Doctoral Thesis Institutional Investors as Perfect Monitors:

(1)

Doctoral Thesis

Institutional Investors as Perfect

Monitors:

The theory behind institutional investor activism in

the Netherlands

August 17, 2007 RijksUniversiteit Groningen Faculty of Economics Thesis Supervisor Drs. M.E. Helmantel Co-Assessor Prof.dr. F.M. Tempelaar Author Onno Meijering 1257854 O.W.Meijering@student.rug.nl JEL Classification: G20, G23, G30.

Key words: Corporate Governance; Shareholder Activism; Institutional Investors; Pension

(2)

Preface

In front of you lies the doctoral thesis I have written for completing my Finance study at the Faculty of Economics, Rijksuniversiteit Groningen. Over the past six months I have dug into the world of institutional investor involvement. In doing so, I have depicted an in-depth image of the theory behind institutional investor activism in the Netherlands.

Writing a thesis comes with many pitfalls. Although I was convinced I would not fall into one of these clichés, I have succeeded in falling in every single one of them. Nonetheless, the process has proven to be a very intensive, challenging, and above all very instructive one. The result is this thesis, which I am proud to present to you.

Although writing a thesis is an individual task, its realization is not done without the help of others. Consequently, there are a couple of people I am greatly indebted to. First of all, I would like to thank my thesis supervisor Mark Helmantel for his expertise and for keeping me sharp and on the right track when it was absolutely needed. Furthermore, I would like to thank my thesis writing colleagues Ferdinand, Marcel, Martijn en Rutger for their support encompassing many enjoyable coffee and lunch breaks at the University Library. I cannot imagine me finishing this thesis in time without them, let alone with so much fun. Subsequently, I would like to thank the weather gods for disposing me of any summerly distraction.

But most importantly, I would like to thank my parents for supporting me in every way possible, not only over the past six months, but during the entire six years this study has taken me.

This thesis marks the end to my student days in Groningen. With melancholy I will look back to these beautiful six years. My time at “Huize JVG”, as well as the many trips to Groningen centre have brought me both friends and memories for life.

I hope the reader will enjoy reading this thesis, Onno Meijering

(3)

Executive Summary

Corporate governance has received a great deal of attention since the beginning of the 21st century. Numerous scandals in various developed countries have drawn renewed attention to the way in which firms operate in the interest of their stakeholders. Corporate governance problems arise from the infamous separation of ownership and control (Fama and Jensen, 1983a), which results in an agency problem (Jensen and Meckling, 1976), in which the objectives of utility maximizing shareholders and management differ, thereby harming shareholder value through agency costs.

One solution put forward by the literature to (limit) the agency problem is the presence of a large investor (Shleifer and Vishny, 1997). In this respect, special attention is directed to institutional investors. In most corporate governance codes, they are strongly encouraged to take the lead in active monitoring. In doing so, institutional investors are appointed as ultimate guardians of long term shareholder interests. This thesis discusses the implications of this proposition, while focusing on the Netherlands.

First, the ideal owner concept is defined. After that, the various institutional investors are tested along this ideal owner framework. From this, it comes out that institutional investor categories differ significantly in their incentives to monitor. Of all institutional investors, pension funds come closest to the ideal monitoring picture. However, when conducting an empirical survey into the ownership structure of the Dutch market, it turns out that pension funds lack the significant stakes deemed necessary to make a difference. Banks, insurance companies and mutual funds, on the other hand, are the dominant players in the market. Simply stated, it seems that pension funds have the incentives, but lack the stakes, and banks and mutual funds have the stakes, but lack the incentives.

In addition, it is demonstrated that although Dutch pension funds extensively describe their corporate governance policies and voting behaviour, there may not be much substance behind this. Some activities undertaken, as well as the rather passive attitude towards activism are indications that point in this direction.

Two examples from the US and the UK show what “active activism” should look like. CalPERS and Hermes also lack the muscle arising from significant stakes, but they attempt to move troubled companies by entering into a dialogue with management, building coalitions with other shareholders and publicly raising their concerns.

(4)

1. Introduction and Problem Statement

Corporate governance has received a great deal of attention since the beginning of the 21st century. Numerous scandals in various developed countries have inspired governments and regulators to draw renewed attention to the way in which firms operate in the interest of their stakeholders. Although the problems seem to be a modern phenomenon, they have been with us for a long time. Ironically, even the world’s first ever corporation, the Dutch VOC, experienced governance problems some 400 years ago (Cools, 2005). It appears to be that wherever the infamous separation of ownership and control (Fama and Jensen, 1983a) exists, governance problems will be present. More recently, this has been manifested in a couple of major scandals, destroying billions of shareholder value and leaving thousands of people unemployed and disposed of their pensions. More importantly, these scandals devastated people’s trust in public corporations. It was through the sheer size and social consequences of the recent scandals that governments and regulators felt obliged to act. This resulted in stricter legislations and corporate governance codes being introduced in most of the developed countries. The most prominent of these were the 2002 Sarbanes-Oxley legislation in the US and the 2003 Combined Code in the UK. In the EU, the European Commission issued a directive in 2003, requiring every listed company in all EU member states to publish the national code to which it subscribes (Solomon and Solomon, 2004). At the moment, most of the countries have a code installed. In the Netherlands, a corporate governance code (Dutch Corporate Governance Code) was installed at the end of 2003.

What most of these codes have in common is that they put a lot of emphasis on shareholder power and the monitoring role of shareholders. Firms are obliged to provide shareholders more information on their operations, and shareholders are given more discretion to intervene with the firms’ strategy and decisions. Put more generally, the control of the firm should be given back to its ultimate owners.

(6)

given the role of ultimate guardians of the long term interests of the firm, and thereby its share- or stakeholders.

This proposition is at least disputable. First, while the academic literature recognizes the above mentioned advantages of concentrated ownership and institutional investor involvement, it also depicts some severe potential disadvantages (Shleifer and Vishny, 1997; Gillan and Starks, 2003; Solomon and Solomon, 2004). Furthermore, it is incorrect to generalize all varieties of shareholders. Shareholders, and particularly institutional investors, come in many categories, with each category having its typical characteristics. As I will demonstrate, it is rather ingenuous to expect all these categories to have equivalent, corresponding structures, resources and incentives to perform the “legendary monitoring role” (Monks and Minow, 2004), attached to them by most modern literature and governance codes. These characteristics should be taken into consideration when discussing institutional shareholder monitoring.

Indeed, the Dutch market has experienced a flip side of the coin recently. Over the past couple of years, firms like Vendex KBB, VNU, Stork and, very recently, ABN Amro have been subjected to misconceived shareholder activism. More shareholder rights and less takeover protection invited hedge funds and private equity firms to take control of firms and impose their judgment on how the firm should be managed. This happened often to the anger of the incumbent management and employees of the firm.

