PART IV

MONITORING AND CORPORATE GOVERNANCE IN THE UNITED STATES

Introduction

Commentators have suggested that forms of corporate governance can be identified along several different axis. Perhaps the most commonly suggested spectrum stretches from passivity to control. The United States is generally cited as an example of 'passive' corporate governance while Germany is given as an example of active, or 'controlling' corporate governance. In this schema, Japan would be mid-spectrum characterized by a variety of 'voices' typically exercised within the keiretsu. Even though corporate owners in the United States generally play a passive role in corporate governance, some owners -- particularly some institutional investors -- are becoming more and more active. Their activity, however, falls far short of the controlling role owners play in Germany or that cross-holdings play in Japan.

In slightly different terms Coffee sees a polarity of liquidity and control. Liquidity is the ability to move into or out of an investment quickly and is associated with low levels of control. In governance schemes where ownership control is high (e.g. Germany) liquidity is substantially reduced. Likewise, Porter writes of fluid (and transient) versus dedicated forms of capital allocation and the OECD has written of market and institutionally based governance systems. Gilson proposes a polarity of stability and adaptability (or 'mutability'), which is somewhat different than the others. By stable systems he means something similar to dedicated or institutional arrangements, but by adaptable he means something quite different than necessarily market (that is, liquidity) based. Adaptability is not (yet) identified with the US although he sees the US as potentially more adaptable than others. Mutability is the ability to change as situations change so that governance serves as what he calls 'an equilibrating device'. Thus for Gilson the goal is not a theoretically optimal governance system, but "...the ability to mutate in response to an ever more quickly changing world [so that]... institutional structure facilitates prompt and low cost organizational responses to changes in industrial technology." Thus, technological changes becomes the driver of governance.

Whatever terms various authors use, the core idea is similar: If the agency problem is to be addressed in ways other than purely through a market for corporate control, then there must be some form of monitoring and/or relationship investing which will either supplement or replace that market. Briefly what is meant by monitoring is some type of formal or informal organizational accountability of managers to the benefit of owners, or chains of owners. A specific form of monitoring is 'relationship investing' in which an owner (or groups of owners) hold a significant block of equity in a particular firm, thereby establishing a long-term position and by virtue of their ownership block can exert leverage on management. This is unusual in the US context currently, though some authors believe it will become more important in the future.

In this part we will explore the issues that surround monitoring and relationship investing in the United States. In the next part we summarizes the major policy implications and suggestions that flow from this summary.

The Political Model of Corporate Governance

As noted, both monitoring and its more active form, relationship investing are means of attempting to reduce agency costs which inhere in the relationship between either owners and managers, or fiduciary agents and other (fiduciary) agents. Reducing agency costs involves attempting to align the interests of managers with those who watch them, or if not possible, then to limit their opportunistic abuse of position. Thus, monitoring is an attempt to reduce or limit agency costs which relies neither primarily on optimal contracting nor on the incentive mechanisms of the market for corporate control, although it recognizes each as important. As Grundfest argues, inherent in the idea of monitoring is a view of management as an 'independent constituency in its own right' or an interest group.

Pound defines the 'political model of governance' as an approach, "...in which active investors seek to change corporate policy by developing voting support from dispersed shareholders, rather than by simply purchasing voting power or control...", as would be the case in a market or transactions based system. While there is a range of monitoring and relationship investing, all have in common the initial (or total) bypassing of the market for corporate control. Most commentators agree with Pound when he writes that a major economic advantage of the 'political' model is that it is more flexible than the takeover market since, "...it can address specific problems at a corporation without imposing changes in control, changes in management, and the enormous transactions costs attendant to them." Similarly Grundfest argues that, "Capital-market behavior is...guided by a jointly determined political [regulatory] and economic equilibrium, and an understanding of the political marketplace is essential to appreciate the role that capital-market mechanisms can...play in corporate governance." Black argues that shareholder monitoring is part of a series of 'imperfect constraints' on corporate managers which may also include the market for corporate control, the product and capital markets, the market for corporate managers, incentive compensation arrangements, creditor monitoring and the risk of bankruptcy.

