The Emergence of Shareholder Value In the German Corporation

Michael Bradley and Anant Sundaram*

October 2003

* F.M. Kirby Professor of Investment Banking and Professor of Law, Duke University (bradley@duke.edu), andAssociate Professor of Finance, Thunderbird, the American Graduate School of International Management(sundaram@thunderbird.edu). Sundaram acknowledges support from the Thunderbird Research Center. We thankMohan Gopalan, Tida Srichiraratna and especially Pia Mussel for excellent research assistance. We are grateful toHendrik Bessembinder for comments.

2

Abstract

We analyze the effects of changes in the purpose of large German corporations from stakeholder-orientedorganizations to shareholder-oriented organizations during the decade of the 1990s. We document thistransformation by first examining the annual reports of large firms at strategic points in time relative to significantchanges in German corporate law. We find that changes in the law over this period both reflected and facilitated afundamental shift in the operations of German corporations as evidenced by their adoption of stock- and option-based incentive compensation plans, adoption of US GAAP-based (or related) accounting systems, ADR listings,and restructuring activity. We also document the emergence and adoption of the rhetoric of shareholder value amongGerman managers, the public, and the media. Detailed empirical analysis shows that German firms that embracedshareholder value as their corporate purpose and operating strategy realized a gain in equity values over the decadeof the 1990s, as well as a significant increase in their Betas relative to the S&P 500, when compared to lessshareholder-oriented firms. We interpret the Beta shifts as evidence that focusing on shareholder value leads firms toadopt entrepreneurial risk-taking strategies that reflect shareholder, rather than stakeholder, concerns. We conjecturethat the increase in Betas might also be due to the adoption by some German firms of a similar operating philosophyto that of the traditionally shareholder-oriented US corporation. Finally, we show that German firms that embracedshareholder value-orientation during the 1990s realized significantly greater growth in their Market-to-Book ratiosand market capitalizations relative to their less shareholder-oriented counterparts.

3

1. Introduction

For the past two decades scholars have been debating the nature of the evolution of

corporate governance structures and practices throughout the world. Much of the debate centers

on the degree to which corporate governance structures are converging toward an optimal form.

Two camps are well represented in the literature. There are those who believe that global

competition in capital markets, product markets, and the market for corporate control will result

in an optimal governance structure to which all economies will inevitably be driven. The general

belief among these scholars is that this optimal form resembles the Anglo-American corporation

to varying degrees. Alternatively, there are those who believe that path dependencies, culture,

and historical serendipity will prevent convergence to any one optimal form.

The early literature denied the existence of an optimal corporate form and cautioned

against drawing any generalizations to pronounce one particular form superior to the alternatives.

It was argued that the different governance structures found in the leading economies throughout

the world, such as the US, Germany, Japan and the UK were ‘efficient’ forms in their own

respective ways, since these corporate economies could not have become successful and endured

for so long had there been major deficiencies inherent in their governance structures.1 The

literature subsequently split into two camps: the ‘path dependency camp’ which holds that path

dependencies, cultures, and historical serendipity prevent the convergence to any one optimum,2

and the ‘optimal governance camp’ which holds that global competition will manifest an optimal

governance structure to which all economies will inevitably be driven.3

The path dependency camp argues that the persistence of alternative corporate forms

results from economic structures being dependent on those that preceded it. Factors contributing

to this dependence are ‘structure-driven’ (e.g., informal rules and ownership structures) and

‘rule-driven’ (e.g., laws). Efficiency considerations, interest group politics, and rent-seeking

behavior militate against revolutionary change. In contrast, the ‘optimal governance camp’

points to a unidirectional trend toward convergence. While governance systems in the European

1 See, for example, Romano (1993); Prowse (1994); Kester (1997); Shleifer and Vishny (1997); Bratton andMcCahery (1999).

2 See, for example, Roe (1994); Roe (1996); Bebchuk and Roe (1999). Others have expanded on arguments for path-dependency to consider the roles of national culture and emebeddedness of social norms and values (Licht, 2001).

3 See Bradley, Schipani, Sundaram, and Walsh (1999); Hansmann and Kraakman (2000).

4

Union (EU) and in Japan, traditionally viewed as communitarian or stakeholder systems of

governance, were adopting features of the Anglo-American model - characterized by a system

that focuses on shareholder value - there was no evidence of the reverse, implying that the so-

called contractarian model was optimal.4 Further, this move toward the contractarian model

seemed to trump path-dependence, since the structures and rules that lead to persistence are

themselves endogenous in the long-run.5

This debate is certainly interesting and has provided rich cross-country, comparative

analyses and data. But the literature may be missing a larger issue. It is not so much that

international governance structures may or may not be converging to an optimal form, rather the

fundamental issue is whether the overarching purpose or objective of the public corporation is

changing throughout the world. It is our assessment that such a transformation is, in fact, taking

place and this transformation is toward a system that is shareholder-oriented. As such, we believe

that we are not witnessing a convergence but rather a conversion - a conversion to the notion that

the purpose of the business corporation is to enhance shareholder value.6

Corporate governance itself is just a process that serves a larger purpose, namely the goal

or purpose of the corporation. The outward manifestation of a governance system tells us little

about the underlying goal that guides the firm’s activities. We may find outward manifestations

of the governance process that seem to be different (e.g., labor representation versus no labor

4 This literature grounds its arguments on the belief that the globalization of product and capital markets and theemergence of a worldwide market for corporate control will lead firms to adopt a contractarian governance structure.

5 See Easterbrook (1997). As Bradley et. al (1999, p. 79) argue, “…the efficiencies and contractarian imperatives ofglobal competition [would] propel changes in the underlying legal, political, and social infrastructures in ….communitarian economies” and thus attenuate the influence of factors that cause persistence. More recently, thearguments have come full circle. Some recent scholarship raises the thesis that the arguments for path-dependence,convergence, and even more specifically, convergence to the American norm are not only wrong since there aredifferent types of “embedded capitalism” (see e.g., Branson, 2001).

6 Convergence implies gradual, evolutionary change; conversion implies rapid, revolutionary change. Convergencesuggests that there may be changes in the observable manifestations of governance at the corporate level, i.e., in theinternal mechanisms of governance such as managerial compensation/reward systems, board structure andcomposition, and so forth; conversion suggests that we should also observe changes in the external governancecontext, e.g., changes in the market for corporate control, the product market, the capital market, ownershipstructure, and regulatory environment. Convergence is consistent with arguments for social and culturalembeddedness, with law, interest group politics, and culture being exogenous variables and therefore the persistenceof inefficient governance forms; conversion is consistent with the law-and-economics viewpoint that embeddingforces such as law, interest-group politics, and culture are themselves endogenous. Convergence implies that twodifferent governance systems are moving toward each other and will meet somewhere in between; conversionimplies a unidirectional movement of one system to the other.

5

representation in the boardroom) and yet, upon deeper examination, find that both processes

could be serving an identical corporate purpose (e.g., ‘maximizing shareholder value’).

Our thesis is that in order to understand comparative corporate governance, we would be

better served by focusing attention on whether there is convergence in corporate purpose, rather

than a convergence in form. We do so by undertaking a detailed empirical assessment of

developments in the third largest industrialized economy in the world—that of Germany. The

German experience offers us a rare, and fortuitous ‘laboratory setting’ to study a transformation

in corporate purpose and consequently, governance practices. Specifically, we examine the

changes in corporate governance practices and their efficiency implications, as well as the

changes in the legal-political-economic environment of the large, publicly-traded German

corporation during the 1990s. In this examination, we ask the following questions: (1) Are the

transformations in the German governance system that we observed over the decade of the 1990s

consistent with the traditional stakeholder-oriented German corporation, or are we witnessing the

third largest industrial economy in the world moving toward a shareholder-oriented corporate

economy? (2) What are the efficiency and share price implications for German companies

adopting such practices? (3) What are the implications of these changes for the traditional

governance practice embodied in the stakeholder doctrine of ‘codetermination,’ which provides

for an explicit role for labor in the governance process? (Prior to the decade of the 1990s,

codetermination had been the centerpiece of German corporate law.)

2. German Governance: A Brief Historical Overview

A substantial amount of scholarship has been devoted to understanding the governance of

the large, publicly-traded German corporation (for a detailed assessment and references, see

Bradley, Schipani, Sundaram and Walsh (1999))7. We present only a brief summary of the

salient features of the traditional German governance system in order to provide a backdrop for

the dramatic changes we have witnessed during the 1990s.

There are two basic organizational forms for large-scale German companies: stock

corporations or “Aktiengesellschaften” (“AG”) and limited liability companies “Gesellschaften

7 Early contributions include Roe (1993); Prowse (1994); Roe (1994). Others have examined transitions in Germangovernance during the late 1990s, e.g., Baums and Birkenkaemper (1998); Gordon (1999); Logue and Seward(1999); Nowak (2001); see also the 1999 Symposium Issue of the Columbia Journal of European Law.

6

mit beschrankter Haftung” (“GmbH”). The primary distinction between the two organizational

forms is that the shares of AG firms are traded on organized exchanges.8 An important difference

between German corporations and their Anglo-American counterparts is the existence of a two-

tiered board structure. By law, all AGs and large GmbHs are required to be governed by a

supervisory board (“Aufsichtsrat”) and a management board (“Vorstand”). The supervisory

board, which is responsible for strategy formulation, appoints, oversees and, if necessary,

disciplines or even dismisses members of the management board. The management board has

responsibility for strategy implementation, i.e., the day-to-day operations of the firm. The

Aufsichtsrat can include outside directors, while the Vorstand consists of senior company

managers (equivalent to the officers in a US company). Members of the management board

cannot serve as members of the supervisory board and vice versa. The most notable aspect of the

German governance system is that the supervisory board must be comprised of a specified

fraction of employees, depending on the size and industry of the firm. This practice, known as

‘codetermination,’ is a legislated guarantee of employee involvement in the strategy and

operations of the firm.

German corporate law was initially codified in 1937, and subsequently modified in 1965.

Under the 1937 law, the role of the board of directors and the objective of the public corporation

are defined as follows (the original German words are in parentheses): “The managing board is,

on its own responsibility, to manage the corporation for the good of the enterprise and its retinue

(Gefolgschaft), the common weal of the folk (Volk) and the state (Reich).”9 Nothing specific was

mentioned in German corporate law about shareholders until the 1965 revision. The law also

provides that if a company endangers public welfare and does not take corrective action, it can

be dissolved by an act of the state. Despite the relatively recent recognition that shareholders

represent an important constituency (see discussion in the next Section), corporate law in

Germany has historically made it clear that shareholders are only one of the many stakeholders

on whose behalf the managers must manage the firm.

8 In 1998 there were more than 700,000 limited liability companies (GmbH) as compared to about 4,200 stockcorporations. Of the latter about 700 were, at the time, listed on the various German stock exchanges. These dataimply that most stock corporations in Germany are also privately-held.