In sum, handing over more power to shareholders is not straightforwardly advantageous as predicted by the literature and anticipated by regulators. This leads me to question the validity of the propositions in most governance codes of more shareholder power and protection, and particularly the idealized monitoring picture of large, institutional investors. Is this really a solution to the agency problems arising from the separation of ownership and control, and therefore a blessing for the corporate environment? Does a concept like the “ideal owner” (Monks and Minow, 2004) exist in practice? I will try to answer these questions by reviewing the literature on the subject, by examining the different features of the various institutional investors and by performing an empirical survey in the Dutch market. The focus will be on institutional investors, in order to check their ability to perform the “legendary monitoring role”. In doing so, I will depict a complete picture of the background of institutional investor involvement.

(7)

2. Literature Review and Theoretical Background

In this section, the literature on concentrated ownership and institutional investor monitoring will be discussed. In doing so, I will provide a thorough insight in the theory behind shareholder activism and institutional investor involvement.

First, I will give a description of the origin of corporate governance problems. More specifically, I will discuss the separation of ownership and control and the agency theory. Hereafter, the concept of the ideal owner will be introduced in section 2.2. In section 2.3, the boundaries of shareholder involvement will be discussed. Subsequently, in section 2.4, the pros and cons of concentrated ownership in general will be discussed, while in sections 2.5 and 2.6 the scope will be narrowed down to institutional investors. Finally, the empirical literature on concentrated ownership and firm performance will be discussed in section 2.7.

2.1 Agency Theory and the Separation of Ownership and Control

Corporate governance problems are an unintended consequence of the economic system the Western world has developed over the past four centuries. It deals with the initially plain concept of ownership, as shareholders are generally referred to as the “owners” of a company. The first ever companies were financed by the owners themselves or by a few rich individuals. However, as companies grew both in quantity and dimension, traditional financing proved to be insufficient. This induced developments of the stock market, which were aimed at improving transferability and liquidity, in order to make the stock market accessible for a broader audience. However, these developments made it harder to exercise classic ownership rights, as perceived by the basic (legal) definition of ownership.

(8)

corporate governance struggles, and consequently has dominated the corporate governance literature: the separation of ownership and control.

Although the actual term “separation of ownership and control” was first introduced by Fama and Jensen (1983a), the issues involved were initially mentioned in the classic Berle and Means (1932) book, which depicted a corporate environment in which the ‘true’ owners of a company, the shareholders, had little influence over company management and were rendered impotent by the wide dispersion of ownership as described above. Berle and Means (1932) named this phenomenon the “dissolution of the atom of property”.

This situation left shareholders with only one option when dissatisfied with the companies’ behaviour; sell the shares. This phenomenon has been labeled the preference of “exit” over the more proactive approach of “voice” (Solomon and Solomon, 2004).

Based on the foregoing observations, Fama and Jensen (1983a,b) definitely defined the concept of separation of ownership and control. In their definition, decision management (initiation and implementation) is separated from decision control (ratification and monitoring), as, in the modern corporate environment, residual risk bearing is separated from decision management. They go on to conclude that, in order for a corporation to survive, contracts have to be designed in such a way that the decision control lays with the risk bearing agents, in this case the providers of capital. Stated in simpler terms, the eventual control of corporations must lie with the shareholders, as they bear the ultimate risk arising from the decisions made by the management.

The separation of ownership and control, amplified by dispersed ownership,has created the infamous “agency problem”, first explored by Ross (1973), but extensively described in a theoretical way by Jensen and Meckling (1976). The agency theory defines an agency relationship as a contract under which one or more persons (the principal(s)) engage another person (the agent) to perform some service on their behalf which involves delegating some decision making authority to the agent (Jensen and Meckling, 1976). Obviously, this definition fits the relation between shareholders and management like a glove. The core of the agency problem is that, when both principal and agent are utility maximizers, there is good reason to believe that both parties’ interests do not correspond to each other, and consequently, due to information asymmetries, the agent is very likely to make decisions that are not in the best interest of the principal. Indeed, the critical assumption of the agency theory is that the individual goals of the principal and agent conflict.

(9)

to more fundamental problems like pursuing pet projects, short-termism, empire building and management entrenchment (Shleifer and Vishny, 1997; Demsetz, 1983).

The agency problems as defined above lead to a significant reduction in shareholder welfare, referred to as agency costs. According to Jensen and Meckling (1976), agency costs are made up of three components. First, there are monitoring costs, beared by the principal while attempting to monitor the agent (i.e. management). Monitoring by shareholders is expensive, as it involves time-consuming activism and sophisticated reward systems, drawn up in expensive contracts. Second, there are bonding costs, which arise from the agent attempting to convince the principal that he is accountable and is pursuing the principal’s objectives. In our particular case, these costs may involve providing extra (sensitive) information about the firm and arranging meetings with the shareholders. The final category of agency costs consists of the dollar equivalent of the reduction in welfare arising from the decisions made by the agent which are, despite the optimal monitoring and bonding activities conducted, not value-maximizing for the principal. These costs are generally referred to as residual loss (Jensen and Meckling, 1976).

The above discussion presents the substantial problems arising from the separation of ownership and control. How can these agency problems be solved, or at least controlled? The classic approach offered by the literature is that of effective contracting, in which contracts are designed in such a way that managers’ interest (i.e. remuneration) is aligned with that of the shareholders (Coase, 1937; Jensen and Meckling, 1976; Fama and Jensen, 1983a,b). Although this sounds very straightforward, in practice the situation is more complicated. Shleifer and Vishny (1997) document two problems with respect to assessing the compliance of the contracts. First, they argue that, because of interpretation difficulties, contracts are very hard to enforce by outside courts. Second, and more relevant for this study, investors are often too small and dispersed to put effort and resources in exercising their control rights and monitoring role. This gives rise to the notorious free rider or collective choice problem (Grossman and Hart, 1980), in which it is uneconomical for a single shareholder to dedicate effort and resources to monitoring, as other shareholders will profit from this effort for free. This phenomenon is also called “rational ignorance” (Monks and Minow, 2004).

(10)

exercising the classic ownership rights. The rationale behind this perception, as well as its pros and cons, will be discussed in section 2.3.

(11)

2.2 The Ideal Owner Concept

From the preceding discussion it can be concluded that the theory assigns a decisive role to (large) shareholders in creating and maintaining a healthy corporation, and consequently a healthy corporate environment. An intriguing concept in this respect is that of the “ideal owner”. Monks and Minow (2004) introduce this concept, stating that this must be someone who has the “information, ability, and alignment of interest with other corporate constituencies to provide the optimal level of monitoring” (Monks and Minow, 2004, p. 181). Translated in more practical terms, this implies an owner with a sufficiently large stake, a long term perspective and no other interest in the firm except for his ownership.

