46, pp. 89–99.

Belsley, David A./Kuh, Edwin/Welsch, Roy E. (1980), Regression Diagnostics, Identifying Influential Dataand Sources of Collinearity.

Berndsen, Hans-Peter (1979), Unternehmenspublizität.Buchmann, Ruth/Chmielewicz, Klaus (1990), Finanzierungsrechnung, Empfehlungen des Arbeitskreises

“Finanzierungsrechnung” der Schmalenbach Gesellschaft, in: zfbf, Sonderheft 26.Busse von Colbe, Walther (1981), Kapital- und Finanzflußrechnung, in: Kosiol, Erich (ed.), Handwörter-

buch des Rechnungswesens, 2nd edition, Sp. 891–901.

Chr. Leuz

206 sbr 52 (2/2000)

Busse von Colbe, Walther (1990), Funds Flow Statement, in: Grochla,Erwin et al. (eds), Handbook ofGerman Business Management, Volume 1, Sp. 1000–1016.

Chmielewicz, Klaus/Caspari, Britta (1985), Zur Problematik von Finanzierungsrechnungen, in: DieBetriebswirtschaft, Vol. 45, pp. 156–169.

Coenenberg, Adolf G./Schmidt, Franz (1978), Die Kapitalflußrechnung als Ergänzungsrechnung desveröffentlichten Jahresabschlusses, in: Zeitschrift für Betriebswirtschaft, Vol. 48, pp. 507–516.

Cooke, Terence E. (1989), Disclosure in the corporate annual reports of Swedish companies, in:Accounting and Business Research, Vol. 19, pp. 113–124.

Diamond, Douglas W. (1985), Optimal release of information by firms, in: Journal of Finance, Vol. 40,pp. 1071–1094.

Ewert, Ralf/Wagenhofer, Alfred (1992), Unternehmenspublizität und Konkurrenzwirkungen, in: Zeit-schrift für Betriebswirtschaft, Vol. 62, pp. 297–324.

Feltham, Gerald A./Gigler, Frank B./Hughes, John S. (1992), The effects of line-of-business reporting oncompetition in oligopoly settings, in: Contemporary Accounting Research, Vol. 9 (Fall), pp. 1–23.

Förschle, Gerhart/Lust, Peter/Kroner, Matthias (1999), § 297, Beck’scher Bilanz-Kommentar, Der Jahres-abschluß nach Handels- und Steuerrecht, bearbeitet von Dieter Budde et al., 4. Auflage.

Gebhardt, Günther (1984), Kapitalflußrechnungen als Mittel zur Darstellung der Finanzlage, in: DieWirtschaftsprüfung, Vol. 37, pp. 481–491.

Gebhardt, Günther (1999), Empfehlungen zur Gestaltung informativer Kapitalflußrechnungen nachinternationalen Grundsätzen, in: Betriebs-Berater, Vol. 54, pp. 1314–1321.

Gebhardt, Günther (1999), Kapitalflußrechnungen, in: Castan, Edgar et al. (eds), Beck’sches Handbuchder Rechnungslegung, C 620, pp. 1–89.

Goebel, Andrea/Fuchs, Markus (1995), Die Anwendung der International Accounting Standards in denKonzernabschlüssen deutscher Kapitalgesellschaften, in: Der Betrieb, Vol. 48, pp. 1521–1527.

Görges, Thomas/Schulte, Jörn. (1994), Publizitätspolitik börsennotierter Teilkonzerne – Eine empirischeUntersuchung, in: Die Wirtschaftsprüfung, Vol. 47, pp. 561–574.

Gray, Sidney J./Radebaugh, Lee H. (1984), International Segment Disclosures by US and UK MultinationalEnterprises: A Descriptive Study, in: Journal of Accounting Research, Vol. 22 (Spring), pp. 351–360.

Haller, Axel/Jakoby, Stephan (1994), Verbreitung und Entwicklungsstand der Finanzierungsrechnung inDeutschland: Eine empirische Analyse, in: Der Betrieb, Vol. 47, pp. 641–649.

Harris, Mary S. (1998), The Association between Competition and Managers’ Business Segment Report-ing Decisions, in: Journal of Accounting Research, Vol. 36, Spring, pp. 111–128.

IdW (1978), Stellungnahme HFA 1/1978: Die Kapitalflußrechnung als Ergänzung des Jahresabschlusses,in: Die Wirtschaftsprüfung, Vol. 31, pp. 207–208.

IdW/SG (1995), Die Kapitalflußrechnung als Ergänzung des Jahres- und Konzernabschlusses, Gemein-same Stellungnahme HFA 1/1995 des Hauptfachausschusses und des Arbeitskreises “Finanzierungs-rechnung” der Schmalenbach-Gesellschaft/Deutsche Gesellschaft für Betriebswirtschaft e.V., in: DieWirtschaftsprüfung, Vol. 48, pp. 210–213.

Jakoby, Stephan/Maier, Jürgen/Schmechel, Thomas (1999), Internationalisierung der Publizitätspraxis beiKapitalflußrechnungen – Eine empirische Untersuchung der DAX-Unternehmen für den Zeitraum1988 bis 1997, in: Die Wirtschaftsprüfung, Vol. 52, pp. 225–238.