A quite different approach is taken by Kester. For him governance deals with problems of coordination and control, which are of two types: those associated with the separation of ownership and control, which is the focus of much agency theory (e.g. Jensen/Meckling; Berle/Means); and those associated with contractual exchange among separate commercial enterprises, which is the focus of transaction cost economics when applied to corporate governance (e.g. Coase and Williamson). Both should be seen as governance concerns, but he distinguishes between the former as 'corporate governance' and the latter as 'contractual governance.' Most of the 'governance' literature focuses on the former, while Kester suggest that there is a rich possibility in the latter as well. The difference between contractual and corporate tends to characterize the Anglo-American tradition from the Japanese in particular. For Kester, "...the best practice requires not only a corporate governance system that minimizes agency costs, but also a parallel contractual governance system that minimizes transaction costs, broadly construed." Similarly, Roe discussed 'relational contracting' as a potentially significant part of governance, while Porter discusses what he sees as competitive advantages for US firms in expanding contractual governance.

Monitoring by Institutional Investors

In his 1992 work Porter was unduly pessimistic about the desire and ability of institutional investors (especially those with indexed portfolios) to monitor. He viewed the short-term nature of "monitoring and valuing stocks [through the market] on current earnings" as a destructive and self-fulfilling prophesy. Although the proxy system could and was used as the "only direct means of imposing ...[institutional investors'] views on management", it was, as noted, an expensive and clumsy means, typically used as a last resort in crisis situations. His conclusion was that "...despite their large aggregate holdings, American institutional agents have virtually no real influence on management behavior."

Yet it was exactly beginning in the early 1990's that public pension funds began to actively monitor their 'passive' portfolios primarily because their indexed funds reflected the market as a whole. If they could not sell (by definition), they had to care. Porter's study entirely missed this dynamic of indexed funds due to his singular focus on fragmented aggregated holdings. However the benefit of monitoring and targeting underperforming firms is captured by Pound when he writes: "The percentage of stock held by a specific investor is largely irrelevant to the decision to monitor. What is relevant is the raw dollar amount at stake, relative to the economic costs and benefits of becoming informed. With a sufficiently high raw dollar level of holdings, the incentive to become active is great, even if the proportion of stock held is very small." Thus, free rider problem inherent in collective action situations does not pertain as long as the institution believes that its returns (either firm specific or for all underperformers) will be increased through monitoring. In this sense it appears that Porter mistook an agency problem (which exists in all monitoring situations) for a collective action problem. Richard Koppes, Chief Counsel of CalPERS, concludes, "...the cost-benefit analysis of shareholder activism is an essential step in developing strategy, and that some strategies, such as direct discussions with management at selected corporations [informal monitoring], are an inexpensive means to an effective end."

Institutional Voice

Some large activist public pension funds and more recently private funds have begun to monitor due to the sheer size of their holdings. Yet the small percentage of shares held in any single firm and their highly diversified portfolios has been the focus of a significant number of studies. Of particular concern has been revisions in the law to permit larger concentration of holdings without running afoul of current SEC and other regulations, as well as various ways institutions can coordinate 'voice'. (In Part V we discuss various legal reform proposals. Here we concentrate on coordination of voice.) Black makes a distinction between what he calls institutional 'voice' and institutional 'control.' The former requires agents to watch agents, often agents who may not have identical interests and for that reason may let greater amounts of daylight into the board of directors in a sort of checks and balance approach to governance. Institutional control suggests the domination of one or a small number of institutions over a particular firm, as in the German model. Roe suggests that a structure of multiple intermediaries, "...can deter opportunism by monitoring one another, impel action in a way that a single blockholder might not, and facilitate power-sharing, not domination." It also avoids or minimizes the political problem in the American tradition of deep distrust of concentrations of wealth and power. Yet what is striking about the current reality is that few large institutions have amassed anything close to the five percent limit (before severe SEC regulation kicks in). As Roe point*s out, even a number of institutions owning near five percent would need to have on-going coordination in order to 'wrest control' from management.

In order to enhance institutional voice, Black advocates that institutions slim portfolios so they hold between 5-10% of the stock in a particular firm (but not over this). Communication among these significant minority holders would then be easier, permitting stronger influence in selecting board members, but at these levels of ownership no single owner is likely to dominate the board or its selection. Thus, a half dozen or so owners could collectively influence major corporate actions, usually through their influence on the board, but no one institution could dominate or act effectively alone.