The large stake will provide the shareholder with both the incentives and power to extensively monitor the investee company. In addition, a large investor can be expected to possess the resources required for monitoring. The long term perspective preserves that the investors’ interest is in line with the long term interest of the firm, as well as that of other stakeholders. The final characteristic is necessary to preserve independence.

(12)

2.3 The Boundaries of Shareholder Involvement

Before continuing to the concentrated ownership and institutional investor discussions, there are two general issues that need to be addressed when discussing matters like more shareholder involvement, relationship investing and increasing the dialogue between (institutional) shareholders and management. These are limited liability and the boundaries of shareholder involvement.

When looking for the ideal owner, it has to be kept in mind that there are certain limits to shareholder involvement. According to Monks and Minow (2004), shareholders do not have the expertise, resources, or the right to get involved in matters of day-to-day management, and should not become involved in second-guessing ordinary business. This is the essence of the contract between shareholders and their investee companies, which implies that in exchange for limited liability shareholders will have a more limited scope of authority and a restricted agenda. It is also consistent with the “business judgment rule” for directors apparent in most western legal systems.

Limited liability is the principal element of today’s stock market (Solomon and Solomon, 2004). It states that investors are only liable for the amount they invest or have invested. This ensures that corporations are able to take significant risks without threatening the personal resources of their owners. Although the concept is in principle applied only to financial issues, it can be extended to a broader field. This is where the legal boundaries of shareholder interference come around. In principle, investors are not liable for any activities undertaken by management. But what happens if a certain policy is co-designed by active shareholders and fails, with, for example, huge societal costs? This could make shareholders vulnerable to lawsuits, especially in legal-action-prone countries like the US and, to a lesser extent, the UK. This would significantly shift the limited liability border, and it is of course highly questionable whether any investor is willing to take this risk, or whether this is a desirable situation in the first place.

Another legal boundary is that of inside information. As described earlier, when a firm enters into a detailed dialogue with its core shareholders, the risk of providing price-sensitive inside information is substantial. Therefore, this dialogue should be very cautiously managed. This implies explicitly setting the appropriate and, perhaps even more important, inappropriate issues for shareholder involvement.

(13)

2.4 Concentrated Ownership as Agency Problems Solution

As mentioned earlier, ownership concentration has been acknowledged by the literature as an important mechanism to control agency problems and improve investor protection. There are several reasons why this could be the case. First, a substantial minority investor has the incentive to collect information and control management. This results from the fact that his stake is generally too large to instantly sell in the market, if the large investor is dissatisfied with the firm’s policies. Furthermore, because of the large stake, the investor will be able to earn back its resources spent on monitoring, as the resulting absolute return will be more significant. This way, the “voice” approach is vindicated and the free rider problem avoided. Second, a large investor has enough voting power to put pressure on management in some cases, or even to oust them through a proxy fight or a takeover. This latter mechanism is studied by Shleifer and Vishny (1986), who, among other things, find that the larger the stake of the large minority shareholder, the greater the chance of a takeover, when the shareholder is dissatisfied with the firm’s performance. This could of course be due to a share build up as part of an takeover intention. Consequently, Shleifer and Vishny (1986) see takeover activity not simply as a market activity, but also as the execution of power by large shareholders in order to solve agency problems. Shleifer and Vishny (1997) further enhance this view when they state that “Takeovers can…be viewed as rapid-fire mechanisms for ownership concentration” (Shleifer and Vishny, 1997, p. 756).

The third and final reason for concentrated ownership to be acknowledged as a solution to agency problems is that both the magnitude and the illiquidity of a large investor’s stake is very likely to provide him with a long term perspective (Solomon and Solomon, 2004). Theoretically, this is particularly the case for institutional investors, as will be discussed later on.

Despite the theoretical benefits, concentrated ownership may come at a cost as well. An obvious cost is excessive risk bearing by underdiversified investors (Demsetz and Lehn, 1985). However, the demonstrated concentrated ownership structures around the world found by LaPorta et al. (1999) suggest that underdiversification is not perceived by large investors as much of a cost as the relinquishment of control.

(14)

expropriation is more likely and potentially more significant. First, straightforward expropriation of other shareholders, as well employees and management, is likely to occur when the large shareholder has superior voting rights and/or control over the firm through a pyramid structure. This might, for example, induce him to pay special dividends only to himself. The fact that shares with superior voting rights trade at a large premium might be considered as evidence of significant private benefits of control that come at the expense of other minority shareholders (Shleifer and Vishny, 1997).

Second, expropriation becomes potentially more significant when the other investors have different patterns of cash flow claims in the company. For example, if the large investor is an equity holder, he might use his dominant stake to force management to undertake high risk investments, since he shares in the upside, whereas the other investors (especially creditors) bear the costs of failure. Related to this problem is the situation in which different equity holders have different investment objectives, particularly with respect to the time horizon. Recently, this problem has manifested itself in the Netherlands, as hedge funds and, to a lesser extent, private equity firms invested in large corporations and dictated their relatively short term perspective on to the firm and other share- and stakeholders. Although these actions do not necessarily have to be disadvantageous for the market as a whole, it provoked a lot of social unrest and started a social discussion on which this thesis is primarily based.

Besides the underdiversification and expropriation threats, the literature names some other, less materially negative effects of concentrated ownership. These arguments include inhibited information production in the stock market through reduced liquidity (Holmstrom and Tirole, 1993), and reduction of the probability of takeover attempts through aggressive counterbidding by the incumbent large shareholders (Burkart, 1995).

A final argument against ownership concentration, and in favor of dispersed ownership, is made by Burkart et al. (1997). They argue that excessive control resulting from ownership concentration leads to diminished manager discretion, which in turn could lead to less manager initiative. This could result in underinvestment, i.e. managers foregoing positive NPV projects, thereby harming future firm value. In this view, there is a trade-off between the gains from monitoring and those from managerial initiative, that is between control and initiative. Burkart et al. (1997) see a firm’s ownership structure as an ex ante instrument to solve this trade-off.

(15)

point, the large shareholder(s) exercises too much control over the company, uses this control to pursue private benefits, and profitability drops. Stulz (1988) presents this theory in a formal model, giving rise to the so-called roof-shaped relationship between ownership structure and performance.

The above discussion provides an overview of the benefits and costs of concentrated ownership in general. From this discussion it can be concluded that concentrated ownership is not by definition advantageous. Particularly, the expropriation threat is potentially a severe disadvantage, which could harm minority shareholders considerably. Consequently, it pays to take a deeper look into the investors exercising this concentrated ownership. Therefore, the discussion will be narrowed down to a very important group of large shareholders: the institutional investors. But first, two issues not to be neglected when talking about shareholder involvement will be discussed.