9 Bradley et. al (1999), footnote 243 at page 52, quoting Vagts (1966).

7

Equity ownership structure in Germany differs quite substantially from that observed in

the US. As of the mid-1990s, approximately 14% of the shares of German corporations were

owned by banks, and about 40% by other German corporations (both constituencies hold

virtually nothing in the US). Less than 17% of the equity was owned by households (as

compared to approximately 50% in the US), a proportion that has steadily declined since the

1950s. Historically, ownership of German corporations has been concentrated in the hands of

financial institutions and other corporations. As of the mid-1990s, roughly 25% of the listed

German firms had a single majority shareholder, and such majority holdings accounted for about

65% of the value of all listed stocks. Importantly, a substantial portion of equity in Germany

traded in the form of bearer stock, unlike registered stock in the US. As a result, such equity was

left on deposit with the hausbank of the corporation, which handled matters such as dividend

transfers and record-keeping. German law allowed banks to vote such equity on deposit by

proxy, unless depositors explicitly instructed their bank to do otherwise.10 Compounding the

influence of banks was the ‘voting caps’ provision in many company charters, whereby non-bank

shareholders could not exercise more than 5% to 10% of the total votes, regardless of the

proportion of shares they owned. Furthermore, even when a company elected to have its shares

listed on an exchange, the common practice was to list only non-voting shares.

Historically, German firms have relied heavily on intermediated debt. Market debt

instruments played a relatively minor role in the financing of the typical German corporation.11

Dividend payout rules were designed to protect creditor interests.12 The market for corporate

10 As a result, banks directly or indirectly controlled a large portion of the equity in German companies. In a studyreported by the Organization for Economic Cooperation and Development (OECD, 1995) of large, non-majority-owned firms in Germany, Baums and Fraune found that nearly 85% of the voting shares in 1992 was controlled bybanks and their associated investment funds.

11 The minor role of market debt is explained by the restrictions placed by the German government on the issuanceof such debt. Issuance of commercial paper and domestic bonds was discouraged until 1992 by complicatedauthorization procedures and transfer taxes. The issuance of foreign currency bonds was prohibited until 1990, andthe issuance of Eurobonds was subjected to maturity restrictions. Also until 1990, there was a 1% tax on the value ofall new equity issues, and secondary trading in equities was subject to a transaction tax. Companies also paid a 1%tax on their net asset value, a tax that must be paid even if the firm is not profitable.

12 German law stipulated that dividends may not be paid out from paid-in capital, even if such paid-in capitalincludes a premium over the face value of equity. This provision made it impossible for German firms to undertakeshare repurchases. Moreover, companies were required to retain a portion of their profits as reserves, serving asadded security for creditors.

8

control was poorly developed.13 Hostile takeovers and leveraged buyouts were virtually non-

existent.14 There was formal takeover law. Disclosure was considered inadequate by Anglo-

American standards.15 Stocks and options as the basis for CEO compensation were virtually

unheard of, with compensation comprising a basic salary plus bonus. As a result, the average

salary for CEOs of large firms in Germany were about one-half to one-third of that for their US

counterparts (Murphy (1998)).16

In summary, the historical German governance system was characterized by little reliance

on external capital markets; a small role for individual share ownership and a disproportionately

large role for concentrated institutional/inter-corporate ownership; relatively permanent capital

providers; boards comprising functional specialists and insiders with knowledge of the firm and

the industry; a relatively important role for banks as financiers, advisors, managers and monitors

of top management; a preference for bank-debt financing; an emphasis on salary and bonuses

rather than stocks and options as the basis for top management compensation; and a relatively

poor disclosure regime from the standpoint of outside investors. In many respects, the German

governance system emphasized the welfare of employees and creditors over the interests of

shareholders. The market for corporate control was largely absent. Put simply, the traditional

German governance system can be characterized as a substantially communitarian, stakeholder-

oriented system, with little or no regard for shareholder welfare.

However, the German governance system has been far from static over the past decade.

Significant changes have transformed the governance and legal systems into a more shareholder-

oriented environment. As we will see below, these changes began to gather momentum in the

early 1990s.

13 Prowse (1994) reports that during the period 1985-89, only 2.3% of the market value of listed stocks was involvedin mergers and acquisitions, compared to over 40% in the US.

14 There were only four hostile takeovers in Germany since World War II through the late 1980s.

15 In an OECD survey of corporations across the US, Germany and Japan, the companies were rated relative toOECD guidelines for “full,” “partial” or “not implemented” disclosure. Two thirds of the US firms surveyed met the“full” disclosure standard and the other one third of US firms surveyed had “partial” disclosure. In contrase, none ofthe German firms surveyed met the OECD’s “full” disclosure requirement.

16 German CEOs were limited by law to a “reasonable” compensation. Their compensation averaged about 21 timesthat of an average German worker (Fukao (1995)) at a time when the average American CEO used to make at least120 times the salary of an average American worker (Bradley et. al (1999)).

9

3. Changes In German Governance: Regulations

The change in corporate focus of the German corporation could not have happened if the

regulatory environment of public corporations in Germany had not changed as well. In this

Section we review the major changes in German corporate law that have facilitated the change in

corporate purpose that we observe. We describe the legal milestones during the past decade, and

designate three broad regulatory regimes for German corporations over this period: the

‘codetermination’ regime (the prevailing norm until early 1990), the ‘transition’ regime (1990 –

1994), and, the ‘shareholder value’ regime (1995 – present).

As already mentioned, the supervisory board of a large, public German corporation must

be comprised of a specified fraction of employees. This practice, known as codetermination, is a

legislated guarantee of employee involvement in the strategy and operations of the firm. The first

codetermination law was enacted as the 1951 Coal and Steel Codetermination Act, which applied

strictly to those two industries. Under the legislation, the supervisory boards of firms in these

industries had to have equal representation between employees and shareholders. In 1952, this

legislation was extended to cover to all industries, requiring that a third of the seats be given to

employee representatives in companies with 500 or more employees. The most recent version of

the law, the Codetermination Act of 1976, requires that employees elect half of the members of

the supervisory board in all companies with 2,000 or more employees.

Codetermination occurs at two levels within the German corporation. At the company

level, employee input into the governance process is assured by their representation on the

Aufsichtsrat. At the plant level, codetermination occurs within groups known as “Works

Councils” (“Betriebsrat”). German law specifies the size and composition of works councils for

each firm, again depending on its size and industry. For large corporations, (defined as those

with more than 600 employees) a specified number of the members of the work council must be

“freed from work” in order to attend to corporate matters.

Prior to 1990 there was no unified capital market law in Germany. Rules and regulations

governing the issuance and trading of securities were found in various parts of the law, such as

corporation law, securities exchange law, and banking law. The Prospectus Act of 1990, which

was the major component of the ‘First Financial Market Promotion Act,’ was the first legislative

act in German history to have as its primary goal the protection of investors in German capital

markets. The Act governed the prospectus requirements for all first-time securities that are

10

offered to the public and had not previously been registered to trade on a German stock

exchange.

Far-reaching securities legislation was enacted by the German Parliament via the ‘Second

Financial Market Promotion Act’ (SFMPA), which came into effect in January 1, 1995.

Although the intent of the law was to define and curtail the practice of insider trading, the

creation of a system and a process to monitor, detect, prosecute and punish violators had a

significant effect on all aspects of the German equity market. The legislation created the Federal

Supervisory Office on Securities Trading (FSO), an agency roughly equivalent to the Securities

and Exchange Commission (SEC) in the US. In an attempt to provide the Supervisory Office

with sufficient information to prosecute violations of the insider-trading laws, the legislation

mandates that issuers inform the FSO of any new non-public facts that could have an effect on

the issuer’s financial position or general business activities and which could, if publicly known,

have a significant effect on the price of its securities.17

Legislation in 1998 was the next major step in the transformation of the German

corporate environment. The ‘Third Financial Market Promotion Act,’ otherwise known as the

“Control and Transparency in Business” Law (abbreviated as ‘KonTraG’ in German)

implemented changes that affected the laws governing the activity of corporations, stock

exchanges and importantly, accounting practices. In announcing the new legislation, the Federal

Ministry of Justice issued the following press release, in English:

“The adoption of further mandatory provisions in the Germancorporation law ought to be avoided as far as possible. Instead of strictdirectives, it is preferable to leave companies more leeway in organizingthemselves. Control should be provided by the existing supervisory boards andthe markets. The law should actively keep pace with public corporations as theygear up to the requirements and expectations of international financial markets.This also means that corporate strategy needs to be more strongly orientedtowards shareholder value.”18 (Italics ours).

17 This mandate is in contrast to the “disclose or abstain” philosophy toward insider trading in the US. As long asinsiders do not trade on non-public information, they need not disclose this information. Although compliance is anissue, the Germans have adopted a continuous disclosure process whereby insiders must divulge any “material”information even if they have no intention of either buying or selling the firm’s securities. Interestingly, the Actprovides no private right of action for corporate stockholders. Only the FSO has the right to prosecute a violation ofGermany’s insider trading laws.

18 Interestingly, in the German version in this last sentence concerning the shareholder value orientation the term“Steigerung des Unternehmenswertes” (‘increase of the value of the company’) is used.

11

The Act covered a wide range of issues of importance to shareholders, including the conduct of

the annual general meetings, the use of proxy voting rights of banks, the enabling of stock option

programs, allowance of share repurchases (which had been previously disallowed under German

law), disclosure of inter-corporate share ownership, limitations on serving on multiple boards,

and the use of voting caps.19

The 1998 KonTraG was followed by the Tax Reduction Act of 2001 (effective January

2002), which eliminated capital gains tax from the sale of stock by corporate shareholders. The

basic purpose behind this legislation was to provide incentives for the inter-corporate

ownership—estimated to be about 40% of the German stockholdings—to be unwound, and

equally important, to give the German banks the incentive to reduce their equity holdings of the

corporate sector. The 2001 tax law was, in turn, followed by the Securities Acquisitions and

Takeover Act of 2002, which for the first time formally regulates mergers with, and acquisitions

of, publicly traded German firms. The Takeover Act addresses rules governing tender offers,

Supervisory board roles in takeover situations, stock versus cash-based offers, hostile takeovers,

and minority squeeze-outs.20

We thus observe three distinct regimes, each with its own implications for the conduct

and purpose of the German corporation: a period spanning the late 1930s to 1990 representing a

stakeholder-oriented ‘codetermination’ regime, with labor and banks playing substantial

governance roles; followed by a ‘transition’ regime from 1990 to 1994, a period in which

legislation was enacted that provided greater protection to shareholders’ interests somewhat

indirectly; and finally, from 1995 and on, a ‘shareholder value’-oriented regime that not only

directly focuses on shareholder value creation and shareholder protection concerns, but also

seeks to limit the influence of other stakeholders, especially banks. We present the relevant time

line in Figure 1. We will return to this classification later in the paper, when we present our

analysis of the implications of these distinct regimes for the wealth of German stockholders.

19 For instance, banks are required under KonTraG to advise shareholders of alternative ways of voting their shares.Also, they may no longer vote proxies at an annual general meeting if, at that meeting, they are also exercising votesof their own holdings in the company of more than 5%. The KonTraG facilitates the introduction of stock optionprograms as part of remuneration for top management and other employees. After the KonTraG, share repurchasesare generally allowed (although there are still a number of restrictions that companies must abide by). Listedcompanies must list their holdings of more than 5% in other large corporations in their annual reports. An individualcannot be a member of more than 10 supervisory boards. Finally, the KonTraG denies the use of voting caps thatlimit the percentage of votes by individual shareholders.