Why have US institutions avoided slimming down portfolios , and not taken larger positions in firms? That is, why have they resisted becoming relational investors with significant stakes rather than coalition monitors? Analysis is divided between the majority who see law and regulation (read: the political tradition) as the primary inhibiting factor and those (most especially Coffee) who see the desire for liquidity (apart from the legal constraints) as the primary factor. On the legal side concerns about 13-d restrictions which define (in some highly uncertain situations) two or more persons as a 'group' who, as a group, might be 'beneficial owners' of five percent or more if their holdings are aggregated. Such aggregation, in Briggs' words, can be determined by the courts "if they have agreed to act together 'for the purpose of acquiring, holding, voting or disposing' of the issuer's equity securities." Such agreement need not be in writing and can be inferred ex post from circumstantial evidence. The danger of triggering a 'group' is clearly an inhibiting factor for accumulation of even relatively small percentages, and an incentive for caution around voting and other strategies. Numerous other restrictions are detailed by Black, some of which are discussed below.

Multiple Monitors

In Black's distinction between institutional control and institutional voice, he sees voice as both more desirable and more possible in the US context. However, for him voice has several dimensions. First, voice would mean asking "one set of agents (money managers) to watch another set of agents (corporate managers)." Secondly, it would mean that different types of institutional agents would join together to 'exercise influence' over corporate managers. Finally, in a third dimension of voice, it would encourage corporate managers to 'watch their watchers.' In Black's view, much of this checks and balance system should be informal.

Grundfest advocates that multiple monitors use 'just vote no' campaigns directed against management director slates as a whole, or individually targeted directors, as CalPERS has been doing in 1995 and 1996. While typically these campaigns achieve only minority support (however, often upwards of 35%), they are symbolic. But Grundfest argues that symbols are often effective as public flags that something is wrong, and can be important first steps toward correcting perceived problems, usually by putting pressure on the board. He quotes Delaware's Chancellor William T. Allen saying, "'just vote no' is like corporate chicken soup; it couldn't hurt, and what if it helps." One example of a 'just vote no' campaign was the 1994 campaign directed against the Westinghouse corporation initiated by the New York City Teachers Union pension fund which gathered over 35% of the proxies for a no confidence vote. The Westinghouse campaign was relatively successful in that it allowed shareholders to cast a no confidence vote about a firm which, in Grundfest words, "...there [was] general consensus that management or strategy [was] deficient."

Roe also argues in favor of multiple monitors, summarizing as follows:

Multiple blocks may induce intermediaries to act in a way that a single isolated blockholder [as in Germany] might not. The multiple blocks may solve the weakness problem without creating a dominating block that recentralizes authority in a dominating intermediary.

The idea of multiple monitors is also key to Pound's idea of the political model. He suggests seven types of recent evidence that monitoring is increasing and 'strengthening' the political model. First, that proxy contests are moving away from control as the goal to sending a message, increasingly settled by negotiations between dissidents and management. Second, the use of formal voting challenges to provoke debate (as in 'just vote no') over specific policies, rather than as a attempt to get control. Third, the use of public lobbying without a vote, such as forming a 'shadow board'. Fourth, private, informal negotiations, as CalPERS has been conducting in the past few years. Fifth, the development of 'smart voting systems' which target firms on the basis of economic performance, such as the New York, Wisconsin and California pension funds use. Sixth, management responds to these initiatives in a variety of ways, for example, by establishing an 'investor ombudsman'. Finally, Pound suggests that there is a 'new dynamic' between large investors and large corporations, reflected in the meetings between various investors and CEO's at various firms such as GM, Westinghouse, IBM, American Express and others. As the Institutional Investor noted, "The biggest single spur to the growth of the shareholder movement ....occurred when pension fund activist began linking governance issues...to concerns about cooperate performance."