2.5 Institutional Investors as Active Investors

When discussing concentrated ownership as solution to governance problems, special attention is most of the time directed to institutional investors. This is no surprise, given the fact that their asset holdings have dramatically increased over the past couple of decades. In all developed countries they represent a significant portion of the equity markets, in some they are even a majority force.

Some authors see institutional investors as the embodiment of the ideal owner (e.g. Solomon and Solomon, 2004; Agrawal and Knoeber, 1996). However, the appropriate role for institutional investors in corporate governance has been the subject of continuing debate (Gillan and Starks, 2003).

However, a fact frequently overlooked is that institutional shareholders come in many categories, with each category having its own structure, incentives and obstacles affecting their ability to perform the ideal owner role. Furthermore, their position in corporate governance is more complicated than one initially would perceive. These features will be discussed in this section. In doing so, the theory behind institutional investor involvement will be illustrated.

(16)

subsection. In addition to this, institutional investors are presumed to have a long term perspective, to be mostly independent of the investee company and can be expected, through the tie with their beneficiaries, to keep a broad stakeholder view of the firm. Indeed, looking at it this way, this appears to be a perfect completion of the ideal owner characteristics. Agrawal and Knoeber (1996) seem convinced when they state that: “concentrated shareholding by institutions can increase managerial monitoring and so improve firm performance” (Agrawal and Knoeber, 1996, p.377)

It is also because of these perceptions that legislators and regulators have particularly relied (and still rely) on institutional investors to promote corporate governance (see for example the Dutch Corporate Governance Code, 2003; and the UK Combined Code, 2003).

A concept that is frequently mentioned with regard to institutional investor involvement is that of “relationship investing” (Chidambaran and John, 1998, 2000). The core of this concept is that management and the institutional investor(s) enter into an extensive dialogue with each other, thereby preserving the long term interests of the firm. Relationship investing is derived from the so-called network oriented governance systems (Moerland, 1995), particularly the German and Japanese ones. Here, it is not unusual for large investors (especially banks) to enter into an extensive dialogue with an investee firm, for example by placing an executive or other representative on the supervisory board of investee companies. This is done not only as a function of ownership, but rather as a function of preserving a valuable commercial relationship (Monks and Minow, 2004). Chidambaran and John (2000) argue that institutional shareholders can transmit private information obtained in this dialogue to other shareholders. They provide formal evidence that, under certain conditions, this might be an optimal situation for both the management and the large investor.

An interesting case in this respect is that of Warren Buffett. He is a prime example of a large investor showing the willingness to enter into an extensive dialogue with investee companies, in particular the troubled ones. In some cases, he even accepted a position on the board. However, this is not done while completely ignoring the free rider problem. In many investments, Berkshire Hathaway (Buffett’s investment vehicle), insists on the purchase of a special class of convertible preferred stock, which guarantees a better return than ordinary common stock. This way, the free rider problem is reduced by charging a fee for Berkshire’s perceived ability, as a shareholder, to add value to the company. Indeed, Monks and Minow (2004) suggest that, following this account, Warren Buffett could well be present day’s closest approximation to the ideal owner.

(17)

First, as we have seen in the previous section,concentrated ownership in general might come with significant disadvantages. Particularly, when talking about preferential treatments of institutional investors, as well as starting an extensive dialogue with them, the expropriation threat clearly stands out. The line between useful additional company information and price-sensitive inside information is a very thin one. Theoretically, it is not difficult to imagine an institutional investor expropriating his preferential treatment and good contacts with management, be it through insider trading or forcing the prioritization of their interests to that of other share- or stakeholders. However, in practice this theory might not be that material. By conducting activities like minority shareholder expropriation, insider trading and imposing their particular interests onto management, institutional investors risk major reputational and legal damage, should these activities come out. It is not very likely that many institutional investors are willing to take such risks.

A second reason why institutional investor monitoring might not be as perfect as perceived, is their own governance. However, in assessing the role of institutional investors in corporate governance, the governance of the institutions themselves has generally been underexposed (Schneider, 2000). What stands out in this respect is their complex ownership structure. By definition, institutional investors are not the real owner of their funds. They act on behalf of their clients, be it pension plan participants, insurance policy holders, trust beneficiaries, mutual fund holders etc. These people entrust their money to a fund manager, and rely on him to invest it in an appropriate way. This way, the ownership structure of institutional investors creates an added agency problem (Schneider, 2000; Solomon and Solomon, 2004). This is illustrated in figure 1, derived from Schneider (2000). What can be seen in this figure is that institutional ownership results in “agents watching agents” (Black, 1992). This is likely to be detrimental to monitoring by the institutional investor, as he is caught up between multiple interests, including his own. In fact, Gorton and Kahl (1999) conclude that institutional investors are imperfect monitors, i.e. less ideal owners, because of their own internal agency problems.

(18)

Figure 1: Added agency costs arising from institutional ownership (from: Schneider, 2000, p. 217)

2.6 Monitoring Incentives of Different Institutional Investors

2.6.1 Banks

Banks make up a large category of institutional investors, as trustees for everyone from pension plans to private estates. The most important thing to notice when discussing banks as shareholders is the fact that their shareholding is most likely not their only relationship with the firm. Banks can have a variety of different relationships with firms, ranging from creditor to advisor. Evidently, this commercial interest is a serious violation of the ideal owner characteristic of independence. Accordingly, banks can be expected to vote generally in favor of management, in order not to upset them in the light of a (potential) commercial relationship. However, banks sometimes have a significant credit stake in a firm as well. Although a credit stake does not provide much incentive for active monitoring, as the benefits of performance improvement flow to the equityholders, it can be used to force a certain proposal onto management in specific situations.

(19)

a bank, banks tend to vote in favor of management in antitakeover proposals. Moreover, in the absence of such relationships, banks tend to oppose such proposals.

Concluding, it is fair to claim that, because of their (potential) commercial interests with firms, banks do not comply with the ideal owner picture, as this is a violation of the ideal owner characteristic of independence. As a result from this, they lack the right incentives to thoroughly monitor their investee companies. However, their large credit stakes are a potentially powerful tool for putting pressure on management, although this will probably only be executed in some exceptional cases.

2.6.2 Mutual Funds

The most important distinction of mutual funds from other institutional investors is that they are designed for total liquidity (Monks and Minow, 2004). They are designed for investors who desire the flexibility to come in and out of the fund at any time they like. This leaves the investment manager with a lot of uncertainty about what he will have to pay out at a certain moment. Therefore, he sees his investments as collateral; they are simply there to ensure the flexibility demanded by their shareholders. As Monks and Minow (2004) rightfully notice, this is not the kind of relationship to encourage a long-term attitude towards any particular investee company. For example, in the case of a tender offer at any premium over the current trading price, mutual fund managers are practically compelled to accept it. In addition, because of the fierce competition for new clients, mutual fund managers may favor attracting new investors to maximize fund revenues rather than maximize after-tax return to current investors (Barclay et al., 1998).