20 Prior to this, Germany had only a ‘voluntary’ takeover code.

12

Figure 1Governance Regimes for German Corporations

Codetermination Transition Shareholder Value

1937… Æ 1990 1995… Æ Present

4. Changes In German Governance: The Rhetoric of Governance and Corporate Practice

The German language had no equivalent expression for the English language phrase

‘shareholder value’ until the mid-1990s.21 But as we will demonstrate in this section, there was

not only a dramatic rise in the use of the English-language phrase ‘shareholder value’ in the

German media around 1991, but starting in 1995, a new German word surfaced to describe this

phenomenon.

From 1980 through 1990, there does not appear to have been a single instance in which

the English term ‘shareholder value’ appeared in the German language media (see Table 1). The

first use of this phrase appears to have been in the publication Sueddeutsche Zeitung on July 12,

1991. In 1992, there were two mentions and in 1993, two more mentions. In the following year,

there were 11 mentions. However, the year 1995 seems to have been the ‘breakthrough’ year for

the adoption of ‘shareholder value:’ its use increased nearly 17-fold, to 182 mentions. By the

year 2000, there were well over one thousand mentions of this English phrase. During the period

1991 (when it first appeared) to 2001, there were a total of 6621 mentions of ‘shareholder value.’

(TABLE 1 HERE)

21 Bradley, Schipani, Sundaram and Walsh (1999) at 52, quoting Stephan Wagstyl (1996). As a result, the standardpractice in referring to the phenomenon was (and continues to be) to simply use the English language phrase!

1937 Corporation Law(modified in 1965);

Codetermination lawsof 1951, 1952, 1976

1st FinancialMarket Promotion

Act of 1990

2nd Financial Mkt Promotion Actof 1995, KonTraG of 1998, Tax

Reduction Act of 2001, Securitiesand Takeover Act 0f 2002

13

In 1995, a German word ‘Aktionaerswert’ appeared for the first time in the German

language media (‘Aktionaer’ = shareholder; ‘Wert’ = value).22 According to our research, it first

appeared on March 14, 1995 in a German language publication from Zurich, Switzerland (in

Neue Zuercher Zeitung). The first recorded date of the use of this phrase in Germany itself is not

until five months later, on August 22 (in Frankfurter Allgemeine Zeitung, FAZ). It would appear,

however, that this word did not quite catch on among business writers in Germany. As can be

seen from the data in Table 1, in its first year of use, the new word was mentioned twice. Its peak

usage was six years later in 2000, when it was mentioned 15 times. By 2001, the number of

mentions of this phrase had declined to three. During the seven-year period 1995-2001,

‘Aktionaerswert’ was used a total of 41 times in the German language publications tracked by

the Lexis-Nexis database. A side-by-side comparison of this German word with the German use

of the English phrase ‘shareholder value’ is quite interesting. It would seem that while the usage

of the English phrase has become fairly commonplace, the German language equivalent not only

appears to have not caught on, but its usage may be disappearing as well.

The dramatic rise in the usage of the phrase “shareholder value,” and the attempts to

create a German-language equivalent—albeit unsuccessful—suggests that something significant

did change in the German mindset during the 1990s, at least in terms of the public rhetoric on

corporate governance.23

We now turn to a longitudinal analysis of the extent to which the concept of shareholder

value maximization was embraced by the managers of the largest public corporations in

Germany. Specifically, we examine the annual reports, management discussions, incentive

systems, and disclosure practices of these firms. We have chosen to focus on the so-called DAX

22 We are grateful to Jutta Ulrich (a Professor of German language at Thunderbird) for pointing us to thisdevelopment.

23 In the organizational studies literature, it has been suggested that the study of shareholder value orientation ofGerman corporations has been done with a presumption of a ‘context-free theory’ derived in the Anglo-Americansetting, and that the diffusion of the ‘shareholder value’ ideology did not result in a wholesale adoption of thepractice (Zajac and Fiss, 2001). The authors also suggest that while the rhetoric of ‘shareholder value’ may be on therise, it has been top-down in nature and that there may be no ‘actual implementation of such strategic shift’ (p. 14).We shall provide ample evidence to the contrary below.

14

30 firms.24 In order to measure the change in attitude towards shareholder value, we examined

the annual report of each of the firms in the current DAX 30 at three points in time,

corresponding to the three different time-periods (regimes) that we identified in the previous

Section. Ideally, we would like to have data from the 1980s, clearly a period of Codetermination.

However, the availability of annual reports falls off dramatically prior to 1990. Therefore, we

begin with 1990, a year we have designated as the beginning of a transitional period during

which German firms began to focus on shareholder value. We also examine the annual reports of

these firms in 1995, the year of the Second Announcement and again in 1999, the year of the

Transparency Act. We obtained the annual reports through direct communication with the firms.

We obtained both the English and the German versions of the annual reports.25 In examining the

annual reports, we focused on five factors:

1. The objective of the firm as stated in the “official” letter to the firm’s stockholders;

2. Whether the firm had an employee stock purchase plan;

3. Whether the managers of the firm had stock-based compensation packages;

4. Whether the managers had stock option plans;

5. The firm’s choice of accounting method.

Table 2 reports the results of our analysis of the annual reports of the firms in the current

DAX 30 regarding corporate objectives. Our primary focus is on the language used by the

managers of these firms when communicating their firms’ objectives to their constituencies.

(TABLE 2 HERE)

It is rather astonishing that in 1990 more than half (54%) of the 22 firms in the current

DAX 30 that existed at the time did not even mention stockholders in their annual reports. We

found statements to stockholders in the remaining 46%, but the comments were more in the form

24 One of the criteria for inclusion in the DAX is the depth of the market for a firm’s securities. The DAX 30accounts for more than 70% of the market capitalization of the Frankfurt Stock Exchange and for about 75% of thetotal daily volume of all German equities. Consequently, the equity of firm’s in the DAX trade much morefrequently than the typical German corporation. Foreign investors hold about 40% of the DAX. Arguably, an active,international market for a firm’s shares would make the firm’s management more responsive to shareholderpreferences.

25 Our research assistant Pia Mussell did all of the translation of the German versions. Her knowledge of bothGerman and English enabled analysis of the differences, if any, between the two versions.

15

of a ‘thank you’ note from the firm’s Managers and Directors.26 Importantly, none of the 22

firms mentioned anything about maximizing shareholder value or firm value in their annual

reports in 1990. Yet in 1999, 27 of the 30 firms (93%) not only addressed shareholders directly,

but also stated that the firm’s objective was to either increase the value of the firm or increase the

value of shareholder wealth. (Only 2 of the 30 firms did not address stockholders directly in their

1999 annual report.) Clearly there had been a sea change between 1990 and 1999, at least

regarding the rhetoric of creating shareholder value among the managements of Germany’s large

public corporations.

In 1995, 58% of the firms mentioned value-orientation as their corporate objective; 27%

of the firms used the English term ‘shareholder value’ in the German version of their annual

report. In 1999, 93% mentioned value-orientation as the main goal (however, only 17% used the

English term ‘shareholder value’ in the German version of their annual report). In 1995, only

27% expressed an intent to ‘increase the value of the company’; but in 1999, more than twice as

many companies as in 1995 (62%) stated that their objective was to ‘increase the value of the

company.’27

We next examined the annual reports for any indication that the firms in the DAX 30 paid

more attention to the market price of their firms’ stock. The data in Table 3 address this. In 1990

only 8 firms (36%) had employee stock purchase programs. By 1999 that number had increased

to 19 firms or 63% of the DAX 30. Obviously, the incentives for employees to maximize the

value of their firms’ equity increased dramatically during the decade of the 1990s.

(TABLE 3 HERE)

26 Typical among these comments are those from Bayer AG: “Dear shareholders, . . . at the same time we thank allof our business partners and finally you, our stockholders, for your confidence in our Company,” (1990, p.7); fromDegussa-Huels AG, “Dear Shareholders . . ., the Board and employees of Degussa thank you for your continuedsupport,” (1998/90, p.7); from Preussag AG, “Message to our shareholders . . . the merger of Preussag AG andSalzgitter AG has created a group with new qualities, not only for our shareholders, employees and the public, butindeed for competition as well,” (1990, p.12); and finally from RWE AG, “From the Chairman (German version:Dear Shareholder) . . . the trust and support of you, the shareholders, is all important,” (p. 4-6).

27 In the years following 1995 companies seemingly tried to avoid the English term “shareholder value” and startedusing other terms. Instead of talking about increasing “shareholder value,” they talked about increasing the “value ofthe company” (“Steigerung des Unternehmenswertes”).

16

A similar pattern emerges in terms of the incentives of corporate managers to maximize

shareholder value. In 1990 not one firm in the DAX 30 reported a stock-based compensation

plan for their managers. However, by 1995 seven firms (27%) had adopted such compensation

plans and by 1999, the number jumped to 22 or 73% of the firms in the DAX 30. The adoption of

stock-option plans followed at a much slower pace. None of the firms in the DAX 30 granted

options to their managers in 1990.28 In 1995 only one firm (Siemens) did so. However by 1999

almost half of the DAX 30 (47%) firms had instituted stock option plans for their management.

We also examined if and when the firm adopted either international accounting standards

(IAS) or US GAAP. A change to a more rigorous accounting method would evidence an attempt

by managers to facilitate better monitoring of the firm’s activities in order to attract outside

investors. A change to IAS or GAAP can also be seen as a commitment device designed to

assure investors that the firm’s management will act in their interest, i.e., become more

shareholder-oriented. None of the firms in the current DAX 30 operated under either IAS or US

GAAP in 1990. By 1995 five firms or 25% of the sample had changed to one or the other

accounting system, and by 1999, 83% had done so.

Since accounting under US GAAP is a prerequisite for listing on a US stock exchange,

those firms that have issued American Deposit Receipts (ADRs) or have their shares listed in the

US are a subset of the previous sample (line 4 in the Table 3). In line 5 of the table we report the

number of DAX 30 firms that had either their shares or an ADR listed during the indicated time

period. The number of firms in the current DAX 30 that had shares or ADRs traded in the US

grew from zero in 1990, to 3 in 1995, to 15 or 50% by the year 2000.

The data presented above paints a clear picture of a significant change in corporate

purpose over the past decade for the vast majority of firms in the current DAX 30. Perhaps the

most direct evidence of this change is the fact that in 1990 not one firm in the sample even

addressed the fate of the firm’s stockholders in their annual reports, whereas in 1999 only two

firms ignored shareholder concerns in their annual report. More dramatically, in 1999 most firms

avowed a goal of either maximizing shareholder value or maximizing the value of the firm.

As we all know, talk is cheap. Actions are driven by incentives, not rhetoric. But here

again, we see a definite trend, as reflected in the incentive systems for both employees and

managers to maximize shareholder value. The increased popularity of employee stock purchase

28 At least, none were reported in their 1990 annual reports.

17

programs, stock-based compensation plans and executive stock-options over the past ten years

all suggest that the employees and managements of these companies had become more focused

on creating shareholder value by 1999 than they were a decade prior. We also see the adoption of

IAS or GAAP as a commitment device by managers to signal that they are acting in the interest

of the firm’s stockholders.

None of this evidence amounts to much unless it is ultimately reflected the performance

of the firm and its equity. We will turn to that issue in the Section after the next. But before we

examine the effects of the changing environment in the 1990s on the performance of German

stocks, we review the rationale, philosophical underpinnings, and more important for our

purposes, the specific empirical implications of a shift in emphasis toward shareholder value

maximization. In other words, as a firm’s operating policies change from focusing on

stakeholder welfare to shareholder welfare, what effects should we expect to observe in its

operating and financial policies, and its stock market performance?

