1.3.The Board of Drector Many outsi e larg U.5. its are muc g Lin the c .The the0mo时ankp rs toc the United States But.four fac ship is ce al h to th he hoa 's ady do re well-kno
1.3. The Board of Directors 31 Many outside directors in the largest U.S. corporations are CEOs of other firms. Besides having a full workload in their own company, they may sit on a large number of boards. In such circumstances, they may come to board meetings (other than their own corporation’s) unprepared and they may rely entirely on the (selective) information disclosed by the firm’s management. Insufficient incentives. Directors’ compensation has traditionally consisted for the most part of fees and perks. There has often been a weak link between firm performance and directors’ compensation, although there is a trend in the United States towards increasing compensation in the form of stock options for directors.58 Explicit compensation is, of course, only part of the directors’ monetary incentives. They may be sued by shareholders (say, through a class-action suit in the United States). But, four factors mitigate the effectiveness of liability suits. First, while courts penalize extreme forms of moral hazard such as fraud, they are much more reluctant to engage in business judgements about, say, whether an investment or an acquisition ex ante made good economic sense. Judges are not professional managers and they have limited knowledge of past industry conditions. They therefore do not want to be drawn into telling managers and directors how they should run their companies. Since corporate charters almost always eliminate director liability for breaches of duty of care, it is difficult for shareholders and other stakeholders to bring a suit against board members. Second, firms routinely buy liability insurance for their directors.59 Third, liabilities, if any, are often paid by the firms, which indemnify directors who have acted in good faith. Fourth, plaintiff’s lawyers may be inclined to buy off directors (unless they are 58. Yermack (2004b), looking at 766 outside directors in Fortune 500 firms between 1994 and 1996, estimates incentives from compensation, replacement, and opportunity to obtain other directorships. He finds that these incentives together yield 11 cents per $1,000 increase in firm value (shareholder wealth) to an outside director. Thus, performance-based incentives are not negligible for outside directors even though they remain much lower than those for CEOs (e.g., $5.29 per $1,000 increase in firm value for the median CEO in 1994, as reported by Hall and Liebman (1998)). 59. As well as officers (these insurance policies are labeled directors and officers (D&O) insurance policies). extremely wealthy) in order to settle. Overall, for Black et al. (2004), as long as outside directors refrain from enriching themselves at the expense of the company, the risk of having to pay damages or legal fees out of their own pocket is very small in the United States,60 as well as in other countries such as France, Germany, or Japan, where lawsuits are much rarer. This undoing of the impact of liability suits has two perverse effects: it makes directors less accountable, and, in the case of indemnification by the firm, it deters shareholders from suing the directors since the fine paid in the case of a successful suit comes partly out of their pocket. Avoidance of conflict. Except when it comes to firing management, it is hard even for independent directors to confront management; for, they are engaged in an ongoing relationship with top executives. A conflictual relationship is certainly unpleasant. And, perhaps more fundamentally, such a relationship is conducive neither to the management’s listening to the board’s advice nor to the disclosure to the board of key information. In view of these considerations, it may come as a surprise that boards have any effectiveness. Boards actually do interfere in some decisions. They do remove underperforming managers, as we discussed in Section 1.2. They may also refuse to side with management during takeover contests. A well-known case in point is the 1989 RJR Nabisco leveraged buyout (LBO) in which a group headed by the CEO made an initial bid and the outside directors insisted on auctioning off the company, resulting in a much more attractive purchase by an outsider. It should be realized, though, that the cosy relationship between directors and management is likely to break down mainly during crises. Directors 60. It was a shock to directors when ten former executive directors of WorldCom agreed to pay a total of $18 million from their own savings and ten former Enron directors paid $13 million (still, the insurance companies are expected to pay out the bulk of the money: $36 million for WorldCom and $155 million for Enron The Economist, January 15, 2005, p. 65). It is hard to predict whether this indicates a new trend, as these cases involved extreme misbehaviors. D&O insurance policies are less prevalent in Europe because of the lower probability of lawsuits, but they are likely to become very widespread as lawsuits become more common
32 1.32月 ng the B o the b As we observed,bou bat the top to be to be g