The Partial Institutionalization of Insurgency

In the near future it is likely that activist institutions will continue using proxy filings, engage in 'just vote no' campaigns against incumbent directors, and informal lobby and monitoring. CalPERS, for example, intends to focus its strategy based on these points, especially on what it calls the 'personal accountability of directors', and to focus as well on smaller companies (that is, those not in Fortune's top two hundred firms). In the past CalPERS has set the pace for governance activity, and we would expect this to continue in a general way. In an August 1995 policy statement CalPERS defined its current monitoring focus as follows:

...As a result of its focused activities at selected companies [in the past, CalPERS]...has also gained the attention of other [non-targeted] companies, as the common issues in a peer group and concepts of corporate governance in a general sense. Taking advantage of this phenomenon, CalPERS seems to 'move the herd' rather than follow it, while still retaining the risk-reward reliability of an indexed portfolio.

In sum, what seems to be occurring is a partial institutionalization of insurgency based on the fragmented ownership patterns of institutional shareholders. Under these circumstances monitoring must take a widespread coalition form rather than either a control form or even an institutional voice form. Since it is unlikely that the underlying ownership structure will change rapidly (if at all), a number of proposals for better coordination among monitors have been offered. For example, some authors have suggested that various 'trade groups', such as the Council of Institutional Investors, should play a more important, albeit, informal role in coordination. A number of proposals for increasing monitoring while reducing the fragmentation of ownership have suggested that there should be 'lead monitors' whose holdings are greater than those typically held by the large indexed funds. Such institutions would of necessity reduce their portfolios from approximately 1000 or more to a few hundred positions, without increasing risk, assuming the capital asset pricing model is correct. This would enable institutional holders to divide up monitoring responsibility to a greater extent than currently. Yet such a change is likely to occur only if there are significant regulatory and legal changes on a more widespread basis that has yet occurred.

Generic Impediments to Monitoring

The US federal regulations that provide impediments to monitoring fall into two broad categories: those that regulate voice and those that regulate exit. As the authors of Lifting All Boats observe,

...a fundamental tradeoff exists between 'exit' and 'voice'...if voice is more feasible and attractive, institutions will rely less on exit and have an incentive to hold larger equity stakes in individual companies and monitor them more closely. The challenge for public policy is to strike a balance of incentives within the market for exit and voice. Too little liquidity can prevent the deployment of resources...but where liquidity is the only choice...this can impose costs in the form of excessive transactions costs and shortened investment time horizons.

Much of the focus has been on voice (e.g. S.E.C. communication rules), while less has been on exit. The following discusses current regulation of voice and exit while Part V discusses various proposal to deregulate voice and exit laws, and establish new regulations in other areas of exit and voice.

The area of legal and regulatory impediments to effective monitoring contains a vast and complex legal and political literature. In what follows, we do not attempt a comprehensive review of that literature, we merely attempt to highlight the most important topics.

Regulation of Voice: Four ExamplesRegulation of Voice Four Examples

Section 13(d) of the 1934 Securities Exchange Act

Section 13(d) of the 1934 Securities and Exchange Act requires 'any person or group' that beneficially owns 5% or more of a public company's stock to file a disclosure statement with the SEC. "While the filing isn't complex it involves substantial litigation risk which means heavy legal bills win or lose." As Briggs notes in his discussion of the 1992 communication liberalization, the main problem regarding communication is from 13(d) since there is a new 'risk calculus' between the post-1992 communication standards and the potential liability under 13(d). "The SEC's reforms [on communication] probably do not go far enough. Shareholder free speech of all sorts, especially speech advocating action, continues to be chilled by the proxy rule and by section 13(d)." The authors of Lifting All Boats tend to agree: "...the real threat ... is that the information disclosed can easily become the subject of securities litigation brought by the target company.... [The rules under 13(d)]...have a substantial regulatory barrier to collective action by shareholders (if the group agreeing to vote together collectively owns 5 percent or more)."

Section 16(b) of the 1934 Securities Exchange Act

This section limits the ability to sell after a 10% threshold has been passed. Coffee argues the law is very imprecise about under what specific circumstances a 'group' of holders become subject to the law, as in a voting group. While this provision of the law clearly regulates exit (its intent is to prohibit short term movements in and out of the market), one of its provisions does affect voice because 'group' status may be inadvertently triggered by communication. If this happens, the law requires a 'reporting' company (a company in the legal sense of an investment company) to turn over profits made on a purchase and sale within a six month period. In yet another section of 16(b), as the authors of Lifting All Boats state, "...even it they [institutions] stay below the 10 percent threshold, they may still be subject to its recapture provision if they have participated in nominating and electing a member of the corporation's board....Thus, an institutions owning only 1 percent may be subject...if it belongs to a shareholder group that has elected a director..."