5. The Rationale for, and Implications of Shareholder Value Maximization

In the Anglo-American, especially American, system of governance, the logic of

shareholder value maximization is so taken for granted that its origins, meanings, and

implications are often just assumed to be understood. As is well known, in the US, the

shareholder value logic finds its roots the in famous Dodge v. Ford decision rendered by the

Michigan Supreme court in 1919.29 The philosophy behind the Dodge v. Ford decision has been

subsequently reflected in the Model Business Corporation Act (and its revisions), in the ALI

Principles of Corporate Governance, and in the various corporate laws of the fifty US states.

29 The issue in this case was whether or not Henry Ford could use the substantial accumulated earnings of the FordMotor Company to build a steel foundry for the purpose of reducing the price of automobiles to the Americanconsumer. This action was brought by the Dodge brothers, as shareholders of Ford Motor Co. to compel thepayment of these accumulated funds as a dividend. At the time the company had 1500 shares outstanding. HenryFord owned 225 and the Dodge brothers owned 50. In the much-cited (and celebrated) decision, the court stated:“There should be no confusion (of which there is evidence) of the duties which Mr. Ford conceives that he and thestockholders owe to the general public and the duties which in law he and his co-directors owe to protesting,minority stockholders. A business corporation is organized and carried on primarily for the profit of thestockholders. The powers of the directors are to be employed for that end. The discretion of directors is to beexercised in the choice of means to attain that end, and does not extend to a change in the end itself, to the reductionof profits, or to the non-distribution of profits among stockholders in order to devote them to other purposes.”(Italics ours.)

18

Yet, it is useful to revisit the following questions: What is the rationale for shareholder

value maximization, i.e., why should mangers act solely on behalf of maximizing the wealth of

the firm’s residual claimants, and not on behalf of other stakeholder constituencies such as

employees, suppliers, and creditors? Why has it become such a dominant corporate purpose in

Anglo-American countries, and increasingly so in other major economies around the world?

What specific implications for firm conduct and performance follow from the adoption of

shareholder value maximization as the corporate objective? In a recent study that synthesizes and

analyzes literatures from law, finance, strategy, and organizational studies, Sundaram and Inkpen

(2003) address the first two questions at length. Specifically, they argue that there are five

rationales for the primacy of the shareholder value maximization rule and that these rationales

illustrate why this has become the most preferred among all available objective functions that

managers could rely on as a guide in the conduct of their business.

Rationale 1: Maximizing Shareholder Value Maximizes the Value of the Whole Firm

Synthesizing the work of Alchian and Demsetz (1972), Easterbrook and Fischel (1983),

and Macey (1989; 1991) among others, Sundaram and Inkpen argue that only residual cash flow

claimants have the incentive to maximize the total value of the firm, since claimants to

committed or fixed cash flows—such as employees and debtholders—have no incentive to

increase the value of the firm beyond the point at which their claims are assured.

Control rights should go to shareholders since, as residual claimants, they are the

constituency that will value this right most. Shareholders have the greatest incentive to induce

firms to engage in activities that fixed claimants would consider excessively risky, since they

gain all of the benefits from the success of risky activities, but because of limited liability, stand

to lose only the amount of their initial capital investment. Fixed claimants do no better whether

the firm performs ‘spectacularly well’ or just ‘well.’ Recognizing the incentives for managers to

take on excessively risky projects, particularly when the firm is in financial distress, creditors

will ‘price-protect’ themselves and adjust the price they would pay for their fixed claims to

compensate for the prospect that the firm subsequently will undertake activities that lead to an

expropriation of their wealth. Thus, it will be the shareholders, not fixed claimants, who bear the

costs of any anticipated excessive risk taking. In essence, managing on behalf of shareholders

forces managers to go beyond effort levels that would suffice were they were to manage the firm

19

on behalf of its fixed claimants. By going beyond the requirements of such committed claims,

managers increase the size of the pie for all constituencies.

A corollary of this argument is that managing on behalf of fixed claimants (stakeholders)

will lead the firm forego positive-NPV projects, lower its value, and thereby increase its cost of

capital.

Rationale 2: Only Shareholder Value Maximization Is Consistent With Appropriate

Entrepreneurial Risk-taking Incentives

Sundaram and Inkpen (2003) draw upon the work of Bradley, Schipani, Sundaram and

Walsh (1999) to argue that managers often bear unacceptable amounts of non-diversifiable

human capital risk in one asset: the firm for which they work. Even though they may be

diversified in their personal portfolio and the firm’s stock, this may not be sufficient to

compensate for the risk of unemployment. This can lead managers to focus on the total risk

characteristics of the firm’s cash flows, when what actually maters for shareholder value

maximization is risk that cannot be diversified away, i.e., the systematic risk of the assets

contained in a well-diversified portfolio. Put simply, managers who manage on behalf of

shareholders care about the systematic risks of a firm’s cash flows in the context of a diversified

portfolio and are only as risk-averse as the shareholders at the margin, whereas poorly diversified

managers care about the systematic plus any unsystematic risks of the firm’s cash flows and thus

tend to be more risk averse than their shareholders. Managing on behalf of fixed claimants such

as debtholders and employees exacerbates the incentive for entrepreneurial risk aversion since

fixed claimants are also driven to minimize total risk. Such entrepreneurial risk aversion can lead

to an over-reliance on decisions to stabilize cash flows such as survival or maximizing market

share. Entrepreneurial risk aversion can lead to unprofitable diversification, excessive hedging,

and avoidance of high-growth investment opportunities. Again, a corollary, as with Rationale 1,

is that such behavior would lead to an increase in the firm’s cost of capital and a decrease in its

market value.

Rationale 3: It Is Not Possible to Maximize More Than One Objective Function

Having more than one objective function creates the potential for confusion and dithering

in managerial decision-making. As Jensen (2001, p. 10) argues, “multiple objectives is no

20

objective,” since “it is logically impossible to maximize in more than one dimension at the same

time unless the dimensions are monotone transformations of one another. The result will be

confusion and lack of purpose that will fundamentally handicap the firm in its competition for

survival.”30 Put simply, it is not possible to manage on behalf of multiple constituencies when

their goals are in conflict. Even if it were possible to do so, it may not be socially desirable to

allow managers the unfettered freedom to do so. Shareholder value, on the other hand, is a

single-valued metric that is also observable and measurable and whose maximization is

consistent with value maximization for other constituencies as well.

Rationale 4: Stakeholders Can Become Shareholders, But the Reverse Is Not Easy

Sundaram and Inkpen point out that there is no reason why non-share-owning

stakeholders, who are concerned about the potential abuses of shareholder value maximization,

could not choose to become shareholders themselves. In the case of a publicly traded firm, these

constituencies can buy shares in the open market, thereby becoming shareholders and availing

themselves of all shareholders’ rights. However, it is not as easy for the reverse to happen. For

example, it is impossible for a stockholder to demand to become an employee or a supplier of the

firm. This being the case, shareholders would be denied the opportunity to participate in the

governance process in those firms that manage on behalf of alternative constituencies.

Rationale 5: The Law Fills the Judicial Void for Stakeholders

The interests of stakeholders such as employees, suppliers, bondholders, communities,

and customers are protected by contract law and by regulation (Bradley et. al, 1999: 24-29;

Ramseyer, 1998; Hansmann and Kraakman, 2000: 10).31 Shareholders have no recourse through

contract law. Essentially, as Macey & Miller (1991) argue, the legal system derived the fiduciary

30 This builds on the earlier work of Fama and Jensen (1983), and invokes the Arrow Impossibility Theorem (Arrow,1950): Under certain conditions for ‘fair’ decision-making, there is no consistent method of making a fair choiceamong three or more alternatives.–i.e., there is no majority-voting procedure that can always fairly decide anoutcome when more than two alternatives are involved. For example, apart from shareholders, if there are multipleclasses of stakeholders, then there is no fair way to democratically decide on whose behalf the board should vote tomanage the firm.

31 This is even more so in countries such as Germany where, as we have discussed previously, the extent ofprotection afforded to stakeholders far exceeds anything that such constituencies can hope for in the Anglo-American setting.

21

system for filling gaps that arise in the terms of shareholders’ implicit contracts with

management. It is not that stakeholders have all the contracting power they need or that

shareholders have no judicial recourse whatsoever. Rather, the argument is that stakeholders

have greater ability to explicitly contract with the firm and thereby have the backing of the

judicial system to step in to fill the inevitable voids created by the limitations of explicit

contracting.

Rationale 6: Expanding the Firm’s Equity Base

A final rationale for adopting a shareholder-value orientation, and one that is particularly

relevant in this study since we focus on large corporations—many of which are global in scope

and size—is to attract new equity investors from the global capital pool. This is especially true

for firms operating in environments without well-developed capital markets, which, as we

documented in a previous section, was the case with Germany in the early part of the 1990s.

In a series of influential papers, La Porta, Lopez-de Silanes, Shleifer, and Vishny (see, for

example, La Porta et. al. (1999)) show a positive relation between the existence of legal and

regulatory protections of (minority) shareholders’ rights and the existence of a well-developed

capital market. If, during this period, German firms were attempting to attract new equity capital

it would behoove them to act at least outwardly as if they were adopting a more shareholder-

oriented philosophy, and to adopt operating practices such as stock- and option-based

compensation plans, IAS or GAAP accounting, and so forth to signal a shift to shareholder-value

orientation.

Empirical Implications

The preceding arguments suggest a number of empirically testable implications when a

firm shifts its corporate objective away from stakeholder value-orientation toward shareholder

value-orientation:

• The adoption of a more shareholder value-oriented governance structure and pursuit

of a more shareholder value-oriented operating strategy should increase the value of

the firm’s outstanding equity;

22

• Since firms that adopt a more shareholder-oriented operating strategy will be more

willing to assume higher entrepreneurial risk, we are likely to observe an increase in

the Betas of these firms;32

• Shareholder-oriented firms should have higher Market-to-Book ratios and faster

growth in their equity values, compared to less shareholder-oriented firms.

6. Empirical Effects of Greater Shareholder Value Orientation

In the first part of this section, we examine changes in the characteristics of returns to the

stockholders of firms in the current DAX 30 over the decade of the 1990s. The purpose is to

examine whether the increased focus on shareholder value that we documented in Section 5 is

associated with increased returns to the stockholders (higher Alphas), higher Betas (covariances

with the market), and whether we can observe any effects of the change in corporate purpose on

shareholder value. In the second part of our empirical analysis, we examine the impact of this

change on the firms’ market values (specifically, the Market-to-Book ratios), and the rate of

growth in their market capitalization.

In order to conduct our empirical analysis, we construct an index of shareholder

orientation for the DAX 30 firms. We use this index to classify the firms in our sample into those

that are more shareholder oriented (MSO) and those that are less shareholder oriented (LSO). Of

course, we recognize that any such exercise is fraught with possible selection biases, omitted

variables, multicollinearity problems and other misspecifications. Nonetheless, we believe that it

is still useful to examine if any discernible patterns emerge between the two groups of firms.

