SZW/RSDA 2/2011 167
Regulation of Say on Pay: Engineering Incentives for Executives and Directors Experiences from the United States and Implications for Regulation in Switzerland Dr. oec. HSG et lie. iur. Lukas Miller, MA UZH
The debate about the compensation of executives effect, this could lead to weaker bargaining power of and directors is a discussion about incentives and directors in relation to executives when they have to agency costs. This article analyzes basic tools to contract jbr new employment agreements with their reduce agency costs and also assesses the ongoing executives. Moreover; shareholders often do not have debate about the future regulation of the compensa- the time or ability to process complex disclosure tion of executives and directors. It draws upon leg- about executive compensation. This will lead to unin- islative experience from the United States. Recently formed voting behavior of rationally apathetic share- proposed legislation in Switzerland attempts to em- holders. Additionally, some shareholders, e.g., insti- power shareholders with the redraft of the applicable tutional investors, may prefer to stay on good terms section of the Swiss Code of Obligations (CO). The with the CEO or directors because they want to have main motivation behind this draft law is the reduction a good long-term relationship with the board and the of excessive executive compensation. Directors and executives. This article advises against implementing shareholders with a higher degree of independence a specific salary cap for so called "very high com- might be less conflicted in their decisions but they pensation "and also advises against the implementa- might also have a lack of firm-specific know-how. In tion of tax burdens for executive compensation.
Table of contents public in 2002. Various other scandals have been discussed since then in the tabloid press. Many di- 1. Public Outrage Causing Regulatory Intervention rectors or executives have paid themselves sums that II. Managerial Power and Agency Costs have caused public outrage , 2 and this has triggered 1. Principal Agent Problem regulatory activities such as the upcoming Swiss ref- 2. Other Factors Influencing Agency Costs erendum "against rip-off salaries"' and the related 3. Remedies against the Principal Agent Problem 3.1 Monitoring regulatory debates in the Swiss parliament.' Propo- 3.1.1 Independent Directors: Golf Club Syndrome vs. Expertise 3.1.2 Shareholders as Monitors of the Com- LL.M. candidate, 2011, Columbia Law School; Postdoc- pensation of Executives and Directors toral Researcher at the Faculty of Law of the University of 3.1.3 Shareholder Approval of Say on Pay Zurich. The author would like to thank Angela K. Chen for 3.2 Incentive Alignment her valuable comments and suggestions in preparing drafts 3.3 Compensation as a Screening Device vs. of this Article. The views and opinions expressed in this Anchoring Article are the author's own. III. Compensation Arrangements in Mergers and See Beat Gygi, Managerldhne als Privatsache, Einstellung Acquisitions der Strafuntersuchung gegen ABB-Pension re Barnevik 1. Say on Golden Parachute Agreements und Lindahi, Neue ZOrcher Zeitung (October 6, 2005), 21. 2 See Lucian A. Bebchuk/Yaniv Grins fein, The Growth of 2. Extend Disclosure to a Broader Group Executive Pay, Oxford Review of Economic Policy, 21 of Individuals (2005) 300-302. IV. Unintended Incentive Effects of Taxation See e.g.
Electronic copy available at: http://ssrn.com/abstract=1881176 pp
Mulle':Regulation of Say on Pay: Engineering Incentives for Executives and Directors SZW/RSDA 2/2011 168
nents of these regulatory movements demand say on the shareholders.' In order to examine the proposed pay for shareholders regarding the compensation for changes to the pay-setting process, it is helpful to directors and executives. In the U.S., say on pay or look at the experience in the U.S., where a similarly more broadly: Accountability and Executive Com- complex set of problems exists. The main assump- pensation - is being addressed by §§ 951 to 956 of tion of this paper is that directors and executives in the Dodd-Frank Wall Street Reform and Consumer a company are incentive-driven. 7 Based on the eco-
Protection Act (hereinafter "Dodd-Frank Act") . 5 This nomic discussion of legal issues, this article provides issue has become even more prominent since the suggestions for future regulation. bailouts of giant banks all over the world carried out to prevent an imminent meltdown of the global finan- cial system. II. Managerial Power and Agency Costs As of today, the compensation of executives and directors is solely a duty of the directors of Swiss 1. Principal Agent Problem companies limited by shares. The proposed initia- In the typical model, the pay-setting process for tive wants to shift this duty at least partially to executives is basically influenced by the top execu- tives' managerial power in a company. Economists assume that setting managerial pay is the product of Aktienrechtsrcvision, Fine Standortbestimrnung per Ende arm's length contract negotiations. Such a pay-setting 2010, Zurich/St.Gallcn 2010, 247-283; Hans- Ce/i Vogt/ process, would, according to theory, always produce Emanuel Se/i iwow/Karin J'Viedmnei; Die Aktienrechtsrevi- an efficient outcome. 