In short, the constant need to attract new money in combination with ongoing competition forces mutual fund investment managers to focus on the short term, as he might not even exist anymore in the long term.

(20)

the foregone opportunity of increasing portfolio value through shareholder activism” (Davis and Kim, 2006, p. 3)

However, arguments in favor of mutual fund activism have also been made. Solomon and Solomon (2004) use the fierce competition among the many mutual funds to pose that engagement and dialogue have become an area of competitive advantage for institutions. Especially in today’s “green hype” environment this is a valid argument, as mutual funds are fighting each other for the crown of “greenest” and most responsible investor.

Furthermore, the empirical evidence on mutual fund activism is very mixed (Schneider, 2000). Many of them review proposals on proxy statements, and sometimes contact management (Useem, 1993), and some of them have been among the most activist of all institutional shareholders (Schneider, 2000).

2.6.3 Insurance Companies

Insurance companies are a relatively underexposed class of institutional investors. In the Netherlands, they are a significant market force, as over a quarter of all listed companies have an insurance company owning over 5 % of their stock (source: Amadeus database).

Insurance companies, perhaps more than any other class of institutional investors, have a complex relationship with the companies in which they invest (Monks and Minow, 2004). First, it is very likely that they hold some debt securities of an investee company, as debt instruments are very well suited for insurance companies’ needs and consequently form a large part of an insurance company’s investment portfolio. Second, insurance companies also have (or would like to have) a commercial relationship with their investee companies. Third, insurance companies have no obligation to report their voting behaviour to their clients. Furthermore, insurance company clients are, due to transparency difficulties, the least likely of all institutional investors’ to intervene with the company’s voting and monitoring behavior.

Concluding, insurance companies do not seem to be in the driver’s seat to promote corporate governance. Their monitoring incentive profile is very similar to that of banks. It should therefore be no surprise that the insurance industry consistently votes with management (Monks and Minow, 2004).

2.6.4 Pension Funds

(21)

Nederlandsche Bank Pensioen Monitor, 2006). Given these facts alone, a good understanding of this group of investors is essential in understanding the current and particularly the future state of corporate governance. In addition, and important in light of this study, pension funds are most often seen as the closest institutional approximation to the ideal owner.

The theoretical arguments in favor of the latter proposition are relatively straightforward. First, pension funds, and their holdings, are both large and sophisticated enough to minimize the collective choice problem. Second, their long payout horizon, typically 30 years, guarantees the long term perspective desired by proponents of the ideal owner. Third, the fact that their clientele is very divergent makes sure that their behavior reflects a good representation of the general public interest. As Monks and Minow (2004) state it: “It is virtually inconceivable that something would be in the interest of pensioners that is not in the interest of society at large” (Monks and Minow, 2004, p. 143). Fourth, pension plans are less, sometimes even not, restricted by the commercial conflicts of interest we encountered at the previous investor categories. Fifth, pension funds are to a large extent index investors. This makes their investments both diversified and permanent. Although this comes with some significant disadvantages as well, as will be discussed later on, the major advantage is that passive investing should come with active owning (Monks and Minow, 2004). Pension funds are principally stuck with their investments, which provides them with a strong incentive for active monitoring and initiating improvements when dissatisfied with the firm’s performance or policies. Finally, pension funds are subject to a sophisticated legal and supervision system, which ensures proper and ethical (investment) policies.

Although the above arguments are both numerous and intuitively appealing, the actual situation is more subtle and complex. Some characteristics of pension funds lead to incentives not in line with those of the ideal owner, nor accommodating the advantages of pension fund ownership as stated above.

(22)

actuarially defined payout or a market rate of return. The result of this is biased diversification and insufficient innovation (Roe, 1993), which means overinvestment in large companies and underinvestment in emerging companies. In addition, it is very unlikely that pension funds will encourage significant risk taking in investee companies, nor is it very likely that they invest in risky emerging companies, such as IT start up companies.

Second, the prudent nature of most pension fund investment policies has resulted in a strong preference for index investing. This has a couple of implications for their monitoring ability. It implies that pension funds, especially the larger ones, have shareholdings in an immense number of companies. Consequently, they can not be expected to be sufficiently informed about every investee company and extensively monitor them, as this would require uneconomically vast amounts of resources. Therefore, pension funds can be visible, but they cannot be very specific about strategic issues regarding a certain company. This issue will be further dealt with in section 3.

A third argument to the detriment of pension fund monitoring is their own governance structure. This is generally a complex structure, particularly in the case of public pension funds like the PGGM and ABP funds in the Netherlands. The trustees of these funds are a very diverse group, consisting of political appointees, elected officials, employee (union) representatives, employers’ representatives and a wide range of financial experts. Although the actual investments are managed by qualified professionals, or outsourced to professional investment banks, the investment policy is determined by the above mentioned group. These trustees all have their individual interests, perspectives and outside pressures affecting their investment and monitoring preferences. For example, Romano (1993) argues that public pension funds are subject to pressures to take actions that are politically popular, but could harm the funds’ investment performance. In addition, the salary of trustees is a nominal fee, with no incentives built in for superior investment performance (Annual reports ABP and PGGM, 2006). The investment managers of ABP and PGGM do have such performance-related pay, but this part never exceeds 25% of total pay.

(23)

afraid that, as a result of the large scale outsourcing, investment expertise would deteriorate at pension funds’ trustees. This would be a detrimental development, as the trustees are the people ultimately responsible for the investment policies and performance. In the Netherlands, pension funds traditionally manage most of their assets internally, so this potential

disadvantage should not be as material here as in the UK. When discussing the positive characteristics of pension funds as owners, the lack of

commercial conflicts of interest was mentioned. This advantage is only partly applicable, that is, only material in the case of public pension funds. Private funds do have incentives not to overly interfere with certain investee companies, as they risk retribution, for example in the form of retaliation in the market place, or an invitation to other private pension funds to adopt a similar aggressive view of their company (Monks and Minow, 2004, Schneider, 2000). The empirical literature on pension fund activism is inconclusive about its effectiveness. Del Guercio and Hawkins (1999) find that activist pension funds are generally successful in furthering their stated objectives onto the investee company. However, they also find that this does not lead to significant improvements in financial performance. This result is supported by Wahal (1996). Smith (1996) finds a more mixed result when he studies firms targeted by CalPERS (the California Public Employees’ Retirement System). He finds that shareholder wealth increases for firms that adopt proposed changes by CalPERS, and decreased for firms who do not. However, he finds no significant change in operating performance.

2.6.5 Conclusion

The theoretical discussion on institutional investors above has provided an extensive view on their characteristics and ability to perform the ideal owner role. Consequently, what can be the overall conclusion from this exposition?