6.1. The DAX 30 Sample and the DAX 30 Shareholder Orientation Index

Since firms in the DAX became shareholder-oriented over time and to differing degrees,

we have both time-series and cross-sectional data. We construct an index of ‘shareholder value

orientation’ by collapsing these observations into an index score for each firm in our sample. In

our scheme—admittedly arbitrary, and potentially prone to specification biases—each firm is

32 As Fama (1976) argues (see pp. 255-256), if there is no systematic relationship between correlations of returns onsecurities and standard deviations of returns, then there will be a relationship between the standard deviations ofsecurity returns and their average covariances with the returns on other securities. Securities with larger standarddeviations of returns will tend to have larger average covariances with the returns on other securities.

23

given points for the various data categories that we defined in Section 5.33 On some of the

attributes, firms are given an additional point if the designated action occurred or was adopted in

1995 as opposed to 1999. We believe that the earlier a practice is initiated or adopted, the greater

the effect it should have on the firm’s stock price.

We assign the highest value, a score of three points, to the adoption of a stock-based

compensation plan for managers, since we view this as direct evidence of an attempt to align

management and shareholder interests. If this adoption occurred earlier, in 1995 as opposed to

1999, we award an extra point. If the firm also adopts a stock option plan or creates a share

purchase program for its employees, we award an additional point each for such adoption. We

assign the second highest value, a score of two points, if the firm adopts a more transparent

accounting system in the form of either the IAS or the US GAAP, or listed an ADR on the

NYSE; an additional point is awarded if this occurred in 1995 instead of 1999. Finally, we award

one point if the managers mentioned in a particular year’s annual report that their objective was

to maximize shareholder value or firm value—one point for each year (1990, 1995, 1999). Thus,

the maximum possible number of points under this scheme is 12, and the minimum zero. A firm

with a score equal to or above the median is placed into the ‘MSO’ (more shareholder

orientation) sample, and a score below the median puts the firm in the ‘LSO’ (less shareholder

orientation) sample. A table detailing the scoring of each firm is included in the Appendix.34

This scoring scheme generates a range of scores from zero for KarstadtQuelle, to 11 for

Bayer. Table 4 lists the DAX 30 sample firms in decreasing order of their Shareholder

Orientation Index (hereafter called the DAX 30 Index). Because of data availability issues, we

also indicate which firms are in the sample that we use in our empirical tests.

(TABLE 4 HERE)

33 Despite these potential difficulties, many recent studies have, however, constructed and used such governancescores or indexes; see, for example, Gompers, Ishii, and Metrick (2003); Durnev and Kim (2002). In the context ofGerman firms, at least two recent studies have used such scores (Hoepner (2001) which is discussed in more detailbelow, and Tuschke and Sanders (2003)).

34 We have prepared a summary of the report to shareholders for each firm for each of the three chosen years. Thissummary is available upon request. Note that in the table in the Appendix, NA (not applicable) refers to firms thatwere not in existence at the indicated time.

24

6.2 An Alternate Sample: The Hoepner Shareholder Orientation Index

Since most of the firms in the DAX 30 increased their shareholder focus over the decade

of the 1990s, it could be that there is not sufficient cross-sectional variation among them, thus

potentially compromising our statistical tests for differential share price effects.

Fortunately, however, we have access to an alternate sample. A 2001 study by Professor

Martin Hoepner (of the Max-Planck-Institute for the Study of Societies) of large publicly traded

German firms during a time period similar to ours provides an alternate sample that we can use

to calibrate, and test the robustness of the results from our DAX 30 sample.35

Hoepner’s paper discusses the shareholder value-orientation of the 40 largest listed non-

financial German companies. Hoepner examined three dimensions of shareholder value: he

defines these as the ‘communicative’ dimension, the ‘operative’ dimension and the

‘compensation’ dimension. Hoepner measures the extent of shareholder orientation of the

companies in his sample by deriving a shareholder value score (hereafter called the ‘Hoepner

Index’) intended to capture these three dimensions, based on factors such as the clarity of the

annual report, the existence of an investor relations department, non-fixed top-management

compensation, and the implementation of shareholder-oriented profitability metrics. The firms

and the corresponding Hoepner index scores are shown in Table 5.36

(TABLE 5 HERE)

6.3 Empirical Analysis

In this section we examine the empirical effects of the changes in German corporate law

and the purpose of the large-scale, publicly traded German corporations over the past two

decades. Specifically, based on the hypotheses developed in Section 5, we test for differences

between the MSO and LSO samples on the following variables: (1) the mean return over various

governance regimes; (2) the Betas over the various governance regimes relative to the S&P 500

index of U.S. stocks; and (3) the variance of the returns to the relevant portfolios over the three

regimes.

35 “Corporate Governance in Transition: Ten Empirical Findings on Shareholder Value and Industrial Relations inGermany,” Max-Planck-Institute Working Paper No. 05/2001.

36 The major finding of the Hoepner study is a positive and significant relation between his index and the number ofshares held by institutional investors. He did not investigate the returns to the equities of these firms.

25

We divide the time periods as follows:

Period Description

1/1980 – 12/1989 Codetermination

1/1990 – 12/1994 Transition to Shareholder Value

1/1995 – 12/2002 Shareholder Value

We obtained the returns to the common stock of the firms in our sample from the

Datastream database. This database contains returns dating back to 1970. However, several of

the current DAX firms either did not exist in the 1970’s or did not have shares traded in public

markets. In event studies such as these, there is always a tradeoff between the length of the time

period studied and the number of firms in the sample: the longer the time period, the more

precise the estimates of the relevant parameters; however the longer the time period required, the

fewer the number of firms in the sample. As a result, we decided to begin our analysis in 1980.

This choice reduces the number of firms in our sample to 22, which we felt was the minimum

acceptable.37 We then divided the time period from 1980 to 2002 into three sub-periods: 1980-

1989 (the period of codetermination); 1990-1994 (the period of transition to shareholder value);

and 1995-2002 (the period of shareholder value).

The eleven firms that we classified as having “more shareholder orientation” (MSO) have

scores that range from 11 to 8, with a mean of 9 and a standard deviation of 1. The eleven firms

that have “less shareholder orientation” (LSO) have scores that range firms 7 to 0, with a mean

of 5 and a standard deviation of 2. In order to test for the relation between the means and Betas

of the two groups over the three time periods, we run a seemingly unrelated regression (SUR)

analysis. This technique corrects for the serial correlation in the contemporaneous returns to the

two groups. We have two equations in the model, one for each group. In each regression we

include a dummy variable corresponding to the three time periods under study. We also include

three interaction variables consisting of the product of the dummy variables times the return on

the S&P index. Specifically, the system of equations we estimate is:

MSOt = A1,1 D1 + A1,2 D2 + A1,3 D3 + B1,1 D1 S&Pt + B1,2 D2 S&Pt +B1,3 D3 S&Pt + e1,t

LSOt = A2,1 D1 + A2,2 D2 + A2,3 D3 + B2,1 D1 S&Pt + B2,2 D2 S&Pt +B2,3 D3 S&Pt + e2,t

where MSOt is the weekly return to the portfolio of shares of the eleven DAX 30 firms we deem

more shareholder value-oriented, and LSOt is the weekly return to the portfolio of shares of the

37 In other words, eight of the firms in the current DAX 30 do not have data dating back to 1980.

26

DAX 30 firms that we deem less shareholder value-oriented. The Di are dummy variables that

equal 1 over the indicated period and zero otherwise, and S&Pt is the weekly return to the

Standard & Poors’ 500 index. The results of this analysis are presented in Table 6.

(TABLE 6 HERE)

The data in Table 6 show that the mean return to the two portfolios are positive in the

first period, however they are not different from each other. None of the constant terms (Alphas)

in periods 2 or 3 is significantly different from zero. Thus, there is no evidence that the mean

return to the MSO firms is any different than the mean return to the LSO firms in any of the three

periods.

The estimates for the Betas, however, reveal interesting patterns. First, note that the Betas

of both the MSO and the LSO firms relative to the S&P 500 returns are significantly greater than

zero for all three periods.38 Second, the Betas for both sub-samples increase over time: The Beta

for the MSO firms goes up from 0.30 in the codetermination period, to 0.37 in the transition

period, to 0.65 in the shareholder value period (the Beta goes from 0.28, to 0.40, to 0.55,

respectively, for the LSO firms).39 The last row of Table 6 indicates, moreover, that there is a

significant upward Beta shift in the shareholder value period (Period 3) relative to both the

previous periods, while there is no significant upward shift between the codetermination period

(Period 1) and the transition period (Period 2). The most intriguing result, however, is the

significant difference between the Betas in shareholder value period for the two sub-samples.

The data indicate that the Beta for MSO firms (0.65) is greater than the Beta of the LSO firms

(0.55), with the t-statistic of the difference being 3.67. Thus, the shift to shareholder value-

orientation in Germany appears to be associated with a greater covariance between the returns to

the MSO firms and the firms in the S&P 500. This is consistent with the hypothesis that,

compared to the LSO firms, the MSO firms are being priced similarly to the shareholder-oriented

US firms in the S&P 500.

38 This is, perhaps, to be expected. As Harvey (1991) showed, developed-country equity market indices, includingthat of Germany, are co-integrated across countries and behave as though they are ‘globally’ priced. In other words,regardless of the underlying cause of the co-integration, we should expect similar effects across both the MSO andthe LSO sub-samples.

39 Such a Beta shift is consistent with the time-varying price of covariance risk that Harvey (1991) has documented.

27

An obvious critique of the analysis presented in Table 6, and one suggested by the

findings in Harvey (1991), is that, since firms in both groups are moving toward shareholder

value-orientation, there may not be sufficient variation in the two samples to detect a significant

effect on mean returns and Betas. We are able to (at least, partially) circumvent this issue by

performing the same tests as above on the independently-developed Hoepner sample, which we

discussed in the previous Section. The Hoepner Index is constructed to have a mean of zero and

standard deviation of 1. (See Table 5.) We classify those firms with a score greater than zero as

being ‘More Shareholder Oriented’ (MSO) and those with a score of zero or less as being ‘Less

Shareholder Oriented’ (LSO). We are able to find returns data for 16 firms in the Hoepner MSO

sample, and 13 in the Hoepner LSO sample. The mean Hoepner Index in the MSO sample is .72,

with a standard deviation of .43. The mean score in the Hoepner LSO sample is –0.97, with a

standard deviation of .71.

As with the DAX 30 sample, we performed a SUR analysis on the returns to the shares of

the firms in the Hoepner sample. The results are reported in Table 7. The data show that the

mean return for the Hoepner MSO sample is significantly greater that the mean return for the

Hoepner LSO sample in the third period. Thus, unlike the DAX sample, the Hoepner sample

reveals a significantly higher return to the MSO sample over the third period. Similar to the

DAX 30 results, we find statistically significant Betas for both samples in all three periods, with

the Betas for the “shareholder value” period being significantly greater than those for the

“codetermination” period.

(TABLE 7 HERE)

As in the case of the DAX sample, the Beta is significantly greater for the Hoepner MSO

sample than the Hoepner LSO sample in the “shareholder value” period. Again, we are tempted

to conclude that this increase in the relative covariances with the S&P index is evidence of the

returns to more shareholder value-oriented German firms being generated by market factors

similar to those for their shareholder-oriented counterparts in the S&P index.

It may be argued that the increase in the Betas of the German firms in the “shareholder

value” period relative to the S&P is due to the fact that a number of German firms had ADRs

during this time period and not due to any increased focus on shareholder value per se. In order

28

to test for an ‘ADR effect,’ we re-estimate the parameters of the model excluding the four firms

in the Hoepner sample that had ADRs listed on the NYSE. The results of this test are reported in

Table 8.