8 A better model for the setup sion unter Corporate-Governance-Aspekten, Bestandes- aufnahme, einige Auslegungsversuchc and ciii paar wich- of negotiations describes the relationship between the tige Anliegen vor der Erstberatung im Nationairat, AJP 18 (2009) 1375-1388. Ralph Malacrida/Til/ Spillmann, Pay- back Time?, GesKR 3 (2008) 345-358. For an overview of the recent debates and legislative actions see Hansueli See Art. 716 of the Swiss Code of Obligations (CO); Peter Sc/,Jchli, Die Karten der "Abzocker"-Debatte werden neu &5c/di, Schweizer Aktienrecht, 48 Ed., Zurich/Basel/Ge- gemischt, Neue Zfircher Zeitung (February 25, 2011), 31; neva 2009, § 12 N 237-262; Daniel Daenikem; Vergfitung Hermann Biirgi, Indirekter Gegenentwurf zur Volksinitia- von Verwaltungsrat und Gesch ftsleitung schweizerischer tive "Gegen die Abzockerei", Langes Seilziehen f r Publikumsgesellschaften, Eine Bestandesaufiiahme aus mehrheitsfiihige Losung, ST 85 (2011) 50-51 and Roil rechtsvergleichender Sicht, SJZ 101 (2005) 384-388; Pa- Sethe/Moriiz Seile,; Entwicklungen im Gesellschaftsrecht te, Forsimoser/Arthur Meier-Hayoz/Peter Nobel, Schwei - und im Wertpapierrecht/Le point sur Ic droit des sociétés zerischcs Aktienrecht, Bern 1996; § 28 N 121; Gui; M. et des papier-valeurs, SJZ 106 (2010) 510-511. Similarly Gachte,; Managementvergutungen: Grundlagen, Kompe- in the U.S.: William P Allen/Reinier Kraa/anan/Guhan tcnzen und Verthhren, Zurich 2008. However, Art. 29 of Subramanian, Commentaries and Cases on the Law of Bu- the Swiss Stock Exchange and Securities Trading Act siness Organization, 3d Ed., New York 2009, 335-338. (SSETA; B rsengesetz, BEHO) is limiting the power of The regulation of say on pay has parts of its roots in the the board directors to grant certain compensation; see in- United Kingdom. This paper discusses only evidence from fra Part 111.1; see generally Rudolf Tsr/ian//Jacques J//L the U.S. For the situation in the U.K., see e.g. generally: lane//Hans-Jakob Diem, in: Basler Konimentar zum Brian R. C/se/fins/Randall S. Thomas, Should Share- B rsengesetz/Finanzmarktaufsichtsgesetz, Rolf Watter/ holders Have a Greater Say over Executive Pay: Learning Nedim Peter Vogt (Eds.), 2 Ed., Basel 2011, Art. 29 N 24 from the US Experience, Journal of Corporation Law BEHG [hereinafter: BSK BEHG]; Rudolf Tschdni/Jacques Studies 1 (2001) 277-316; Joseph A. McCahery/Luc Rca- Iffiand/Hans-Jakob Diem, Offentliche Kaufangebote, neboog, Managerial Remuneration: The Indirect Pay-for- 2" Ed., Zurich/Basel/Geneva 2010, N 541-542 and 747. Performance Relation, Journal of Corporate Law Stud- An unofficial translation of the SSETA can be found here: ies 1(2001)317-332; CynthiaA. Williams/Jo/in M. Conley
Electronic copy available at: http://ssrn.com/abstract=1881176 SZW/RSDA 2/2011 Muller: Regulation of Say on Pay: Engineering Incentives for Executives and Directors 169
board of directors in corporations and the sharehold- The first tool, monitoring, is costly for the prin- ers as a principal agent problem.' The principal agent cipal (viz., the shareholder), and he may not be the problem describes a situation where the principal perfect monitor in every case. The agent usually has transfers the control of ownership to an agent. The better and more specialized abilities to serve in such principal cannot perfectly monitor the agent. Fur- a position than the shareholders do, i.e., he can run thermore, the principal wants to maximize the value the business better than the principal and can camou- of his company, whereas the agent primarily wishes flage some of his actions. In the past, payments were to maximize his own utility. Maximizing the value camouflaged using the Generally Accepted Account- of the company is only a secondary objective for ing Principles which allowed companies to issue the agent. Additional assumptions in this model are employee stock options without recognizing them that the principal is risk neutral and the agent is risk as expenses. 14 Another problem is that pension fund averse.'0 The principal can diversify his risk at rela- managers investing in pharmaceutical companies tively low costs with portfolio strategies on the stock normally do not know how to conduct research in market. An agent, however, cannot diversify his job pharmaceuticals (i.e., there is an information asym- position for practical reasons (e.g., being CEO at ten metry). In general, requiring full disclosure of all ma- companies at the same time). This conflict of incen- terial compensation elements is a necessary building tives and information asymmetry results in agency block of enhancing transparency and reducing agency costs. Agency costs consist of the following com- costs.'5 However, even with full disclosure, monitor- ponents: monitoring expenditures by the principal, ing is still a complex and costly undertaking.'' In bonding expenditures by the agent and the residual fact, even specialized lawyers and accountants are loss of the agency relationship. ' 1 In economic terms, not always able to interpret company disclosures cor- "compensation is excessive when it is greater than it rectly in every aspect,' 7 since tools necessary for in- would be if agency costs were zero." 2 terpretation, such as Black-Scholes option pricing,' 8 There are several reasons leading to excessive are rarely discussed comprehensively in the core cur- compensation