Fiduciary Liability

The fiduciary standard under which pension plans operate requires that all actions be taken in the sole interest of the beneficiaries and that those actions should be those of a 'prudent man.' Failure to properly discharge a fiduciary duty exposes pension fund administrators to personal liability. Thus, Black points out that under the 1974 Employee Retirement Income Security Act (ERISA) which explicitly adopted the prudent man rule as well as under the common law of trusts, the "prevailing watchword...is caution" thereby favoring broad diversification and passivity, while concentrated ownership and activism with its accompanying voice are 'dangerous.' These rules reinforced similar provisions in the 1940 Investment Act which had codified such conservatism and its chilling effect on shareholder voice

Section F of the 1934 Security Act: Proxy voting procedures, access and agenda setting

The traditional avenue for shareholders to express their voice has been the proxy system under which proposals are submitted to a vote of the shareholders. Consequently, any proxy voting rules that significantly limit the scope of proxies or limit shareholder ability to communicate with each other are a sever limitation on shareholder voice. A number of rules do. In particular proxy rules define 'solicitation' which exposes a shareholder to onerous reporting requirements and potential liability so broadly that even a communication that seems on its face not to solicit can still be ruled a 'solicitation'. This is still a problem even though the S.E.C.'s 1992 revisions in communication rules loosen this restriction to some degree. S.E.C. Rule 41a-8 protect limited rights of shareholders to use a company's proxy statement to offer proposals, while severely limiting such statements in a variety of other ways by predetermined standards on what is 'appropriate' substance for such resolutions. Rule 14a-7 limits the ability to gain access to shareholder lists. The cumulative impact of these and other restrictions makes the costs of, for example, electing a single non-management slate director or making a counter-proposal to management extremely expensive. It also opens up various types of institutions to legal liability. Roe summarizes the pre- and post-1992 situation: "Managers control the proxy machinery [thereby effectively controlling agenda setting]....managers can find out how shareholders vote."

Regulation of Exit: Four Examples

While there is greater regulation of voice, there is also important regulation of exit (selling a position), especially for an individual or group deemed to 'control' a firm. This, of course, discourages an institution from taking a position which might be construed as 'controlling'.

Insider Information Restriction on Long-Term Indexed Investors

Coffee argues that since indexed investors are of necessity long-term investors (and thus potentially effective monitors) regulation which prohibits access to insider information is an effective restraint on good monitoring since such monitoring logically requires insider information. Long-term investors cannot profit from short-term uses of insider information (the purpose of the restrictions), and thus a distinction between traders and long-term investors might best govern insider information rather than the current blanket prohibition.

The Securities Act of 1933

Under the Act an 'individual' or 'group' in 'control' of a firm is taken to be an 'affiliate', and is restricted in its ability to resell any of its shares except after a registration statement is filed or an exception permitted. The effect is to put some restriction "on liquidity before resale is likely to be economically feasible" for the individual or group, potentially forcing the individual or group to hold equity longer than they otherwise might, according to Coffee.

Section 16(b) of the Securities Exchange Act of 1934

While a disincentive to voice as discussed above, this section also discourages exit by effectively creating a six month period of illiquidity in order to prevent insider trading against 'short-swing' trading profits, that is, profits potentially earned by speculating on short-term movements. Again for long-term holders the effect is to create uncertainty, in Coffee's words, "...because current law is conspicuously imprecise as about when members of a loose association of institutional investors become subject to it."

Insider Trading Liability

An institution which has an employee or agent serving on a corporate board is potentially liable if it trades with insider information, even if "...those making the trades did not use or have access to it." The effect of this is that in order to maintain liquidity, "...the institution must delegate its monitoring role to an outside agent, with whom it must have only limited contacts." This would be the case only when an institution has an employee or agent serving on the board.

Relationship Investing as an Alternative to Monitoring a Fragmented Portfolio

We have devoted significant space to various forms and tactics of monitoring. Most of this monitoring is informal and ad hoc and arises in the normal course of events at institutions with fragmented holdings. However, a number of authors have put forward 'relationship investing' as an alternative to the informal monitoring of fragmented portfolios. We now turn our attention this topic.