(TABLE 8 HERE)

The results of the SUR for the Hoepner sample, excluding the four firms with ADRs, are

similar to those from the previous tests. The only major difference in the results is that the mean

return to the shares in the Hoepner MSO sample is not statistically different from the mean return

to the shares in the LSO sample in the shareholder value period. The t-statistic of the difference

falls from 2.15 in the total sample to 1.72 in the sample excluding firms with ADRs. Thus,

although the mean return is greater for the MSO sub-sample, the difference is not as great as it is

in the total Hoepner sample. This is evidence of a small ADR effect.

Since we find a significant increase in the Betas of the firms in the MSO sub-samples in

both the DAX and Hoepner samples, we turn to an examination of the change in variances of the

returns to the S&P 500, the DAX 30 and its components and the Hoepner sample. The weekly

standard deviation of the returns to the returns to the various portfolios over the three sub-periods

are reported in Table 9.

(TABLE 9 HERE)

Interestingly, the standard deviation of the weekly returns of the S&P 500 is 2.86% in the

“codetermination” period (1980-1989) and 2.88% in the “shareholder value” period (1995-2002).

This implies that the increase in the Betas documented for the German firms in the sample is due

to an increase in the standard deviation of the of the returns to these firms, or an increase in the

covariances with the returns to the S&P, rather than a decrease in the standard deviation in the

returns to the index. The data show that the standard deviation in the DAX index increased from

2.42% to 3.15% 10% over the same period. The data also show that the standard deviation of the

DAX 30 MSO sample increased more than that of the DAX 30 LSO sample, whereas the

opposite is true for the Hoepner sample. In the latter sample the standard deviation of the returns

in the MSO firms increased more than the increase in the MSO firms over the two time periods.

29

(The larger increase for the firms in the LSO sample is apparently due to the unusually low

standard deviation of this portfolio in the first period.)

We next examine the time-series of the ratio of the market value of equity plus the book

value of debt to the book value of assets (Market-to-Book ratios) for each of our sub-samples

over the decade of the 1990s. We also examine the annual growth rate in market capitalization

over the same period. The data are taken from the Worldscope database.

Table 10 shows that the average Market-to-Book ratio for the entire period is positive and

significant for both MSO and LSO firms in the DAX Sample. However, the ratios are not

significantly different from each other. The slope coefficients tell a different story. While the

slopes are positive and significantly different from zero for both, the slope coefficient for the

MSO firms is significantly greater than that for the LSO firms. The results indicate that the

growth in Market-to-Book was greater for the more shareholder-oriented group. The results for

the Hoepner sample, also reported in Table 10, tell a similar story. Here again, although the

average Market-to-Book ratio is greater for the MSO firms, the t-statistic of the difference from

the Market-to-Book ratio of LSO firms is not significant. Consistent with the results for the DAX

sample, the slope coefficient for the MSO firms is significantly higher than that for the LSO

sample. These results reinforce the interpretation that the Market-to-Book ratio increased

significantly more for the more shareholder value-oriented German firms during the decade of

the 1990s. The faster growth in the Market-to-Book ratios is consistent with the higher Betas for

the MSO firms.

(TABLE 10 HERE)

Table 11 shows that the annual growth rate in equity market capitalization over the

relevant time period was significantly higher form the MSO firms than the LSO firms in the

DAX sample. This difference is also positive for the Hoepner sample; however, it is not

statistically different from zero.

(TABLE 11 HERE)

30

Finally, we examine the restructuring activity of DAX firms as measured by their

divestitures and asset sales during the 1990s, compared to all German firms and to EU firms

(Table 12). The data are from the SDC database on mergers and acquisitions.

(TABLE 12 HERE)

First, note that compared to all German firms, DAX firms saw significantly higher levels

of divestiture activity (as a proportion of all corporate control transactions) during the 1990s:

Divestitures accounted for 61.7% of all the corporate control transactions of DAX firms, while

this proportion was 35.2% for all German firms (the t-statistic of the difference in proportions is

6.29). The same holds true when they are compared to their non-German EU counterparts: the

proportion of divestitures as a percentage of all corporate control transactions was 33.5% for

non-German EU firms (the t-statistic of the difference in proportions is 7.04). The results are

similar when we examine cross-border divestiture activity. Compared to all German firms and to

non-German EU firms, DAX firms undertook significantly higher levels of cross-border

divestiture activity as well: Cross-border divestitures accounted for 30.5% of all divestitures

done by DAX firms, while this proportion was 16.0% for all German firms (t-statistic 2.20) and

it was 16.0% for all Non-German EU firms (t-statistic 2.53). The evidence thus seems to indicate

that the period of shareholder value-orientation in the large German corporation coincided with

much higher levels of restructuring activity, presumably reflecting deconglomeration.

7. Conclusion

We began this paper by asking three questions: (1) Are the transformations in the German

governance system that we observed over the decade of the 1990s consistent with the traditional

stakeholder-oriented German corporation, or are we witnessing the third largest industrial

economy in the world moving toward an American-style, shareholder-oriented corporate

economy? (2) What are the efficiency and share price implications for German firms adopting

such practices? (3) What are the implications of these changes for the traditional governance

practice embodied in the stakeholder doctrine of ‘codetermination,’ which, prior to the decade of

the 1990s, had been the centerpiece of German corporate law?

31

The answer to Question 1 is unambiguous. Without a doubt, the large, publicly traded,

German corporation underwent a significant transformation in its corporate purpose during the

decade of the 1990s. German firms moved quite substantially from a stakeholder orientation to a

shareholder orientation, and this move was facilitated by far-reaching changes in German

corporate law during this period. We document the changes in these regulations and present

evidence that many firms took the opportunity afforded by the new regulatory environment to

adopt a more shareholder-oriented philosophy. Changes in operating and restructuring policies

such as the adoption of management incentive compensation plans, better disclosure policies,

ADR listings and the increase in divestitures and asset sales, all signal greater commitment to

align management and shareholder interests. We also observed a significant shift in the rhetoric

of shareholder value among both the German public and among German managers.

But rhetoric is not all that changed. In answer to Question 2, our analysis shows that

firms that adopted more shareholder-oriented governance structures and operating strategies

performed better, along certain dimensions, than did firms that were less shareholder-oriented.

Our most robust finding is that the Beta of the returns to the more shareholder-oriented firms,

relative to the US S&P 500, increased significantly more than the Beta of the returns to less

shareholder-oriented firms. This suggests that the managers of more shareholder-oriented firms

took on higher-risk projects that were more sensitive to global market fluctuations. We believe

that this higher covariance with the returns to large US firms is an indication that such German

firms are affected by similar market forces and perhaps, by similar pricing effects in the global

financial marketplace. We also saw that firms that embraced shareholder value-orientation

realized significant increases in Market-to-Book ratios and equity market capitalization relative

to their less shareholder-oriented counterparts

Our answer to Question 3 is speculative. We feel that even though this transformation

toward shareholder-orientation has been far-reaching, the changed corporate purpose does not,

by any means, signal an end to the German practice of codetermination. Recall the basic

rationale for why shareholder value maximization should be the corporate objective: By

assigning control rights to the residual claimant, the incentive is put in place to maximize the

value of the whole firm. In other words, to the extent that shareholder value maximization is the

only goal that is consistent with enlarging the size of the corporate pie for all claimants, it is

manifestly pro-stakeholder. Any potential conflicts between labor and equity can be overcome

32

by providing a share of the residual claim to labor. This is exactly what many Germans firms

have done, through their provision of stock- and option-based incentive schemes for managers

and employees. There is nothing that would then argue against labor as shareholders having a

voice on the corporate board. If anything, it probably legitimizes and even strengthens the

shareholder-oriented corporate purpose while simultaneously preserving codetermination, a

practice that reflects a uniquely German innovation in corporate governance and one that is

enshrined and cherished in German law. We therefore disagree with those who might argue that

this shift in the corporate purpose could somehow weaken codetermination: we would suggest

the reverse.

We admit the possibility that the governance index-based classification scheme we

developed is prone to potential biases and misspecifications. However, the fact that the results

are nearly identical in empirical tests across two independently constructed governance

classifications—both our Index and the independently developed Hoepner Index—lends

credence to our results. Moreover, our findings are consistent with what we might expect to

observe when the operating strategies of public corporations change from a stakeholder- to a

shareholder-orientation.

33

References

Alchian, A. A. & H. Demsetz, 1972. Production, information costs, and economic organization.American Economic Review, 62: 777-795.

Arrow, K. J. 1950. A difficulty in the concept of social welfare. Journal of Political Economy,58: 328-346.

Baums, T., and A. Birkenkaemper, 1998. Corporate governance in Germany. UniversitatOsnabruck Institut fur Handels-und Wirstschaftsrecht Working paper, October.

Bebchuk, L., and M. Roe, 1999. A theory of path dependence in corporate ownership andgovernance. Stanford Law Review, 52: 127-170.

Bradley, M., C. Schipani, A. Sundaram, and J. Walsh, 1999. The purposes and accountability ofthe corporation in contemporary society: Corporate governance a crossroads. Law andContemporary Problems, 62: 9-86.

Branson, D. M., 2001. The very uncertain prospect of ‘global’ convergence in corporategovernance. Cornell International Law Journal, 34: 321-362.

Bratton, W. W., and J. A. McCahery, 1999. Comparative corporate governance and the theory ofthe firm: The case against global cross-reference. Columbia Journal of Transnational Law. 38:213-298.

Durnev, A., and E. H. Kim, 2002. To steal or not to steal: Firm characteristics, legalenvironment, and valuation. University of Michigan Business School Working paper, August.

Easterbrook, F. H., 1997. International corporate differences: Markets or law? Journal of AppliedCorporate Finance, 9: 23-29.

Easterbrook, F. H. & D. R. Fischel, 1983. Voting in corporate law. Journal of Law andEconomics, 26: 395-427

Fama, F. E., 1976. Foundations of Finance: Portfolio Decisions and Securities Prices, NewYork: Basic Books.

Fama, E. F. & M. C. Jensen, 1983. Agency problems and residual claims. Journal of Law andEconomics, 26: 325-344.

Fukao, M., 1995. Financial Integration, Corporate Governance, and the Performance ofMultinational Corporations, Washington, D. C.: Brookings Institute.

Gompers, P., Ishii, J., and A. Metrick, 2003. Corporate governance and equity prices. QuarterlyJournal of Economics, 118: 107-155.

34

Gordon, J. N., 1999. Pathways to corporate convergence? Two steps on the road to shareholdercapitalism in Germany: Deutsche Telekom and Daimler Chrysler. Columbia Journal ofEuropean Law, 5: 219-241.

Gorton, G., and F. Schmid, 2000. Class struggle inside the German firm: A study of Germancodetermination. The University of Pennsylvania Wharton School Working paper, August.

Hansmann, H. & R. Kraakman, 2000. The end of history for corporate law. Discussion Paper No.280, The Center for Law, Economics, and Business, Harvard Law School.

Harvey, C. R., 1991. The world price of covariance risk, Journal of Finance, 46: 111-157.

Höepner, M., 2001. Corporate governance in transition: Ten empirical findings on shareholdervalue and industrial relations in Germany. Max-Planck-Institut fur GesellschaftsforschungDiscussion paper 01/5, October.