of executives. Primarily, excessive riculum of law or business schools. compensation is a consequence of the transfer of control of the firm by the principal to the agent and the lack of transparency about the agent's actions and performance. The tools most discussed, therefore, in '' SeeBock/i (Fn. 6), § 8 N 251; Lukas Millie,; Eigenkapital- order to reduce agency costs are monitoring (i.e., dis- basierte Vergutung de lege ferenda - Rechtliche und be- closure) and incentive alignment." triebswirtschaftlichc Aspekte, AJP 17 (2008) 527, 530; Hans Caspar von der Crone/PascaiM. K/st/c,; Mitarbeiter- optionen und Rechnungslcgung, Fine Erfassung als Per- sonalaufwand drhngt sich auf, Neue ZOrcher Zeitung See generally Jensen/Meckling (Fn. 7), 305-360; but cf. (September 9, 2000), 28. Lipton/Savitt (Fn. 8), 749-752. 15 Sec Art. 663b" and Art. 663c Sec. 3 CO for the required ’ See Hans Caspar von der Crone, Risiko und Corporate disclosure in public corporations; see generally Ru/f Wat- Governance, in: Hans Caspar von der Crone/Peter Forst- ter/Karioi Maiza,; Transparenz der Vergiitungen rind Be- moser/Roif H. Weber/Roger Z ch (Eds.), Aktuelle Fragen teiligungen von Mitgliedern des Verwaltungsrates rind der des Bank- und Finanzmarktrechts, FS Dieter Zobi, Zurich Geschhftsleitung (Art. 663b"' rind 663c Abs. 3 OR), 2003, 555 and 560-561. GesKR 1 (2006) 349-359; see furthermore supra Fn. 4. See .Jensen/Meckling (Fn. 7), 308-309. See Michael C. Jensen/Kevin]. Murphy, CEO Incentives - 2 Richard A. Posner Are American CEOs Overpaid, and, if It's Not How Much You Pay, But How, Journal of Applied so, What if Anything Should be Done about it'?, Duke Law Corporate Finance 22(2010)69-70. Journal 58 (2009) 1015. 17 See Thomas Lee Hazen, Securities Regulation, Cases and 13 See e.g. Roif Watter/Karirn Maizar, Aktionhrsdemokra- Materials, 811 Ed., St. Paul 2009, 292-305; Posner (Fn. 12), tie - Uber erweiterte Zusthndigkeit der Generalversarnm- 1017. lung und Erleichterungen bei der StimmrechtsausQbung 8 See originally Fischer Black/Myron Scholes, The Pricing in schweizerischen Aktiengesellschaften, in: Peter Breit- of Options and Corporate Liabilities, Journal of Political schmid/ Wolfgang Portmann/Heinz Rey/Dieter Zobi Economy 81 (1973) 637-654 and Robert C. Merton, The- (Eds.), Grundfragen der juristischen Person, FS Hans Mi- ory of Rational Option Pricing, Bell Journal of Economics chael Riemer, Bern 2007, 411-415; Peter Forstmoser, and Management Science 4(1973)141-183. See John C. St rkung der Aktion rsdemokratie - aber wie?, Es solite Hull, Fundamentals of Futures and Options Markets, auch fiber neue Formen von Mitwirkungsrechten nachge- 7th Ed., chapter 13, for an easily understandable explana- dacht werden, Neue Z rcher Zeitung (July 29, 2010), 26. tion of this option pricing model. Pr
1 .M 1ler: Regulation of Say on Pay: Engineering Incentives for Executives and Directors SZW/RSDA 2/2011
The second alternative, incentive alignment, in- First, it is hard to find capable directors. Shareholders volves cleating incentives for the agent that are con- would have to assess and recruit possible CEO can- gruent with the interests of the principal. In public didates even though they may not know the company companies, this is usually done with equity-based very well from the inside. Second, once shareholders payment plans.' 9 However, one problem with equity have found a possible director, they have to nominate compensation plans is the phenomenon of windfall him in order to appoint him as a director. But the gains. What happens if the stock options of execu- fact that shareholders in a public company are dis- tives are only inflating like a bubble because of the persed makes it difficult to obtain sufficient numbers market or simply because central bankers are running of shareholders to take this action. 23 This creates a an expansive monetary policy? Should the manager collective action issue or a free rider problem. For a also get compensated if the company is underper- shareholder that only owns a handful or a few thou- forming in relation to the market in general (e.g., the sand shares, the benefits of being an activist share- market goes up 100% and the company only 20%)? holder and to engage in proxy fights can virtually If agency costs are a problem, windfall gains must never exceed the costs of this engagement. Thus, this not occur because they will undermine the effective- will lead to "rationally apathetic" shareholders. 24 In ness of incentive alignment plans. 2 Setting the right order to address this problem, proxy rules for share- incentives is difficult; e.g., if the principal pays the holder use ought to be made less costly so that the agent for increasing net profit by 1% a year, the agent benefits from shareholder activism may outweigh the only tries to raise net profit by 1% even if he could costs. It is necessary to make it easier for sharehold- have increased it by more than 1%. The rationale be- ers to gain a voice in company management; to this hind this behavior is that he tries to preserve profit end, it is essential that more meaningful disclosure increasing potential for the following years. Hence, of relevant information is provided. This would lower such incentive alignment plans do not work perfectly. the information collecting and processing costs for The better the monitoring system, the less need for shareholders 25 incentive alignment which could lead, if wrongfully