In relationship investing an institution takes a significant stake in a firm and, in Koppes' and Reilly's words, has a "...long-term commitment, and reciprocity between owner and management over business policy decisions..." While monitoring is widespread, influential and growing, relationship investing is the rare exception rather than the rule. The emergence of significant relationship investing is limited by formidable legal and historic-cultural barriers. Monks and Minow summarize some of the most important legal barriers for all types of financial institutions.

With commercial banks, there is the prohibition of Glass-Steagall; mutual fund holdings are limited by the Investment Company Act of 1940; insurance companies are limited by state law; private pension plans are required by ERISA to diversify as widely as possible; the federal system under FERSA [Federal Employment Retirement System Act of 1986] is limited to equity investment through index funds. These provisions, enacted independently, have a cumulative impact of preventing the financial sector executives from being able to exercise control over commercial sector executives--to keep Main Street independent of Wall Street.

Kleiman et al. review the performance of a number of 'patient' capital investors (loosely 'relationship investors'), which range from 'white-squire' funds that assist firms targeted for hostile takeovers to straightforward purchases of major equity positions. They conclude that early evidence suggests generally positive benefits to all shareholders, but warn against possible future 'backlash' if not handle carefully, and also of possible 'corporate inefficiencies' if relational investors attempt to micro-manage firms.

They (and others) distinguish between negotiated and non-negotiated forms of relational investment, the former making a long-term investment in return for a board seat, for example; while the latter offer typically unsolicited 'suggestions' about what should be done. Negotiated positions typically have been taken by investment banks or investment companies (e.g. Corporate Partners, GE Capital, 1818 Fund, Warren Buffett/Berkshire Hathaway, Inc.). Non-negotiated forms have been pursued by corporate governance funds (e.g. Lens Fund), which focus on underperforming firms by using issues such as shareholder rights, executive compensation, equal voting rights and board of director independence as a means of leverage. Kleiman et al. conclude that the former have been moderately successful, while governance funds (fewer and smaller capitalized firms) less so. They write: "Relationship investing that focuses on a shareholder value perspective undoubtedly will play a major role for a few more years....[yet] investors may discover that relationship investment, not unlike the diversification movement of the 1960s and the LBO activity of the 1980s, is nothing more than a fad of the 1990s."

The other form of 'relationship governance', as discussed above by Kester and by Porter, is contractual governance. Although still uncommon, large firms taking equity positions in small supplier firms is increasing, as, for example, when Ford has taken positions in a number of its supplier companies. The goal, however, is not financial performance monitoring, but rather the closer coordination of firm strategy, product development and delivery, research and development, and the monitoring of quality.

Increasing the Focus on Boards of Directors

Governance activities and much analytic commentary since 1992 have been directed at examining the role of the board of directors in either implementing or impeding the various demands of institutional investors. There is a large and rapidly growing literature on these aspects of governance, especially in the US and the United Kingdom. A passive, CEO dominated board traditionally was seen as the logical, albeit not the legal, result of managerial control. In spite of the political language and legal procedures of democratic process (votes, elections, ballots), the reality in managerial theory and corporate procedure was shareholder disenfranchisement and a CEO dominated self-perpetuating, passive board.

Eisenberg suggested there were fours ways in which managerial domination was achieved. First by 'nominal consent' when (as permitted under proxy rules) the shareholder does not vote at all and management votes its share, or it is voted by a third party broker or depository institution. Second by 'tainted consent' when a conflict of interest of an institutional owner arises but the institution nevertheless votes in favor of management because of the commercial network of the institution and the firm. The third by 'coerced consent' when management links a benefit for shareholders to a provision that may be contrary to their interests (e.g. a cash payout linked to an anti-takeover measure). The fourth by 'impoverished consent' when shareholders are forced to select the lesser of two evils replacing a bad rule or policy with one only slightly better due to management control over the proxy agenda process and the funding of the proxy process itself. Eisenberg concludes that shareholders are effectively 'disenfranchised.'