Jensen, M. C. 2001. Value maximization, stakeholder theory, and the corporate objectivefunction. Journal of Applied Corporate Finance. 14: 8-21

Kester, C. W., 1997: Governance, contracting and investment horizons: A look at Japan andGermany, in Studies in International Corporate Finance and Governance Systems: AComparison of the US, Japan, and Europe, D. Chew (ed.) New York: Oxford University Press.

LaPorta, R., F. Lopez-de Silanes, A. Shleifer, and R. Vishny, 1999. Investor protection andcorporate valuation. National Bureau of Economic Research Working Paper, August.

Licht, A. N., 2001. The mother of all path dependencies: Toward a cross-cultural theory ofcorporate governance systems. Delaware Journal of Corporate Law, 26: 147-205.

Logue, D. E., and J. K. Seward, 1999. Anatomy of a governance transformation: The case ofDaimler-Benz. Law and Contemporary Problems, 62: 87-112.

Macey, J. R. 1989. Externalities, firm-specific capital investments, and the legal treatment offundamental corporate changes. Duke Law Journal, 1989: 173-201.

Macey, J. R. 1991. An economic analysis of the various rationales for making shareholders theexclusive beneficiaries of corporate fiduciary duties. Stetson Law Review. 21: 57-90.

Macey, J. R. & G. P. Miller, 1991. The plaintiffs’ attorney’s role in class action and derivativelitigation: Economic analysis and recommendations for reform. University of Chicago LawReview, 58: 1-54.

Murphy, K., 1998. Executive compensation. In Orley Ashenfelter and David Card (eds.),Handbook of Labor Economics, Vol. 3, North Holland (1999).

35

Nowak, E., 2001. Recent developments in German capital markets and corporate governance.Journal of Applied Corporate Finance, 14: 35-48.

OECD, 1995. Organization for Economic Cooperation and Development Economic Surveys:Germany. OECD: Paris.

Prowse, S., 1994. Corporate governance in an international perspective: A survey of corporatecontrol mechanisms among large firms in the United States, the United Kingdom, Japan, andGermany. Bank for International Settlements Economic Papers No. 41, July.

Ramseyer, J. M. 1998. Corporate law, in The new Palgrave dictionary of economics and law,Peter Newman (ed.), New York: Stockton Press, 503-511.

Roe, M., 1993. Some differences in corporate structure in Germany, Japan and the United States.Yale Law Journal, 102: 1927-2003.

Roe, M., 1994. German ‘populism’ and the large public corporation. International Review ofLaw and Economics, 14: 187-202.

Roe, M., 1996. Chaos and evolution in law and economics. Harvard Law Review, 109: 641-668.

Romano, R., 1993. A cautionary note on drawing lessons from comparative corporate law, YaleLaw Journal, 102: 2021-2046.

Shleifer, A., and R. Vishny, 1997. A survey of corporate governance. Journal of Finance. 52:737-83.

Sundaram, A., and A. Inkpen, 2003. The corporate objective revisited. Organization Science(forthcoming).

Tuschke, A., and G. Sanders, 2003. Antecedents and consequences of corporate governancereform: the case of Germany. Strategic Management Journal, 24: 631-649.

Vagts, D. 1966. Reforming the "modern" corporation: Perspectives from the German. HarvardLaw Review, 80: 23-59.

Wagstyl, S., 1996. Crumbs from the table. Financial Times, September 25.

Zajac, E. J., and P. C. Fiss, 2001. The unusual diffusion of a new corporate ideology: The case ofshareholder value orientation in Germany. Northwestern University Kellogg Graduate School ofManagement Working Paper.

36

Table 1

The Use of ‘Shareholder Value’ and ‘Aktionaerswert’ inGerman Language Media: 1980 – 2001

––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Year(s) Number of Mentions of:

‘Shareholder Value’ ‘Aktionaerswert’––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– 1/1980 – 12/1989 0 0

1990 0 01991 1 01992 2 01993 2 01994 11 01995 182 21996 752 11997 907 61998 1120 61999 1208 82000 1474 152001 962 3

Total 6621 41–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––Source: Lexis-Nexis (based on 33 German language publications covered in this database;includes 6 German language publications from Switzerland and one from Austria).

37

Table 2

The Changing Attitudes of German Corporate ManagersToward Shareholder Value

Number of firms that…… 1990 1995 1999

1. Make no mention of shareholders 12(55%)

5(19%)

2(7%)

2. Address shareholders 10(45%)

6(23%)

0(0%)

3. Mention increase in firm value in both German and English

0 7(27%)

19(62%)

4. Mention increase in shareholder value (English) Mention increase in firm value (German)

0 1(4%)

4(13%)

5. Mention increase in shareholder value (English) Mention increase in shareholder value (German)

0 7(27%)

5(17%)

Exhibit Shareholder or Firm Value Orientation in Annual Reports (3+4+5)

0(0%)

15(60%)

28(93%)

Data from the Annual Reports of the Current DAX 30.

38

Table 3

Increased Focus on Stock Price

Number of having adopted …… 1990 1995 1999

Employee stock purchase plan 836%

12(46%)

19(63%)

Stock-based compensation plan 0(0%)

7(27%)

22(73%)

Stock-option plan 0 1(5%)

14(47%)

ADR listing on the NYSE 0 3(15%)

15(50%)

Data from the Annual Reports of the Current DAX 30.

39

Table 4

DAX Sample

Name SV Score In Sample

Bayer 11 XBayer.Hypo-Und-Vbk. 10 XCommerzbank 10 XDegussa 9 XDeutsche Bank 9 XDeutsche Telekom 9Dresdner Bank 9 XSAP 9Allianz 8 XBASF 8 XDaimlerchrysler 8 XFresenius Med.Care 8Henkel 8Munch.Ruck.Regd. 8 XSiemens 8 X

E On 7 XEpcos 7Man 7 XSchering 7 XThyssenkrupp 7 XAdidas-Salomon 6Lufthansa 6 XPreussag 6 XRWE 6 XInfineon Technologies 5Metro 5Volkswagen 4 XLinde 3 XBMW 2 XKarstadt Quelle 0 X

The shareholder orientation score calculation for each firm is given in the Appendix. The “InSample” is used to indicate the firms used in the subsequent portfolio tests. Firms with scores of8 or more are put into the MSO (More Shareholder Oriented) portfolio and those with scores of 7or less are put into the LSO (Less Shareholder Oriented) portfolio.

40

Table 5The Hoepner Sample*

*Hoepner’s paper discusses the shareholder value orientation of the 40 largest listed non-financial German companies. Hoepner examined three dimensions of shareholder value: hedefines these as the ‘communicative’ dimension, the ‘operative’ dimension and managerialcompensation. Hoepner measures the extent of shareholder orientation of the companies in hissample by deriving a shareholder value score (hereafter called the ‘Hoepner Index’) intended tocapture these three dimensions, based on factors such as the clarity of the annual report, theexistence of an investor relations department, non-fixed top-management compensation, and theimplementation of shareholder-oriented profitability metrics.

The major finding of the Hoepner study is a positive and significant relation between his indexand the number of shares held by institutional investors.

Firm Score Sample Firm Score SampleBayer 1.61 X Deutsche Telekom 0.16VEBA 1.48 Krupp 0.16SAP 1.33 Bunderus 0.04Hoechst 1.20 X Agiv 0.00 XBASF 1.14 X Beiersdort -0.17 XVodafone 1.11 X Volkswagen -0.26 XHenkel 1.09 Rheinmetall -0.31 XDaimler-Chrysler 1.02 X BMW -0.43 XRWE 0.90 X VEW -0.46 XSiemens 0.86 X Metro -0.70Schering 0.74 X AVA -0.81Metallgesellschaft 0.72 X Babcock -1.08 XDegussa 0.55 X Deutz -1.18 XViag 0.55 Karstadt -1.23 XPreussag 0.45 X Bilfinger+Berger -1.25 XMAN 0.36 X Spar -1.28Lufthansa 0.28 X Sudzucker -1.30 XLinde 0.22 X Axel Sp[ringer Veriag -1.70 XContinental 0.21 X Holzmann -1.90 XThyssen 0.17 X Strabag -2.29 X

41

Table 6

Results of a Seemingly Unrelated Regression of the Weekly Return to the DAXComponents on the Weekly Returns to the Standard & Poors 500 Industrials

The system of equations is:

Equation 1: MSOt = A1,1 D1 + A1,2 D2 + A1,3 D3 + B1,1 D1 S&Pt + B1,2 D2 S&Pt + B1,3 D3 S&Pt + e1,t

Equation 2: LSVt = A2,1 D1 + A2,2 D2 + A2,3 D3 + B2,1 D1 S&Pt + B2,2 D2 S&Pt + B2,3 D3 S&Pt + e2,t

MSOt is the weekly return to the portfolio of shares of the Dax firms that are more shareholder oriented. LSOt is the weekly return to the portfolio of shares of the Dax firms that are less shareholder oriented.

The dependent variables in the system are the weekly returns to the indicated subset of DAX firms from 1980 to 2002.

The independent variables in the system are a matrix of dummy variables with elements equal to one over the indicated time period and zero otherwise, and a matrix of interaction terms with elements equal to the product of the return to the the S&P and the dummy variable for the indicated period.

MSO (11 Firms) LSO (11 Firms) MSO - LSO

NOBS Alpha Beta Alpha Beta Alpha Beta

Period 1 1980-1989 521 0.0020 0.3021 0.0021 0.2780 -0.0001 0.02411.85 7.90 2.01 7.47 0.17 -1.00

Period 2 1990-1994 261 0.0005 0.3727 0.0007 0.4046 -0.0003 -0.03190.30 5.74 0.49 6.41 0.28 0.78

Period 3 1995-2002 381 0.0008 0.6540 0.0007 0.5519 0.0001 0.10210.60 14.75 0.53 12.80 0.00 3.67

Differences in Periods

Period 2 - Period 1 -0.0016 0.0706 -0.0014 0.1266-0.83 0.94 -1.08 1.73

Period 3 - Period 2 0.0003 0.2813 -0.0001 0.14730.14 3.58 0.00 1.93

Period 3 - Period 1 -0.0013 0.3519 -0.0015 0.2739-0.06 6.01 0.29 4.81

42

Table 7

Results of a Seemingly Unrelated Regression of the Weekly Return to the Shares of theLargest Forty Non-Financial German Firms on the Weekly Returns to the S&P 500

The system of equations is:

Equation 1: MSOt = A1,1 D1 + A1,2 D2 + A1,3 D3 + B1,1 D1 S&Pt + B1,2 D2 S&Pt + B1,3 D3 S&Pt + e1,t

Equation 2: LSOt = A2,1 D1 + A2,2 D2 + A2,3 D3 + B2,1 D1 S&Pt + B2,2 D2 S&Pt + B2,3 D3 S&Pt + e2,t

MSOt is the weekly return to the portfolio of shares of the Dax firms that are shareholder oriented. LSOt is the weekly return to the portfolio of shares of the Dax firms that are less shareholder oriented.

The dependent variables in the system are the weekly returns to the indicated subset of DAX firms from 1980 to 2002.

The independent variables in the system are a matrix of dummy variables with elements equal to one over the indicated time period and zero otherwise, and a matrix of interaction terms with elements equal to the product of the return to the the S&P and the dummy variable for the indicated period.