First, there should be a fourth characteristic added to those of the ideal owner: a favorable own governance of the investor. Own governance is an important distinguishing factor between institutional investors and their monitoring incentive profile. Accordingly, it determines to a large extent their incentives to monitor.

Second, due to different governance structures, commercial relationships and other obstacles, institutional investors vary greatly in their ability to perform the ideal owner role. This is a very important thing to keep in mind when discussing institutional investors, as a group, as solution to governance problems.

(24)

incentives. Although these disincentives are both substantial and unignorable, I still consider pension funds as the most plausible candidate to fulfill the “legendary monitoring role” of the ideal owner. Therefore, in the further course of this study, the emphasis will be on this particular group of institutional investors.

To draw an overall conclusion, I would like to join Monks and Minow (2004) when they state that: “On both sides of the institutional investor spectrum there are plausible candidates for at least some forms of active monitoring. But the “carrot” of increased shareholder value is not enough to make it happen, in a world where the collective choice problem and political and economical reprises present overwhelming obstacles” (Monks and Minow, 2004, p. 180). In terms of this thesis, it is very difficult to find an investor possessing both the four ideal owner properties, and the willingness to use them to actively monitor the management of the firm. Consequently, the quest for the ideal owner is certainly not a clear cut case and will be continued in section 3, when the practice of institutional investor involvement will be discussed.

2.7 Shareholder Activism and Financial Performance

The strongest incentive for active monitoring should of course be that it pays off. Accordingly, this message is what proponents of shareholder activism and institutional investor involvement have been proclaiming over the past couple of years. Moreover, it is a clear assumption of most corporate governance codes, including the Dutch code, that institutional investors’ intervention in investee companies produces higher financial returns. However, academic research on the impact of shareholder activism on corporate performance and company value has produced mixed evidence (Solomon and Solomon, 2004).

An important thing to keep in mind when studying the relation between activism and firm performance is causality. Most of the time, investors only become active when or after an investee firm is experiencing problems, and remain rather silent otherwise. Consequently, the question is whether performance succeeds activism or vice versa. This is very likely to result in biased activism portfolios in the above discussed studies, and therefore we should be careful with interpreting their results.

(25)

when the actual performance of the investee firm is negative, as it might just have been more negative without the interference of the activist shareholder.

When studying the empirical literature on shareholder activism and firm performance, a couple of things stand out. First, Block purchases by institutional investors have been found by some authors to be positively related to company value, top management turnover, financial performance and asset sales (e.g. Mikkelson and Ruback, 1985; Shome and Singh, 1995; Franks and Mayer, 1997; Bethel et al.., 1998). Strickland et al. (1996) find that monitoring by small shareholders, organized via the United Shareholders Association (USA), results in enhanced shareholder wealth. As earlier mentioned, Smith (1996) also finds evidence for increased shareholder wealth as a result of activism by CalPERS.

Evidence on the contrary is also present. Agrawal and Knoeber (1996) find only a weak relationship between large institutional holdings and corporate performance as measured by Tobin’s Q. Demsetz and Villalonga (2001) find no significant relation between concentrated ownership and firm performance. Furthermore, Wahal (1996) and Karpoff et al. (1996) find little evidence that operating performance of companies which are the target of pension fund proposals improves. This result is supported by Del Guercio and Hawkins (1999). Final strong empirical evidence to the detriment of active monitoring by pension funds is collected by Faccio and Lasfer (2000), who find that blockholding by occupational pension funds in the UK has no significant positive effect on both profitability and company value.

Karpoff (1998) surveys the empirical evidence on the subject up to then. He concludes, while recognizing that the evidence is mixed, that most evidence indicates that shareholder activism can prompt small changes in target firms’ governance structures, but has negligible effect on stock price performance or earnings.

Studies examining the relationship between concentrated ownership and financial performance in the Netherlands are relatively scarce. De Jong et al. (2002) find a significantly negative relation, indicating self-interested behavior of the blockholder. On the other hand, when using a different performance measure, Chirinko et al. (2003) find no discernable impact of concentrated ownership on firm performance. This evidence is in line with the dual role (decreasing monitoring costs, but increasing potential expropriation costs) of large investors as discussed in Stulz (1988). Garretsen et al. (1999) study the relationship between ownership of Dutch firms by financial institutions and firm performance. They reject a linear relation, and in the case of banks, even find evidence for a nonlinear relationship.

(26)

(27)

3. Dutch Ownership Structure and Pension Fund Activism

After having discussed the theoretical issues underlying institutional investor involvement, I will now turn to practice. Whereas the previous section primarily dealt with the ideal owner characteristics of independence and long term perspective, this empirical survey will particularly deal with the other characteristic: significant stake. This will be done by examining the direct ownership structure of the Dutch market. In doing so, the overall goal is to get a thorough overview of the market structure, in order to check which market participants comply with the significant stake characteristic and are, judging on their stake, the influential players in the Dutch market. Furthermore, the emphasis will be on pension funds, as they came out as the most plausible ideal owner candidate after the discussion of the previous section.

3.1 The Dutch Ownership Structure

In order to get a clear picture of the actual involvement of institutional investors in the Netherlands, the direct ownership structure in the Dutch market will be examined. The summarized results are presented in table 1. The figures will be discussed hereafter.

Table 1 Dutch Ownership Structure

BI MF PF PE IND FAM MAN GOV Total

>5% 174 171 1 5 181 69 1 2 604 3%<x<5% 41 50 3 2 19 17 1 0 133 1%<x<3% 154 80 2 2 41 13 0 0 292 Total 369 301 6 9 241 99 2 2 1029 In Percentages in percentages >5% 47,15 56,81 16,67 55,56 75,10 69,70 50,00 100,00 58,70 3%<x<5% 11,11 16,61 50,00 22,22 7,88 17,17 50,00 0,00 12,93 1%<x<3% 41,73 26,58 33,33 22,22 17,01 13,13 0,00 0,00 28,38 Total 35,86 29,25 0,58 0,87 23,42 9,62 0,19 0,19 100

(28)

described below. In total, this resulted in 1029 significant stakes of 1% or higher, of which 737 are 3% or higher, and 604 are 5% or higher.

In order to get a thorough insight of the direct ownership structure and significant stakes, distinctions have been made along two dimensions: size of the stake and owner identity. With respect to stake size, three categories were made: share ownership between 1% and 3%, between 3% and 5%, and >5%. As for owner identity eight categories were created; Banks and Insurance companies (BI), Mutual Funds (MF), Pension Funds (PF), Private Equity companies (PE), Industrial and Other companies (IND), Family and/or Individual shareholders (FAM), Management (MAN) and Government (GOV). Banks and insurance companies are combined, because their respective monitoring (incentive) profiles are corresponding to a major extent, as has been described in section 2.6. Furthermore, in some Dutch cases (particularly ING Group) they are part of the same corporation.