calibrated, to excessive compensation. 2 ' 3. Remedies against the Principal Agent Problem
2. Other Factors Influencing Agency Costs 3.1 Monitoring
Besides the solutions to the principal agent prob- 3.1.1 independent Directors: Golf Club Syndrome lem described above, the market for corporate control vs. Expertise can also discipline managers. The underlying idea of One problem in setting efficient compensation is this concept is that where managers perform badly; the so-called "golf club syndrome", i.e., it is hard to the stock price of a company will plunge. Hence, object to a compensation agreement for an executive this event can attract raiders which want to substi- tute management in order to rearrange the resources of a company. This improvement in management will 23 See John C. Cof'e, Liquidity Versus Control: The Institu- cause the stock price to increase, eventually making tional Investor as Corporate Monitor, Columbia Law Re- view 91 (1991) 1285. the company a profitable investment". This phenom- 24 See Stephen . Bainbridge, Executive Compensation: enon will have an effect on the agency costs. How- Who decides?, Texas Law Review 83 (2005) 1615, 1655; ever, in practice, shareholders rarely throw directors Stephen M. Bainbridge, Director Primacy and Shareholder out of a company. There are several reasons for this. Disempowerment, Harvard Law Review 119 (2006) 1751-1757; Jeffrey N. Gordon, The Mandatory Structure of Corporate Law, Columbia Law Review 89 (1989) 1576 and 1580; Roberta Romano, Answering the Wrong Ques- See generally MO//er (Fn. 14), 527-534. tion: The Tenuous Case for Mandatory Corporate Laws, 20 See Bebchuk/Fried (Fn. 8), 665-666; Posner (Fn. 12), Columbia Law Review 89 (1989) 1607; Ru/f Wutter/Katja 1013-1047. Roth Pel/unda, Shareholder Activism, in: Peter Leibfried/ 21 See Posner (Fn. 12), 1023; see also Jensen/Murphy Dirk Schafer (Eds.), 25 Jahre Unternehmertum, FS Gior- (Fn. 16), 66-67. gio Behr, Zurich 2010, 87-90. 22 See Henry G. Manne, Mergers and the Market for Corpo- 25 But cf. Arthur Meier_Hayoz/Peter Forstrnose,; Schweizeri_ rate Control, The Journal of Political Economy 73 (1965) sches GesellschaftSrecht, 10 0 Ed., Bern 2006, § 16 N 361 110-114. 363. SZW/RSDA 2/2011 1\’I ller: Regulation of Say on Pay: Engineering Incentives for Executives and Directors 171
with whom a director is friends. Directors seeking to tasks with less knowledge in a situation where the lower compensation must address a potentially thorny management has the agenda setting power. This subject, and in addition they may have to say "no" to makes it difficult to bargain and thus increases mana- a friend or a colleague. This is often difficult. 26 Apart gerial bargaining power. 29 from this psychological issue, in situations where a Given this issue, how can independent directors CEO disagrees with the lowering of his compensa- maintain their expertise? One solution is to utilize tion, he might influence the board in a way that the di- advisors."' However, the advisors themselves may rectors who are setting his pay will not be nominated have a conflict of interests. Hiring advisors for the again at the next annual shareholder meeting. The board of directors or the compensation committee CEO could, for instance, ask for an ultimatum where has its own problems. The board and the compensa- the other directors either have to decide whether they tion committee need advisors on which they can rely, will nominate somebody else as director, or look for thus they need advisors acting with objectivity." Take a new CEO. In such circumstances, replacing an un- compensation consulting firms as an example: these popular director is easier than substituting the CEO firms consult companies on their employees' pen- who is running the daily business. One way to avoid sion plans. Assignments will be awarded to the con- the golf club syndrome is by appointing independ- sulting firm by the CEO. Simultaneously, the same ent directors for the pay-setting process. In Swit- consulting company advises the board of directors zerland, independent directors usually set the com- or the compensation committee about the executive pensations of the inside directors and the top-level compensation of their executives. This is an actual executives through a remuneration or compensation conflict of interest which becomes even more evident committee. 27 It is a way to ease conflicts of interests if one compares revenues of these two types of con- in a board of directors. 28 But there is also a tension sulting businesses. Empirical evidence relating to the between expertise and independence. Strengthening 250 largest consulting companies in the U.S. shows the independence of directors will necessarily reduce that the executive compensation advising business is their firm-specific knowledge. As a result of this lack merely an insignificant share of the revenue of such of firm-specific expertise, independence in a context consulting companies. However, if such consulting where the directors have to fulfill difficult tasks such firms lose the pension plan advisory contracts, they as performance evaluation and pay-setting decisions, will go out of business. 