Jensen, Michael C./Meckling, William H. (1976), Theory of the Firm, Managerial Behavior, AgencyCosts, and Ownership Structure, in: Journal of Financial Economics, Vol. 3, pp. 305–360.

Kim, Jae-On/Mueller, Charles W. (1978), Factor Analysis, Statistical Methods and Practical Issues, SageUniversity Papers, No. 14.

Lang, Mark/Lundholm, Russell (1993), Cross-Sectional Determinants of Analyst Ratings of CorporateDisclosures, Journal of Accounting Research, Vol. 31, No. 2, pp. 246–271.

Lang, Mark/Lundholm, Russell (1997), Voluntary Disclosure During Equity Offerings: Reducing Informa-tion Asymmetry or Hyping the Stock?, working paper, University of Michigan.

Leftwich , Richard W./Watts, Ross L./Zimmerman, Jerold L. (1981), Voluntary corporate disclosure: Thecase of interim reporting, in: Journal of Accounting Research, Vol. 19, Supplement, pp. 50–77.

Leuz, Christian (1999), Proprietary versus Non-Proprietary Disclosures: Evidence on Voluntary CashFlow Statements and Segment Reports in Germany, Finance & Accounting Working Paper Series, No.21, JW-Goethe Universität Frankfurt.

Mansch, Helmut/Stolberg, Klaus/von Wysocki, Klaus (1995), Die Kapitalflußrechnung als Ergänzung desJahres- und Konzernabschlusses, in: Die Wirtschaftsprüfung, Vol. 48, pp. 185–203.

Meek, Gary K./Gray, Sidney J. (1989), Globalization of stock markets and foreign listing requirements:voluntary disclosures by continental European companies listed on the London Stock Exchange, in:Journal of International Business Studies, Vol. 20 (2), pp. 315–336.

Meek, Gary K./Roberts, Clare B./Gray, Sidney J. (1995), Factors influencing voluntary annual reportdisclosures by US, UK and Continental European multinational corporations, in: Journal of Interna-tional Business Studies, Vol. 26 (3), pp. 555–572.

Schmidt, Franz (1981), Die Beurteilung der Liquidität mit Hilfe von Kapitalflußrechnungen, in: DieFortbildung, Vol. 26, September, pp. 70–77.

Schneider, Pia (1985), Finanzwirtschaftliche Nebenrechnungen.Schülen, Werner (1992), Angaben zur Gewinn- und Verlustrechnung, in: Castan, Edgar et al. (eds),

Beck’sches Handbuch der Rechnungslegung, Band I, B 425.Schulte, Karl-Werner/Müller, Stefan (1994), Berichterstattung über die Finanzlage in den Geschäftsbe-

richten deutscher Aktiengesellschaften – eine empirische Untersuchung, in: Die Aktiengesellschaft,Vol. 39, pp. 540–559.

Scott, Thomas W. (1994), Incentives and disincentives for financial disclosure: Voluntary disclosure of def-ined benefit pension plan information by Canadian firms, in: Accounting Review, Vol. 69, pp. 26–43.

Serfling, Klaus/Marx, Marita (1991), Die Bedeutung der Richtlinie SFAS No. 95 für die Praxis der Kapi-talflußrechnung, in: Die Wirtschaftsprüfung, Vol. 44, pp. 345–350.

Stahn, Frank (1996), Zum praktischen Entwicklungsstand der Konzern-Kapitalflußrechnung in Deutsch-land, in: Die Wirtschaftsprüfung, Vol. 49, pp. 649–657.

Stahn, Frank (1997), Die Kapitalflußrechnung in der aktuellen Berichterstattung deutscher Konzerne,in: Betriebs-Berater, pp. 1991–1996.

Verrecchia, Robert E. (1983), Discretionary disclosure, in: Journal of Accounting and Economics, Vol. 5,pp. 179–194.

Verrecchia, Robert E. (1990), Information Quality and Discretionary Disclosure, in: Journal of Account-ing and Economics, Vol. 12, pp. 365–380.

Wagenhofer, Alfred (1990a), Voluntary disclosure with a strategic opponent, in: Journal of Accountingand Economics, Vol. 12, pp. 341–363.

Wagenhofer, Alfred (1990b), Informationspolitik im Jahresabschluß.Wallace, Olusegun R.S./Choudhury, Mohammed S.I./Pendlebury, Maurice (1997), Cash Flow State-

ments: An International Comparison of Regulatory Positions, in: International Journal of Accounting,Vol. 32, No. 1, pp. 1–22.

White, Halbert (1982), Maximum Likelihood Estimation of Misspecified Models, in: Econometrica, Vol.50, January, pp. 1–25.

Wysocki, Klaus v. (1999), DRS 2: Neue Regeln des Deutschen Rechnungslegungs Standards Committeezur Aufstellung von Kapitalflußrechnungen, in: Der Betrieb, Vol. 52, pp. 2373–2378.


sbr 52 (2/2000) 207

the development of voluntary cash flow statements in germany and ...

Documents

Transcript of the development of voluntary cash flow statements in germany and ...