MSO (16 Firms) LSO (12 Firms) MSO-LSO

NOBS Alpha Beta Alpha Beta Alpha Beta

Period 1 1980-1989 521 0.0021 0.2862 0.0019 0.2377 0.0002 0.04852.12 8.33 1.97 7.06 0.24 1.92

Period 2 1990-1994 261 0.0001 0.4328 0.0006 0.3751 -0.0005 0.05780.08 7.42 0.42 6.56 -0.45 1.35

Period 3 1995-2002 381 0.0009 0.5811 -0.0009 0.4265 0.0018 0.15460.78 14.59 -0.82 10.93 2.15 5.27

Differences in Periods

Period 2 - Period 1 -0.0020 0.1467 -0.0013 0.13741.16 2.17 -0.80 2.07

Period 3 - Period 2 0.0008 0.1483 -0.0015 0.05140.44 2.10 -0.85 0.74

Period 3 - Period 1 -0.0012 0.2950 -0.0028 0.1888-0.79 5.61 -1.91 3.66

43

Table 8

Results of a Seemingly Unrelated Regression of the Weekly Return to the Shares of theLargest Forty Non-Financial German Firms Excluding Firms with ADRs on the

Weekly Returns to the S&P 500(1980-2002)

The system of equations is:

Equation 1: MSOt = A1,1 D1 + A1,2 D2 + A1,3 D3 + B1,1 D1 S&Pt + B1,2 D2 S&Pt + B1,3 D3 S&Pt + e1,t

Equation 2: LSOt = A2,1 D1 + A2,2 D2 + A2,3 D3 + B2,1 D1 S&Pt + B2,2 D2 S&Pt + B2,3 D3 S&Pt + e2,t

MSOt is the weekly return to the portfolio of shares of the firms that are more shareholder oriented. LSOt is the weekly return to the portfolio of shares of the firms that are less shareholder oriented.

The dependent variables in the system are the weekly returns to the indicated subset of DAX firms from 1980 to 2002.

The independent variables in the system are a matrix of dummy variables with elements equal to one over the indicated time period and zero otherwise, and a matrix of interaction terms with elements equal to the product of the return to the the S&P and the dummy variable for the indicated period.

MSO (12 Firms) LSO (12 Firms) MSO - LSO

NOBS Alpha Beta Alpha Beta Alpha Beta

Period 1 1980-1989 521 0.0022 0.2835 0.0019 0.2377 0.0003 0.04582.09 7.66 1.97 7.06 0.41 1.70

Period 2 1990-1994 261 0.0000 0.4314 0.0006 0.3751 -0.0005 0.05630.03 6.86 0.42 6.56 0.48 1.23

Period 3 1995-2002 381 0.0006 0.5823 -0.0009 0.4265 0.0015 0.15580.50 13.57 -0.82 10.93 1.72 5.00

Differences in Periods

Period 2 - Period 1 -0.0022 0.1479 -0.0013 0.13741.19 2.03 0.80 2.07

Period 3 - Period 2 0.0006 0.1509 -0.0015 0.05140.30 1.98 0.85 0.74

Period 3 - Period 1 -0.0016 0.2988 -0.0028 0.18880.98 5.27 -1.91 3.66

44

Table 9

Weekly Standard Deviation of Returns

DAX Sample Hoepner Sample

PERIOD NOBS DAX S&P MSO LSO MSO LSO

80-89 521 2.42% 2.86% 2.65% 2.44% 2.42% 2.08%

90-94 361 2.67% 2.39% 2.61% 2.79% 2.89% 2.47%

95-02 384 3.15% 2.88% 3.14% 2.91% 2.91% 2.70%

95-98 209 2.94% 2.64% 3.01% 2.87% 2.97% 2.61%

98-02 175 3.38% 3.13% 3.29% 2.97% 2.84% 2.82%

45

Table 10

Results of a Seemingly Unrelated Time-Series Regression of the AnnualRatio of the Market Value of Equity plus the Book Value of

Debt to the Book Value of Assets(1980-2002)

MSO LSO Difference

DAX Sample Constant 1.33 1.34 -0.01t-Statistic 4.42 4.50 0.44

Slope 0.34 0.23 0.11t-Statistic 6.98 4.82 4.22

Sample Size 12 11Hoepner Sample

Constant 1.42 1.95 -0.53t-Statistic 5.44 6.66 1.20

Slope 0.30 0.09 0.21t-Statistic 7.11 1.92 2.88

Sample Size 13 16

46

Table 11

Annual Growth Rate in Market Capitalization(1990 – 2002)

MSO LSO Difference

DAX Sample Annual Growth in Market Cap 26.3% 15.0% 11.3%t-Statistic 3.05

Hoepner Sample Annual Growth in Market Cap 18.4% 17.4% 1.0%t-Statistic 0.08

47

Table 12

The Role of Divestitures in Corporate Control Transactions: A Comparison ofGerman, DAX, and Non-German EU Firms (1991 – 2000)

All German DAX Non-German EU Firms Firms Firms

Total # Corporate 1991-95 5495 361 24945 Control Transac’ns1 1996-00 7305 685 32651

1991-00 12800 1046 57596

# Divestitures 1991-95 2265 208 8715 1996-00 2238 437 10553 1991-00 4503 645 19268

# Cross-border 1991-95 325 114 1390 Divestitures2 1996-00 395 224 1694

1991-00 720 338 3084

Divestitures as % 1991-95 43.6% 57.6% 35.1% all Deals 1996-00 30.5% 63.8% 32.9%

1991-00 35.2% 61.7% 33.5%

Cross-border 1991-95 14.5% 30.3% 16.0% Divestitures as % 1996-00 17.9% 30.7% 16.1% all Divestitures 1991-00 16.0% 30.5% 16.0%

Source: SDC database.1 Only ‘completed’ transactions2 Divestitures to buyers outside the European Union (EU).

t-statistics for Test of Proportions: 1991-2000

Divestitures as % All Deals Cross-border Divestitures as % All Divestitures

All Non-German All Non-German German DAX EU German DAX EU

All German --- –6.29** 1.03 --- –2.20* 0.00DAX 6.29** --- 7.04** 2.20* --- 2.53**Non-German-EU –1.03 –7.04** --- 0.00 –2.53** ---

* Significant at 5%** Significant at 1%

Appendix: Construction of the DAX Shareholder Orientation Index

Total points SV

TOTAL

Adidas-Salomon nothing 0 SV in German 1 SV in German 1 2 IAS 1996 3upper/middle management

19991 6 Adidas-Salomon

Allianz shareholders mentioned 0 increase in value 1 SV in English 1 2 Incentive plan , EVA 1999 3IAS , US GAAP, ADR

20002 1999 1 8 Allianz

BASF nothing 0 shareholders mentioned 0 increase in value 1 1 Incentive share program 1999 3 US GAAP, ADR 1999 2senior

executives 19991 1989 1 8 BASF

Bayer shareholders mentioned 0 SV in German 1 SV in German 1 2Top management CFROI

19954 IAS 1995, ADR 1999 3

Management 1999

1 1990 1 11 Bayer

BMW nothing 0 increase in value 1 nothing 0 1 1990 1 2 BMW

Commerzbank shareholders mentioned 0 increase in value 1 increase in value 1 2Incentive plan executives

19993 ADR 1989, IAS 1999 4 1995 1 10 Commerzbank

DaimlerChrysler NA 0 nothing 0 increase in value 1 1stock appreciation rights

20003

US GAAP, US listing 1999

2employees

19962 1999 1 9 DaimlerChrysler

Degussa-Huels shareholders mentioned 0 shareholders mentioned 0 increase in value 1 1 profit-sharing scheme 1995 4 US GAAP 1999 2 1990 1 8 Degussa-Huels

Deutsche Bank shareholders mentioned 0 shareholders mentioned 0 increase in value 1 1performance-oriented 1996, linked to shareprice 1999

4 IAS 1996 3 1999 1 9 Deutsche Bank

Dresdner Bank nothing 0 nothing 0 increase in value 1 1 Long term incentive plan 1999 3 IAS in part, ADR 1995 3executives and

staff 19991 1995 1 9 Dresdner Bank

Deutsche Telekom NA 0 nothing 0 SV in German 1 1group-wide market oriented

remuneration 19993

US GAAP and listing 1996

3top

management 1999

1 1999 1 9 Deutsche Telekom

E.ON NA 0 NA 0 increase in value 1 1performance-based for

managers 20003 US GAAP, ADR 2000 2 2000 1 7 E.ON

EPCOS NA 0 NA 0 SV in German 1 1EVA targets for management,

annual bonus 19993 US GAAP, ADR 1999 2

Management 1999

1 7 EPCOS

Fresenius Medical Care

shareholders mentioned 0 shareholders mentioned 0 increase in value 1 1profit sharing scheme 1995, management linked to share

price 19994

ADR 1996, US GAAP 2001

3 8Fresenius Medical Care

Henkel nothing 0 shareholders mentioned 0 SV in English 1 1stock incentive program for

900 executives 19993 IAS 1995, ADR 1999 3

900 executives 1999

1 8 Henkel

HypoVereinsbank NA 0 SV in German 1 SV in German 1 2 performance-oriented 1995 4 IAS 1995 3 1999 1 10 HypoVereinsbank

Infineon NA 0 NA 0 increase in value 1 1 US GAAP, ADR 2000 2Management

20001 2000 1 5 Infineon

KarstadtQuelle nothing 0 nothing 0 nothing 0 0 0 KarstadtQuelle

Lufthansa shareholders mentioned 0 SV in English 1 increase in value 1 2profit-sharing scheme,

Chance program for 1000 executives 1999

3 IAS 1999 1988 1 6 Lufthansa

Linde nothing 0 increase in value 1 increase in value 1 2 1990 1 3 Linde

MAN nothing 0 increase in value 1 increase in value 1 2Rate of return driven, profit-

participation plans 19993 IAS 1999 2 7 MAN

Metro NA 0 shareholders mentioned 0 SV in English 1 1EVA targets for management

19993

management 1999

1 5 Metro

Munich Re nothing 0 increase in value 1 increase in value 1 2performance-related 1995, bonus payment linked to

stockprice 19994 IAS 1999 2 8 Munich Re

Preussag shareholders mentioned 0 SV in German 1 increase in value 1 2 IAS 1999 2Option rights

exercised 19991 1995 1 6 Preussag

RWE shareholders mentioned 0 SV in German 1 SV in English 1 2 IAS 1999 2Senior

management 1999

1 1990 1 6 RWE

SAP nothing 0 SV in German 1 increase in value 1 2performance-related bonus

program 1995, Stock appreciation rights 1999

4 US GAAP, ADR 1998 2 1990 1 9 SAP

Schering nothing 0 nothing 0 increase in value 1 1Bonus system for senior

management, incentive plan 1999

3 IAS 1995, ADR 1999 3 7 Schering

Siemens shareholders mentioned 0 SV in German 1 increase in value 1 2compensation tied to

performance 1996, EVA 19994

US GAAP 1999, Listing 2001

2 8 Siemens

ThyssenKrupp NA 0 NA 0 increase in value 1 1Incentive Plan, EVA Management 1999

3 US GAAP 1999 2management

19991 7 ThyssenKrupp

Volkswagen nothing 0 increase in value 1 increase in value 1 2Management and workforce

19991 1995 1 4 Volkswagen

Employee stock

purchase

Shareholder Value (SV) Mention

1990 1995 1999Stock-based compensation

planADR, US GAAP or US

listingStock options plan