32 If the legislature wants to may be a problem. Directors with limited time will overcome this issue and require more independence perform worse when they have to do more difficult of the consulting firms, one should keep in mind that the gain of independence will be paid with the cost of losing expertise relating to the firm and industry 26 See Charles M. Elson, Executive Overcompensation - specific know-how. If the consulting companies are A Board-Based Solution, Boston College Law Review 34 no longer able to provide advisory services for em- (1993) 974. 27 See Hans Caspar von der Crone/Benedict Burg, Sal r- ployee pension plans, they will lose firm-specific governance und Markt f r Fuhrungskr fte, 11.2. (forthcom- knowledge which they would have gained by provid- ing); von der Crone (Fn. 10), 554 and 557-561; Hans Cas- ing such services. As a result, they will be less able to par von der Crone, Arbeitsteilung im Verwaltungsrat, in: Charlotte M. Baer (Ed.), Verwaltungsrat und GeschSftsle,- tung, SSPHW, Vol. 76, Bern 2006, 80-82; G chter (Fn. 6), 128-135. In Delaware, compensation decisions of inde- pendent committees are protected by the business judg- 29 See Elson (Fn. 26), 976. ment rule; see e.g. In re Walt Disney Co. Derivative Litiga- 30 See generally Geoffrey C. Hazard/Edward B. Roe/c, A New tion, 906 A.2d 27 (Del. 2006); see also Smith v. Van Player in the Boardroom: The Emergence of the Indepen- Gorkorn, 488 A.2d 858 (Del. 1985); see also Elson dent Directors' Counsel, The Business Lawyer 59 (2004) (Fn. 26), 973. 1389-1392 and 1395-1412. 28 See Lukas Handschin, Treuepflicht des Verwaltungsrates See Elson (Fn. 26), 979. However, the advisors should be bei der gesellschaftsinternen Entscheidflndung, in: Hans careful to avoid personal liability; BGE 128 111 92; BGE Caspar von der Crone/Roger Z ch/Dieter ZobI (Eds.), 117 II 442; Meier-f-Iayoz/Forstmoser (Fn. 25), § 16 Neuere Tendenzen im Gesellschaftstrecht, FS Peter N 575-576. Forstmoser, Zurich 2003, 171-172 and 179-182; Hans 32 See Majority Staff of the United States House of Repre- Caspar von der Crone/Antonio Carbonara/Larissa Ma- sentatives Committee on Oversight and Government Re- ru/c/a Martinez, Corporate Governance und Fuhrungsor- form, Executive Pay: Conflicts of Interests Among Com- ganisation in der Aktiengesellschaft, SJZ 100 (2004) 409. pensation Consultants (2007), 4-5. 172 MOhler: Regulation of Say on Pay: Engineering incentives for Executives and Directors SZW/RSDA 2/2011
correctly assess the level of executive compensation tive compensation policy and the related contracts to that should be paid. A report of the U.S. Government shareholder meetings or road shows with important shows that there is a positive correlation between the investors. lack of independence of consultants and executive compensation. 33 Although the existence of correla- 3.1.2 Shareholders as Monitors of the Compen- tion does not necessarily mean that there is also cau- sation of Executives and Directors sation, such correlation is nonetheless troublesome. (i) Institutional Investors Another study, however, shows weaker empirical evi- The question of whether shareholders are good dence about the same issue. Hence, other variables monitors is important for making future regulation. could play into the whole pay-setting process: bigger In principle, shareholders can remove directors if firms will hire consulting firms and pay higher exec- they do not like their decisions; they may theoreti- utive compensation . > Cadman, Carter & flillegeist cally put their own nominees on the ballot. 38 How- took a larger sample in conducting their study, and ever, individually the shareholders might not be able their findings may therefore carry more weight in the to put directors on the ballot. The solution could be sense that the outcome could be less influenced by to form a coalition, for instance with mutual funds chance. Based on the empirical research behind their or hedge funds. Even though institutional investors study, the only significant variable which is positively do not have the resources to manage the company at correlated to the executive compensation is the firm a micro level, they can do it at a general level, i.e., size measured in terms of the natural logarithm of by preventing or discouraging undesirable actions the assets; hence, the bigger the amount of assets, the at annual shareholder meetings. 39 Since institutional higher the payments for the executives. 35 investors cannot engage in micromanagernent of the Because of the tension between independence and company, their only option is to strengthen the board expertise, a mandatory mix of inside and independ- in order to make it a stronger check upon manage- ent directors may be a good idea. This is also the so- ment. 48 The pension funds with their specialized in- lution which is regarded as best practice for Swiss vestor knowledge can better watch the companies and public corporations. 36 Finally, if boards hire compen- lower the cost of capital because agency costs will sation consultants, it may be useful in most cases for decrease with the monitoring activity. However, there the boards to bring the consultants to investor road are fundamental limits in the abilities of institutional shows. Through this, institutional investors can en- investors which make them unsuitable as monitors. gage in better discussion about the important issues Some institutional investors, like the management, of executive compensation and can have their ques- need quick returns and necessarily have short term tions properly answered. 