Contestability in the Board Room

As boards have increasingly come to be seen as the key link between institutional monitors or would-be-monitors (whether of a 'relationship investment' or 'monitor as catalyst' nature) and operational management, the aim of reformers has been to create a situation of contestability in the board room. Dale Hanson, former CalPERS CEO, characterizes this as simply 'letting the sun shine in,' in order to make the board accountable. There has also been some interest in the German dual boards which codify a clear division between managers who manage and non-executives who monitor them. While dual board are unlikely to be imitated in the U.S., institutionalization of 'truly' independent directors as effective monitors has been an important focus of reform. Measures to increase board independence include the creation of independent board sub-committees on key issues such as CEO compensation and board nomination. Despite halting attempts at reform the core problem remains: boards are self-perpetuating and self-selecting, even in those cases where the once near absolute power of the CEO is dramatically reduced.

Roe suggests there are three general types of board monitoring: hierarchical, collegial and crisis. Hierarchical relies on specialization as, for example, when a board financial expert focuses on optimizing financial returns while an operating executive focuses on running the firm. Both voices need to be on the board yet not in the same person. The second form is collegial. It depends on directors having proper incentives to be informed and involved. This has been the primary focus of board reform in the recent past. The third type of board monitoring is crisis monitoring. This requires that monitors evaluate the causes of poor results, especially as a means of forestalling crisis and, if a crisis occurs, respond to it appropriately. Well publicized failures in crisis monitoring sparked much institutional investor interest in board restructuring and reform.

Directors have the legal authority to execute almost all the reforms that various corporate critics, monitors and would-be-monitors desire, but in the words of one chairman of a large firm, boards have the power "...to govern, but not the process." The issues are: what should boards do, how shall they do it, and in whose interests (and how conceived) shall they act? An important, if intractable, part of board reform is socio-psychological and cultural. In the words of The Economist:

Whatever people's stated roles, small group dynamics tend to make them agree with each other and to forget their role representing outside interests. Non-executives come to respect the judgment of a successful boss too much, or suppress misgivings so as not to upset him. A better way for shareholders to make a board work for them is to make it their direct responsibility to see that it does.

Critics' Agenda

The critics of CEO dominated boards have focused their attention on the following areas and practices:

If monitoring and board structure have characterized US corporate 'politics' in the last five years, so have they begun to influence global governance issues in some of the leading OECD countries. While the focus of this report is corporate governance in the Untied States, it is appropriate to include a brief discussion of some of the ways in which corporate governance concerns in the United States may be transmitted overseas. The next section briefly turns to this topic.

Globalization of Monitoring and Global Financial Markets: Convergence of Governance Systems?

In 1994 private pension held 7.9% of their equity portfolio in foreign stocks, up from 5.9% in 1991, while public pensions (including TIAA-CREF) held 9.5% in foreign stock. (The largest 25 private, not-for-profit and public funds held an average of 10.5% in foreign stock, while CalPERS, the State of Connecticut and IBM were the largest, holding between 14.3 and 17.7% foreign holdings.) Since many US institutional investors have set as a goal a minimum of 20% of their portfolio in non-U.S. equity, the logical implication of this greater foreign investment is the potential to monitor it. As a consequence, a CalPERS, TIAA-CREF, and some other institutional activist have already begun to develop cross-border monitoring programs. Indeed, directives of the Department of Labor in 1988 and again in 1994, the so-called 'Avon letters', mandated as part of fiduciary responsibility the diligent exercise of voting proxies (the 1988 letter), and in 1994 also mandated that ERISA pension funds engage in "activities intended to monitor or influence corporate management." This is taken to include overseas monitoring by US institutions. These emerging programs have had an impact on corporate governance debates and practices in Japan, France and Germany, with significant potential in other countries as well.

CalPERS (as one of the leading US activist investors) has set the pace in overseas monitoring. While limited overseas activities date back to 1985, it was only in 1994 that significant energy and thought were directed at the international program. Aiming at 20% of its portfolio in overseas equities in the coming years CalPERS sees itself as a permanent long-term investor applying its US experiences overseas: "Shareholder activism improves economic returns in the US There is no reason to believe that similar results are not possible globally." CalPERS, with the assistance of the [U.S.] Council of Institutional Investors, and shareholder organizations from a number of European countries considered forming an "International Corporate Governance Organization." While the efforts of CalPERS and others are tentative and initial, it is likely that they will develop significantly in the coming few years and will increasingly interact with governance activities in many of the OECD countries.