There are a couple of observations that can be made from table 1. First, what immediately stands out is the very small number of significant stakes held by pension funds. Of all significant stakes larger than 1%, only 0,58 percent is in the hands of pension funds. This is quite surprising, as one of the assumptions underlying the theoretical allocation of the ideal owner role to pension funds is that their stakes are generally large.

Second, there are only three categories that hold almost all significant stakes: banks and insurance companies, mutual funds and the industrial/other category. These categories account for 88 percent of all stakes larger than 1%. The industrial/other category requires some further explanation. This category is a sort of repository, not only made up of holdings by industrial, non-financial companies, but also to a major extent of private administration companies, often owner (family)-related. This category is particularly visible at the smaller companies, whereas banks/insurance companies and mutual funds are far more active in the larger part of the sample.

When looking at the differences between the stake size categories, a couple of additional observations can be made.

First, what can be seen from table 1 is that the majority of significant stakes is larger than 5 percent. Large stakes are generally held by banks and mutual funds, whereas pension funds, if even holding a significant stake, hold relatively small stakes. However, the negligible number of pension fund stakes found restricts me from saying anything meaningful about this category.

(29)

strategic holdings, as well as owner (family)- related private administration companies. Both types of stakes can be expected to be relatively large. As expected, the family/owner category largely demonstrates large stakes.

An interesting party in the Dutch market is the ING Group. In table 2 below, their stakes in a selected number of Dutch companies are exhibited. What stands out is that they hold very large stakes in a majority of the Dutch market. In 23 Dutch companies, ING holds a stake of more than 5%, of which the 10,78% in ABN Amro and the 16,94% in Unilever are the most striking. Particularly when compared to the Dutch pension funds, the ING Group is a significant player in the Dutch market.

The fact that most stakes are situated in the >5 percent or <3 percent region, does not predict substantial improvement arising from the recently adopted change in Dutch law, which prescribes that all shareholders with a stake larger than 3 percent should report their stake to the Dutch market supervisor (AFM). Previously, this boundary was set at 5 percent. Based on the used data, the change in law will not result in many additional reported stakes. However, it must be said that by lowering the border, in total the authorities will capture a large majority (71%) of the significant stakes in the Dutch market.

Concluding, it can be said that the main forces in the Dutch market are banks, insurance companies and mutual funds. These investors, together with the more diverse Industrial category, take account for the vast majority of large stakes, of which most are significantly large (>5%). Of these large investors, ING is a very special case, as they hold very large to immense stakes in a significant number of Dutch firms. The theory behind these large stakes would be very interesting to analyze. It could be, for example, that ING invests these funds on behalf of other (foreign) investors. Unfortunately, the available data proved insufficient to make such an analysis.

(30)

Table 2 ING Stakes in Dutch companies

Company Mark Cap (x € 1000) ING Stake %

ABN AMRO Holding 47.162,24 10,78

Akzo Nobel 13.222,86 5,48

Buhrmann/Corporate Express 2.030,89 13,86

CSM 2.124,48 8,57

Fugro 2.491,47 10,11

Gamma Holding n.a. 26,49

Getronics 756,82 7,5

Heineken 17.653,78 5,4

KAS Bank 327,33 13,01

Koninklijke Ahold 12.534,32 7,3

Koninklijke BAM Groep 1.735,20 5,54

Koninklijke DSM 7.559,10 5,24

Koninklijke Philips Electronics 35.121,08 1,01

(31)

3.2 Dutch Pension Fund Activism

3.2.1. Pension Fund Stakes

The previous section showed that pension funds are, judging on their stakes, not a major force in the Dutch market. In fact, they only hold a handful of significant stakes. This is a surprising result, and is a drawback to the relatively positive monitoring profile of pension funds found thus far. Accordingly, in this section, pension funds will be further analysed. This will be done in both a quantitative and qualitative way. The analysis has been focused on the three largest Dutch pension funds: the public ABP and PGGM, and the private Shell Pensioenfonds. First, all holdings of these pension funds were analysed, in order to get a complete picture of Dutch pension funds’ holdings, as well as to check the figures found above. This could be done relatively easy, as both ABP and PGGM publish a document on their websites containing a list of all their participations, together with their respective market values. Unfortunately, the Shell Pensioenfonds did not have such a document available, so we will have to make do with the two public funds.

(32)

Table 3 ABP and PGGM Stakes in Dutch Companies

Company Mark Cap ABP Stake PGGM Stake ABP Stake % PGGM Stake %

Aalberts Industries 1608,36 0,1 0,01

ABN AMRO Holding 47162,24 256 28,5 0,54 0,06

Aegon NV 23435,07 118 17,7 0,50 0,08 Akzo Nobel 13222,86 36,9 6,7 0,28 0,05 Arcadis 964,15 1,9 0,20 Arcelor Mittal 5715,89 67,3 46,4 1,18 0,81 ASML Holding 9141,05 36,2 95,2 0,40 1,04 BE Semiconductor Industries 153,13 1,4 0,91 Boskalis Westminster 2144,98 3,6 0,17 Buhrmann/Corporate Express 2030,89 5 0,25 Corio 4162,91 1523,5 42 36,60 1,01 CSM 2124,48 3,3 0,16 Draka Holding 917,64 0,1 0,01 Eurocommercial Properties 10569,17 51,5 177,7 0,49 1,68 Euronext 10073,87 28,7 8,9 0,28 0,09 EADS 21289,53 165,8 0,13 0,78 0,00 Fugro 2491,47 7,5 0,30 Getronics 756,82 6,5 0,86 Hagemeyer 1982,65 8,5 35,9 0,43 1,81 Heijmans 1002,86 2 0,20 Heineken 17653,78 90,9 7,3 0,51 0,04 Hunter Douglas 2570,39 0,3 26,4 0,01 1,03 Imtech 1294,32 0,1 0,01 ING Groep 74274,55 236,1 631 0,32 0,85 KAS Bank 327,33 22,6 36,1 6,90 11,03 Koninklijke Ahold 12534,32 38,3 0,18 0,31