37 Thus, it is recommended incentives; otherwise the fund managers will not get that directors and executives of Swiss public corpora- paid well and will lose their jobs. Thus, at the level tions bring the consultants who designed the execu- of the institutional investors, the same principal agent problem can occur.4! coffee makes the argument that See Committee on Oversight and Government Reform institutional investors want to have a certain level Report (Fn. 32), 6. of liquidity in order to prevent a bank-run situation See Brian D. Cadman/Mary Ellen Carter/Stephen A. Hi!- when investors all want their money back at once. legeist, The incentives of Compensation Consultants and Because of this precautionary measure, the institu- CEO Pay (February 1, 2009), Journal of Accounting and Economics (forthcoming), available at SSRN:
being able to sell, e.g., their 30% ownership in a pub- tion and documents of the corporation. The members lic company. At the first sign of trouble in a company, of the shareholder committee would have the same fi- the institutional investor has the incentive to liquidate duciary duties (and in particular the duties to observe his investment in order to prevent losses or illiquidity, secrecy) as the members of the board of directors. rather than exercise votes in order to try and improve The second model the "owner model" - is basically the company's situation. 42 However, Cofièe ration- the bundling of the shareholder voice and the empow- ale does not seem persuasive because institutional erment of majority shareholders. It would serve as investors can afford to be less liquid since some of an institutionalized way of communication between them can also have long term goals. Therefore, they the directors and majority shareholders. This model are able to choose the voice-strategy instead of exit- is more or less what the best practice in public cor- ing. Even if the liquidity argument does apply, insti- porations should be because directors should know tutional investors can bolster their liquidity quite eas- what their majority shareholders think. 45 However, ily, for example by borrowing funds. Generally, over one has to be cautious when using these two models. the last years, hedge funds have been able to take over Shareholders in such a committee might indeed have companies with very low use of equity capital. They higher degree of independence than the directors but can buy or lend a share and hedge it. Thus, institu- they could also have a lack of expertise despite po- tional investors will be able to vote with little value at tentially greater access to information. Appointing risk if they are correctly hedged. 43 shareholders with expert knowledge could ease the problem regarding know-how, but these committee (ii) Shareholders' Committee members would still lack firm-specific expertise. 46 Furthermore, even shareholders with expertise can Forstinoser/Hostettler/Vogt discuss the introduc- have the same conflict of interests as the managers tion of shareholder committees as a possibly more of a corporation; for instance, certain institutional efficient monitoring device vis- -vis the board of di- shareholders may wish to be allies of the manage- rectors. 44 They have identified two alternative models ment. 47 They might have a particularly strong incen- which could act as a possible missing link between tive to be friendly to executives when the corporation the annual shareholder meeting and the board of di- uses a pension plan offered by the same institutional rectors. In their first model, the "executive model", investor. In such a case, the benefits of becoming a a few appointed shareholders would form a board shareholder activist are lower than the costs which similar to the board of directors. Majority sharehold- are incurred as a consequence of contesting man- ers, which do not need to be (or are not) directors, agement and losing profitable pension plan assign- would represent the shareholders in a committee. ments from the company. Another reason why the They would have privileged access to the informa- monitoring ability of institutional investors may be diminished is that there is no benefit in acting indi- 42 See Coftde (Fn. 23), 1329-1330. Co/fee uses the semi- vidually as an activist shareholder while free riders strong version of the Efficient Capital Market Hypothesis, can gain profits out of this individual activism this i.e., the market price of a share reflects all publicly avail- able information; nonpublic available information is not included in the share price; see furthermore Ronald .1 Gil- son/Rainier H. Kraak,nan, The Mechanisms of Market Ef- Id. ficiency, Virginia Law Review 70 (1984) 555-567. See Jeffrey N. Gordon, The Rise of Independent Directors, See Marcel Kahan/Edward B. Rock, Hedge Funds in Cor- Stanford Law Review 59 (2007) 1490-1500. Gordon dis- porate Governance and Corporate Control, University of cusses the tension between independence and expertise in Pennsylvania Law Review 155 (2007) 1028-1056 and the board of directors. The same considerations can also 1062-1069; however, letting hedge funds act as activist apply