32 1. Corporate Governance are then more worried about liability and more exposed to the spotlight. Furthermore, their relationship with management has shorter prospects than during good times. And, indeed, directors have historically been less effective in preventing management from engaging in wasteful diversification or in forcing it to disgorge excess cash than in removing underperforming managers. Relatedly, there is evidence that decreases in the share price lead to an increase in board activity, as measured by the annual number of board meetings (Vafeas 1999). Bebchuk and Fried (2004) offer a scathing view of board behavior. They argue that most directors choose to collude with CEOs rather than accomplish their role of guardian of shareholders’ interests. Directors dislike haggling with or being “disloyal” to the CEO, have little time to intervene, and further receive a number of favors from the CEO: the CEO can place them on the company’s slate, increasing seriously their chance of reelection, give them perks, business deals (perhaps after they have been nominated on the board, so that they are formally “independent”), extra compensation on top of the director fee, and charitable contributions to nonprofit organizations headed by directors, or reciprocate the lenient oversight in case of interlocking directorates. A key argument of Bebchuk and Fried’s book is that the rents secured by directors for the CEO involve substantial “camouflage”; that is, these rents should be as discrete or complex as possible so as to limit “outrage costs” and backlash. This camouflage yields inefficient compensation for officers. For example, compensation committees61 fail to filter out stock price rises or general market trends and use conventional stock-option plans (as discussed in Section 1.2); and they grant substantial ability to managers to unload their options and shares. They also grant large cash payments in the case of an acquisition, generous retirement programs, and follow-on consulting contracts. Directors also happily acquiesce to takeover defenses.62 61. Despite their independence (in the United States, and unlike for some other committees, such as the nomination committee, directors sitting on the compensation committee are mostly independent directors). 62. Another example of “camouflaged rent” is the granting of executive loans, now prohibited by the 2002 Sarbanes–Oxley Act. 1.3.2 Reforming the Board The previous description of indolent boards almost smacks of conspiracy theory. Managers carefully recommend for board nomination individuals who either have conflicts of interest or are overcommitted enough that they will be forced to rubberstamp the management’s proposals at the board meetings. And managers try to remove incentives to monitor by giving directors performance-insensitive compensation and by insuring them against liability suits, and “bribe” them in the various ways described in Bebchuk and Fried’s book. Most of these managerial moves must, of course, be approved by the board itself, but board members may find their own bene- fit to colluding with management at the expense of shareholders. While there is obviously some truth in this description, things are actually more complex for a couple of reasons. Teammates or referees? As we observed, board members may actually be in an uncomfortable situation in which they attempt to cooperate with top executives while interfering with their decisions. Such relationships are necessarily strenuous. These different functions may sometimes conflict. The advisory role requires the directors be supplied with information that the top management may be unwilling to disclose if this information is also used to monitor and interfere with management.63 Knowledge versus independence? Parties close to the firm, and therefore susceptible to conflict of interest, are also likely to be the best informed about the firm and its environment. Similarly, professional managers are likely to be good monitors of their peers, even though they have an undue tendency to identify with the monitored. What link from performance to board compensation? Providing directors with stock options rather than fixed fees goes in the right direction, but, for the same reasons as for managers, stock options have their own limitations. In particular, if managers go for a risky strategy that reduces investor value but 63. Adams and Ferreira (2003) build a model of board composition based on this premise and show that, in some circumstances, a management-friendly board may be optimal
lance of US.co shy the wwh a few CalPER 1978 pres ake of the
1.3. The Board of Directors 33 raises the value of their stock options, directors may have little incentive to oppose the move if they themselves are endowed with stock options. Similarly, directors’ exposure to liability suits has costs. While the current system of liability insurance clearly impairs incentives, exposing directors fully to liability suits could easily induce them to behave in a very conservative fashion or (for the most talented ones) to turn down directorial jobs. With these caveats in mind, there is still ample scope for board reform. Save a few legal and regulatory rules (such as the 1978 New York Stock Exchange rule that listed firms must have audit committees made up of nonexecutives), directors and managers faced few constraints in the composition and governance of boards. New regulations and laws may help in this respect, but, as usual, one must ask whether government intervention is warranted; in particular, one should wonder why the corporate charter designers do not themselves draw better rules for their boards, and, relatedly, why more decentralized solutions cannot be found, in which shareholders force (provided they have the means to) boards to behave better. That is, with better information of and coordination among shareholders, capital market pressure may be sufficient to move boards in the right direction. In this spirit, several study groups produced codes of good conduct or of best practice for boards (e.g., the 1992 Cadbury report in the United Kingdom and the 1995 Viénot report in France). Abstracts from the Cadbury report are reproduced at the end of this chapter. Among other proposals, the Cadbury report calls for (a) the nomination of a recognized senior outside member where the chairman of the board is the CEO,64 (b) a procedure for directors to take independent professional advice at the company’s expense, (c) a majority of independent directors (namely, nonexecutive directors free