Koninklijke BAM Groep 1735,20 3,5 33,5 0,20 1,93

Koninklijke DSM 7559,10 53,2 51,8 0,70 0,69

Koninklijke Philips Electronics 35121,08 105,2 5,4 0,30 0,02

Koninklijke Wessanen 744,03 4,5 0,60

Nedschroef Hldg 199,04 0,1 0,05

Nieuwe Steen Investments 799,91 16,3 0,23 2,04 0,03

OCE 1098,79 10 0,91 OPG Groep 1285,21 3,6 0,28 Qiagen 1466,40 1,2 0,08 Randstad Holding 6070,85 4,7 0,044 0,08 Reed Elsevier 9655,49 21,6 3,1 0,22 0,03 Rodamco Europe 1035,9 11,46 24,51 Royal KPN 20770,50 110,3 126,7 0,53 0,61 Royal Numico 7762,42 17,1 0,1 0,22 SBM Offshore 3658,10 5,7 8,6 0,16 0,24 SNS Reaal 3857,13 32,3 0,84 STMicroelectronics 12805,89 21,9 0,65 0,17 0,01 Stork 1313,53 0,3 0,02 TNT 14972,45 67,5 27,5 0,45 0,18 Tomtom 3694,79 47,7 1,29 Unilever NV 35494,86 102,6 11,7 0,29 0,03 Univar 1271,67 2,5 0,20 USG People 2089,13 21,8 1,04

Van der Moolen Holding 193,95 5,4 0,29 2,78 0,15

(33)

3.2.2. Dutch Pension Fund Activism

All three funds, and in particular the two public ones, provide extensive information on their investment policies, voting behavior, AGM attendances and corporate governance policies and viewpoints. They all proclaim to vote on the majority of their investments, are concerned with the long run prospects of the firm and are even willing to enter into a dialogue with the management of individual firms, when deemed necessary. Furthermore, they publish internal corporate governance codes, as well as statements with respect to socially responsible investing (SRI). However, it must be kept in mind that this reporting has been mandatory since the introduction of the Dutch Corporate Governance Code. These documents might be part of a good news show and in so doing represent a certain form of window dressing by pension funds. We already saw that they lack the magnitude of stakes deemed necessary to get things done at any individual firm. The question is whether pension fund activism has real substance or if it is more an act of symbolism. There are a couple of indications that point in the latter direction.

First, given an investment portfolio of over 3500 stocks, it is impractible to extensively monitor all of these investments. Still, all three funds state that they do vote on most of them. This is where a couple of interesting and underexposed phenomena come up: Home bias with respect to activism, the use of both proxy firms and external fund managers, and security lending.

(34)

services….may constitute a mechanism through which highly isomorphic shareholder policies and activities result from very uniform company reports and voting recommendations” (De Jong et al., 2007, p. 6). Accordingly, when discussing institutional investor activism, this group of firms is really one to watch. This particularly applies to the Dutch market, as a study by Eumedion (a foundation of joint Dutch institutional investors) finds that over 70 percent of the Dutch market is in foreign hands. If we apply both the home bias phenomenon to these foreign investors and their presumable use of proxy firms for foreign investments to this case, it can be expected that there will be a lot of proxy voting at Dutch AGM’s.

In section 2.6 the detrimental influence of the use of external fund managers on monitoring by pension funds was already discussed.

The final phenomenon to watch is security lending. This is an activity widely conducted by institutional investors. Although this activity is economically beneficial to both the parties involved and the market as a whole (i.e. short term profit from lending fees, more efficient transaction settlement and increased market liquidity), it might produce conflicts of interest with respect to corporate governance. This results from the fact that voting rights are transferred together with the shares, and are generally not retained when the voting season comes around. This way, for example, a shareholder aiming to get something done at a particular firm can obtain a significant temporary voting block by collecting voting rights through lending from multiple (institutional) investors. Furthermore, the bear fact that institutional investors lend their stock without concerning themselves over the loss of voting rights is an indication of less commitment to the governance of investee firms than initially proclaimed by those investors.

(35)

In sum, the above findings provide some empirical drawbacks on the perceived ideal owner role of institutional investors in general, and pension funds in particular. Although pension funds state that they are actively involved in their investee companies, both the size of their stakes and the scope of their portfolio restricts them from making a significant difference. Furthermore, activities like the large-scale use of proxy firms, security lending and the home bias with respect to activism indicate that in practice, Dutch institutional investors are not as concerned or actively engaged with the governance of their investee companies as they proclaim themselves, and as is perceived by regulators. This observation is supported by the empirical findings of De Jong et al. (2007).

3.3 Two examples from the Anglo-Saxon World

The foregoing analysis of the Dutch market did not provide evidence for an ideal owner role of Dutch institutional investors in general, and pension funds in particular. The findings and arguments put forward tend to oppose the favorable theories on institutional investor involvement described in earlier sections.

(36)

3.3.1 USA: CalPERS

When discussing institutional investor activism in the US, attention is mostly directed to one particular institutional investor: the California Public Employees’ Retirement System (CalPERS) (see e.g. Smith, 1996; Crutchley et al., 1998; Barber, 2006).

CalPERS is the largest pension fund in the United States and one of the largest in the world. More importantly, it has been a leader of institutional investor activism since the mid-1980s. Therefore, it is interesting to examine their policies and activities concerning activism, as their knowledge of and experience with activism can be expected to be superior to any other pension fund or institutional investor. Furthermore, they have actually been successful in getting proposed changes implemented at targeted firms, as well as improving shareholder wealth in the process (Smith, 1996; Barber, 2006).

First, it was checked whether CalPERS holds larger stakes than found at Dutch pension funds. The results for the Dow Jones Industrial Index are depicted in table 4.

Table 4 CalPERS Stakes in Dow Jones Industrial Index

Company Mark Cap bln $ CalPERS Stake Calpers Stake %

3M Co. 54,37 304,70 0,56

ALCOA Inc. 29,05 137,01 0,47

Altria Group Inc. 176,64 48,77 0,03

American Express Co. 68,19 322,12 0,47

American International Group Inc. 174,47 718,35 0,41

AT&T Inc. 229,78 522,29 0,23

Boeing Co. 68,87 362,02 0,53

Caterpillar 41,61 229,71 0,55

Citigroup Inc. 247,42 1.242,89 0,50

Coca-Cola Co. 108,08 520,77 0,48

E.I. du Pont de Nemours & Co. 46,88 171,14 0,37

Exxon Mobil Corp. 410,65 1.901,58 0,46

General Electric Co. 358,98 1.777,71 0,50

General Motors Corp. 18,04 880,24 0,49

Hewlett-Packard Co. 106,27 526,91 0,50

Home Depot Inc. 80,8 361,84 0,45

Honeywell International Inc. 37,12 151,53 0,41

Intel Corp. 114,53 524,01 0,46

International Business Machines Corp. 139,92 674,64 0,48

Johnson & Johnson 182,12 940,65 0,52

JPMorgan Chase & Co. 170,97 831,60 0,49

McDonald's Corp. 52,58 233,54 0,44

Merck & Co. Inc. 95,85 396,88 0,41

Microsoft Corp. 275,85 1.257,99 0,46

Pfizer 179,97 858,16 0,48

Procter & Gamble Co. 200,34 849,35 0,42

United Technologies Corp. 65,29 321,85 0,49

Verizon Communications Inc. 109,19 480,18 0,44

Wal-Mart Stores Inc. 201,36 932,41 0,46