to shareholder committees which work in a similar shareholders can also lead to short-term-thinking since way as hoards of directors. However, shareholder commit- they sometimes maximize short-term gains. tees would probably participate in a dialogue about less a See Peter Forstnioser/Stephan Hostett/er/Hans-Ueli Vogt, complicated matters, but cf. Forstnioser (Fn. 13), 26. Aktion rsaussch sse als mhgliche Neuerung in der AG, A similar problem could occur if one appoints employees Vorschlag f r eine eigenthmerorientierte Belebung und to such a committee because employees may want to keep Verbesserung der Aktion rsdernokratie, Neue Z rcher good relations with their bosses; but cf. Jonathan Michie/ Zeitung (January 27, 2011), 31; Forstmoser (Fn. 13), 26. Christine Gogh ton, Employee Participation and Owner- See also Wafter/Roth Pellanda (Fn. 24), 91-92 mentioning ship Rights, Journal of Corporate Law Studies 2 (2002) the institutional shareholders' committee. 150-152. Muller: Regulation of Say on Pay: Engineering Incentives for Executives and Directors SZW/RSDA 2/2011 174
is similar to the collective action problem discussed man and the two second highest paid executives."' above in relation to non-institutional shareholders. The rationale behind this choice of disclosure is that Furthermore, not all institutional investors have the the compensation of top-level executives is the most same incentives; e.g., some government controlled important issue. A possible loophole is that it lies in mutual or pension funds can have political incenti- the discretionary power of the board of directors to ves. Governmental control can result in manipulating define which employees qualify as executives . 5 ' To il- management in troubled companies solely to prevent lustrate: at Citigroup, commodity trader Andrew Hall the public lay-off of many employees which could bet on changes of the price of oil while it was being politically disadvantage the people behind funds ope- shipped. As soon as the oil tankers entered the wa- rated by government officials. 48 ters of the U.S., the worth of the oil increased greatly. Hall made a $100 million bonus in just one year. 52 (iii) Information Processing Costs People like Hall are not included in the CD&A; nei- In general, it is difficult to be a monitor because ther would they be included in the proposed Swiss there is asymmetric information and because it takes regulation. 53 For the shareholders, it is vital to include effort to process complex information. However, de- people who make important decisions like "do we go spite the potential problems identified earlier, if costs into this business?" in the compensation table. Every- are low and benefits are high, institutional investors one on a level below this, in contrast, is under the re- . can be good monitors as long as the subject is about sponsibility of the CEOs 51 Gordon s ideas have been general rules or matters. However, they may still be included into today's Item 402 of Regulation S-K. 55 ineffective when it comes to deciding very specific Another issue is the question of how much stock- matters which are highly complex. A possible solu- ownership can set incentives for the managers, i.e., tion might be to require the companies to have clear, how much will a manager gain or lose if the stock simple and better disclosure of compensation agree- price fluctuates by one dollar 956 The investor should ments. Gordon has proposed guidelines for a Com- be able to see the marginal exposure of an executive's pensation Discussion and Analysis (CD&A): 4 First, compensation to the stock of the firm. There is no way the CD&A should provide standardized information. to see that in today's compensation disclosure. It is With standardization, the costs for preparing the in- very time-consuming to analyze the executive's mar- formation (from the perspective of the company) and ginal exposure to stock with a case-by-case analysis. to understand the information (from the perspective Shareholders do not excel in this type of analysis. In- of a shareholder) will be lower. Second, one should stitutional investors in the U.S. ship this kind of anal- provide justification of compensation practices to the ysis out to proxy advisor companies. In most cases, shareholders. The company should be required to ad- proxy advisors like RiskMetrics Group and its com- dress any questions which the shareholders may have. petitors make voting decisions for the institutional With this degree of standardization, the summary shareholders. 57 The existence of proxy advisor firms compensation table in the CD&A provides informa- is troubling because important shareholders delegate tion about five people: the CEO, CFO, vice chair- their decisions to them; this creates the impression that shareholders are bad monitors. Given that insti-