from business relationship with the firm), and (d) a compensation committee dominated by nonexecutive directors and an audit committee conferred to nonexecutive directors, most of whom should be independent. In 64. The UK Combined Code (the successor to the Cadbury Code) states that chairmen should be independent at the time of appointment. Table 1.1 Compliance of U.S. companies with a few CalPERS criteria in 1997. Source: Analysis by the The New York Times (August 3, 1997) of data compiled by Directorship from the 861 public companies on the Fortune 1000 list. “Independent” here means “composed of outside directors.” Has outside chairman 5% Only one insider on the board 18% Some form of mandatory retirement for directors 18% Independent nominating committee 38% Fewer than 10% of directors over 70 68% Independent governance committee 68% No retired chief executive on the board 82% Independent ethics committee 85% Independent audit committee 86% A majority of outside directors on the board 90% Independent compensation committee 91% contrast, the Cadbury report recommends against performance-based compensation of directors. In the United States, the largest public pension fund, CalPERS, with $165.3 billion in assets in August 2004, drew in the mid 1990s a more ambitious list of 37 principles of good practice for a corporate board, 23 “fundamental” and 14 “ideal.” CalPERS would like the companies to consider the ideal principles, such as a limit on the number of directors older than 70, but has stated it would be more openminded on these principles than on the fundamental ones. CalPERS monitors the companies’ compliance (in spirit, if not the letter) with these principles and publicizes the results, so as to generate proxy votes for companies that comply least. As of 1997, most firms failed to comply with a substantial number of CalPERS criteria, although some of these criteria were usually satisfied by most corporations (see Table 1.1). While the CalPERS list is stringent and some of its criteria controversial, it illustrates well the investors’ current pressure for more accountable boards. More recently, in the wake of the many corporate scandals at the turn of the century, expert recommendations regarding the board of directors have been bolder. For example, they suggest regular meetings of the board or specific committees in the absence of executives, a policy already adopted by a
tors'co for strict ct20 First,co olely p And
34 1. Corporate Governance number of corporations.65 Such meetings promote truth telling and reduce individual directors’ concern about the avoidance of conflict with management. A number of experts have also recommended self-evaluation of boards; for example, at regular intervals the director with the worst “grade” would be fired.66 There have also been calls for strict limits (e.g., three) on the number of board mandates that a director can accept, for limited director tenures, and for a mandatory retirement age. Monetary incentives have also been put forward. The directors’ compensation would be more systematically related to the firm’s stock value. Here the recommendation is for directors to hold a minimum number of shares in the firm.67 Some experts68 have proposed a direct or intermediated (through an ombudsman) access of whistleblowers to independent directors. This is probably a good suggestion, although it has one flaw and its impact is likely to be limited for two reasons. The drawback of whistleblowing is that companies react to its threat by (a) intensively screening employees in order to pick those who are likely to prove “loyal,” and (b) reducing information flows within the firm, which reduces the benefit of whistleblowing in terms of transparency and accountability.69 Second, employees have relatively low incentives to blow the whistle. If discovered by the company (even formal anonymity does not guarantee that there will not be suspicion about the source of information), they will probably be fired. And whistleblowers notoriously have a hard time finding a new job in other firms, who fear that they will blow the whistle again.70 65. Korn/Ferry International (2003) estimated that in 2003 87% of U.S. Fortune 1000 boards held Executive Sessions without their CEO present. By contrast, only 4% of Japanese boards gather without the CEO present. 66. In 2003, 29% of U.S. boards (41% in Asia Pacific) conducted individual director evaluation reviews (Korn/Ferry International 2003). 67. An example often cited by the proponents of this view is that of G. Wilson, who was for twelve years director of the Disney Corporation and held no share of Disney despite a personal wealth exceeding $500 million! 68. See, for example, Getting Governance Right, McKinsey Quarterly, 2002. 69. More generally, a cost of using informers is that it destroys trust in social groups, as has been observed in totalitarian regimes (e.g., in Eastern Germany, where people were concerned that family members or friends would report them to the Stasi). 