The potential significance of US overseas monitoring is not that US governance practices and principles will become the global standard, but that both liquidity and contestability appear to be increasingly global. This is especially so as global financial markets come to play far more central roles in countries with long histories of closely knit networks between firms and financial institutions, e.g. German and Japan; or with firm's depending directly and indirectly on state support, e.g. France and Italy. In spite of the vast differences in national systems of governance (in market structure, politics, law, culture, organizational incentive structure and the like) what is striking, in the words of The Economist, is that, "such differences suggest that though each country's corporate governance may evolve in a similar direction, they will do so in distinct ways over many decades."

Conclusion and Some New Directions

From this review of emerging forms of monitoring and relationship investing we conclude that while the terms of ownership and control are again strongly contested primarily between institutional owner/agents and manager/agents, there is often little agreement on the forms of monitoring, on the nature of monitoring, and on how to evaluate the relative success or failure of monitoring. Additionally, there are also a number of topics which have yet to be addressed, in particular monitoring when an institution owns a broadly diversified portfolio that, effectively, makes it an owner of the entire economy. We conclude with a brief section discussing these issues.

Monitoring What and to What End?

Many observers and institutional activists (e.g. Black, Pound, Koppes and CalPERS) have pointed out that there are economies of scale to monitoring which overcome the free rider problem, making the cost and benefits of monitoring underperforming firms weigh in on the benefit side. This has typically been measured by an analysis of the targeted firms using various forms of event studies, although conclusions have differed widely. Yet Koppes and Black, for example, among many others, have simultaneously argued that active monitoring 'keeps the herd moving' at a rate that (ceteris paribus) it would not otherwise have. In short, it is claimed that the market and the economy as a whole are positively affected by monitoring the underperformers. It is not clear how one would substantiate such a broad claim, although it is a plausible hypothesis. Nor is it clear (and to our knowledge not discussed) that the most effective way to affect the market as a whole is to target the underperformers, or only target them, especially since there are many possible forms of monitoring and relationship investing.

Kester, Porter, and Roe all argue that economic performance and US competitiveness would be improved if (in addition to what we can call financial monitoring) various forms of contractual relationship investing were developed and formalized into an expansion of the 'nexus of contracts' of the various vendor/supplier relations that currently exist. What the relation between these types of monitoring might be, and what might be learned from the German and Japanese experiences in particular with industrial groups or 'alliance capitalism', is not widely discussed in the literature. But an alliance capitalism of industrial groups, albeit however informal, may well be a very different route to increased economic performance than financial monitoring and relationship investing by financial fiduciaries as intermediaries.

Universal Owners: Competition and Externalities

There is an additional issue, raised as far as we know only by Monks: that is the problem of what he calls the 'universal owner.' A universal owner is typically a large, broadly diversified pension fund (although some mutual funds fit) which owns a cross section of the largest and medium-sized publicly traded firms. As such it effectively owns a small proportion of the entire economy. This raises at least two important questions. The first is competitive. What is the effect, for example, of targeting Westinghouse which recently purchased CBS to effectively compete with, for example, NBC, when the institution owns large stakes in both General Electric, the parent of NBC, and Westinghouse, the parent of CBS? In particular how involved with developing company strategy should institutions be, especially as many institutional investors were involved in advising Westinghouse between 1992 and at least 1994?

The second question universal ownership raises is perhaps more profound than the competitive aspect. It is what Monks terms a collective choice problem: what is good for the group as a whole may not be perceived as being good for the individual firm, citizen, or consumer. Since pension funds (and indirectly, mutual funds) are the long-term fiduciaries for a very large proportion of the population , to what degree should and can they be concerned with those issues that are externalities to the individual firm, but public goods for the society as a whole, e.g. vocational and other education, pollution and energy conservation, occupational health and safety. With some minor exceptions these issues have been addressed neither in the scholarly literature nor in practice.

The dramatic growth of various forms of institutional voice in the last five years, and of the scholarly analytic and advocacy literature concerned with it, has major public policy implications for federal and state deregulation in certain areas and increased regulation in others. In addition the private decisions and changes made by corporations and institutional investors may have significant social and economic consequences which raise new public policy issues, or change the consequences of existing public policy made under often dramatically different conditions. Part VI considers some of the most important aspects of these developments.