70. Consider the example of Christine Casey, who blew the whistle on Mattel, the toy manufacturer, which reported very inflated sales In particular, employers routinely check prospective employees’ litigation record. The proposal of letting whistleblowers have a direct or indirect access to independent directors is therefore likely to be most effective when (a) the sensitive information is held by a number of employees, so that whistleblower anonymity can really be preserved, and (b) the directors can check the veracity of the information independently, that is, without resorting to the whistleblower. Lastly, it must be the case that directors pay attention to the information that they receive from the whistleblower (the Enron board failed to follow up on allegations by a whistleblower). For this, they must not be swamped by tons of frivolous whistleblowing messages; and, of course, they must have incentives to exercise their corporate governance rights. Lastly, the Sarbanes–Oxley Act (2002) in the United States requires the audit committee to hire the outside auditor and to be composed only of directors who have no financial dealing with the firm. It also makes the board more accountable for misreporting. A Few Final Comments Scope of codes. First, codes are not solely preoccupied with boards of directors. They also include, for example, recommendations regarding reporting (auditor governance, financial reporting), executive forecasts to its shareholders (see, for example, The Economist, January 18, 2003, p. 60). Some managers kept two sets of figures, and consistently misled investors. In February 1999, Ms. Casey approached a Mattel director. After being screamed at by executives and basically demoted, in September 1999, she telephoned the SEC. She ended up resigning, filed an unsuccessful lawsuit against Mattel, and in 2003 was still without a job. Zingales (2004) reviews the (rather bleak) evidence on what happens to whistleblowers after they have denounced management and after they quit their firm. To counteract the strong incentives not to blow the whistle, he proposes that whistleblowers receive a fraction (say, 10%) of all fees and legal awards imposed on the company (with, of course, some punishments for frivolous whistleblowing and a requirement to denounce to the SEC rather than in public). Such rewards already exists for people who help the U.S. government to recover fraudulent gains by private agents at its expense (whistleblowers are entitled to between 15% and 30%). Friebel and Guriev (2004) argue that internal incentives are designed so as to limit whistleblowing. In their theoretical model, division managers may have evidence that top managers are inflating earnings. Top management, however, provides lower-level managers with a pay structure similar to theirs so as to make them allies. Friebel and Guriev thus provide an explanation for the propagation of short-term incentives in corporate hierarchies
Table 1.2 Some recent codes of good goverano de n Yes es ar often us the ris
1.3. The Board of Directors 35 Table 1.2 Some recent codes of good governance. Separation of Rotation Frequency ‘Comply Selected Independent chairman–CEO of external of financial or explain’ country-specific directors? roles? auditor? reporting? requirement? governance issues Brazil CVM Code As many Clear Not Quarterly No Adoption of (2002) as possible preference covered IAS/U.S. GAAP1 for split Fiscal boards1 Tag-along rights1 France Bouton Report At least No recom- Regularly, for No recom- No Dual statutory (2002) one-half mendation lead auditors mendation auditors of board given Russia CG Code At least Split required Not covered Quarterly No Managerial boards (2002) one-quarter by law of board Singapore CG Committee At least Recommended Not covered Quarterly Yes Disclosure of pay for (2001) one-third family members of of board directors/CEOs United Kingdom Cadbury Code Majority of Recommended Periodically, Semiannually Yes (1992) nonexecutive for lead directors auditors Combined Code At least Clear Not covered2 Semiannually, Yes (2003) one-half preference per listing of board for split rules United States Conference Board Substantial Separation is Recommended Quarterly, No (2003) majority one of three for audit firm3 as required of board acceptable options by law Source: Coombes and Wong (2004). 1. IAS, International Accounting Standards; GAAP, generally accepted accounting principles; fiscal boards are akin to audit committees, but members are appointed by shareholders; tag-along rights protect minority shareholders by giving them the right to participate in transactions between large shareholders and third parties. 2. In the United Kingdom, the accounting profession’s self-regulatory body requires rotation of lead audit partner every seven years. Combined Code recommends that companies annually determine auditor’s policy on partner rotation. 3. Sarbanes–Oxley Act requires rotation of lead audit partner every five years. Circumstances that warrant changing auditor firm include audit relationship in excess of ten years, former partner of audit firm employed by company, and provision of significant nonaudit services. compensation, shareholders voting, or antitakeover defenses. Second, they are now commonplace. As of 2004, fifty countries had their own code of governance, emanating from regulators, investor associations, the industry itself, or supranational organizations. They differ across countries as shown by Table 1.2, which reports some key features of a few recently drawn codes. Do codes matter? Codes are only recommendations and have no binding character. Probably the main reason why they seem to have an impact is that they educate the general public, including investors. To the extent that they are drawn by expert and independent bodies they carry (real) authority in indicating the conditions that are conducive to ef- ficient governance. They further focus the debate on pointing at some “reasonable” or “normal” practices, a deviation from which ought to be explained. For example, it is often asserted that the 1992 Cadbury Code of Best Practice, by pointing at the cost of con- flating the positions of chairman of the board and CEO, was instrumental in moving the fraction of the top U.K. companies that operated a separation from 50 to 95% in 2004. In performing this educative role, the codes finally may help the corresponding practices enjoy the “network externalities” inherent in familiar institutions: investors, judges, and regulators in charge of enforcing the laws gain expertise in the understanding of the meaning and implications of most often used charters; contractual deviations by individual firms therefore run the risk of facing a lack of familiarity by these parties