48 See Co/lee (Fn. 23), 1323. Take Opel in Germany as an example where the layoff of employees was used as a ne- See Code of Federal Regulations respectively 17 CFR gotiation tactic; Peter Rdsonvi, Firmenrettung nach Re- § 229.402 (Item 402). zept, Flohe Hurden f r deutsche Staatsburgschaften und ' See the instructions to 402(a)(3) oft? CFR § 229.402. Kredite, Neue ZOrcher Zeitung (March 3, 2009), 12. See 52 See David Segal, Trader's $100 Million Payday Poses also a Swiss example where it was difficult to conduct a Quandary for Regulator, The New York Times (August 2, necessary restructuring process: Sonia Wuhrn,ann, Der 2009), A2. Fall Swissair - ein Fall Schweiz, Psychodynamische See supra Fn. 15. Mechanismen eines Desasters, Neue ZOrcher Zeitung 54 However, the design of golden parachute agreements is (April 11, 2002), 15. one exception to this rule, see infra Part 111.2. See Jet/ivy N Gordon, Executive Compensation: If There's 15 See 17 CFR § 229.402. a Problem, What's the Remedy? The Case for "Compensa- See Robert I Jackson, Ii:, Private Equity and Executive tion Discussion and Analysis", The Journal of Corporation Compensation (f orthcoming), 24--26. 57 1358; Lipton/Say/ft Law 30 (2005) 675-702. See Coffee (Fn. 23), (Fn. 8), 753 SZW/RSDA 2/2011 Muller: Regulation of Say on Pay: Engineering Incentives for Executives and Directors 175
tutional shareholders are currently reliant upon proxy apathetic shareholder who does not vote in an in- advisors to make voting decisions, we should create formed manner as long as everything appears right or rules that take into consideration this factor and allow at least acceptable to the shareholder. 6 ' consequently, shareholders to exercise voting power themselves. non-apathetic shareholders would perhaps need to The rules should reduce the amount of work which rely on a consultant or proxy advisor firm who ex- shareholders must do to successfully exercise their plains complex compensation matters. Pursuant to voting power, because shareholders are not good at such delegation, yet another principal agent problem, processing information and do not have the ability to this time between the advisor and the shareholder, do a lot of research. Despite their superior resources could be created because the advisor's primary in- as compared to individual shareholders, institutional terest is to generate advisory fees; reducing agency investors do not have unlimited manpower to do re- costs in the corporation is only a secondary interest. search for the thousands of firms in which they in- On the other hand, the shareholders primarily want to vest. For instance, TIAA-CREF, one of the biggest make an informed decision that reduces agency costs investment funds in the U.S., has to make ca. 13,000 in the corporation. 62 Because of the principal-agent say on pay decisions in the year 2011 with a staff of problem just identified, the costs which are suppos- only seven people. 58 For some Swiss pension funds edly reduced via monitoring (in the form of agency it might be easier since they usually just invest in a costs in the context) will reappear in another form. In few stocks (e.g., the SMI listed shares). However, the other words, if a shareholder must pay high fees to an problem of processing complex disclosure remains. advisory firm in order to obtain effective monitoring The disclosure regime in Switzerland regarding (with the aim of lowering agency costs), the money executive compensation is codified in Art. 663b he saves on the latter is simply spent on the former. and Art. 663c Sec. 3 CO." Additionally, cross-border The shareholders have to find out which alternative listed Swiss public corporations also comply with results in the best benefit for them; if the advisory foreign CD&A rules (e.g., U.S. rules) within the fees (i.e., monitoring costs) exceed the agency costs same disclosure report, whenever possible. 6 How- caused by the executives, the shareholders are better ever, in Switzerland, similar problems occur as in the off not hiring the advisors (et vice versa). U.S. Disclosure is in some respects too complex; yet in other respects it is not comprehensive enough. As (iv) Moral Hazard Problem mentioned, individuals such as Andrew Hall would Another reason why shareholders might not be the not appear in the disclosure under current rules be- best monitors is the fact that the bigger the company cause he probably would not be an executive. There- gets, the more serious the moral hazard problem. This fore, it is difficult to find the right balance between is especially true for systemically-important financial simple disclosure which is understandable by the intermediaries or companies. 63 If a risky business average investor, and a report that discloses informa- strategy adopted by such a company turns out well, tion that is actually material and relevant. For an out- the shareholders can only benefit. In the worst case sider (and even for insiders of the corporation) it is scenario, the government has to bailout the financial extremely difficult to understand how to set executive intermediary in order to prevent collateral damage. compensation incentives for the company in ques- Thus, a shareholder may in fact prefer the more risky tion. Hence, it is hard to tell what kind of actions and approach. This moral hazard problem creates possible performance the executives and directors should get rewarded for. Furthermore, how should shareholders monitor executives and directors based on a report which, as discussed, is very complex to understand? This argument links to the concept of the rationally ' See supra 11.2 at the end. 62 See supra 11.3.1.1. 63 See R diger Fahlenhrach/RenØ M. Stultz, Bank CEO in- See supra Fn. 37. centives and the Credit Crisis, NBER Working Paper See generally Watter/Maizar (Fn. 15), 352-359; see also 15212,
176 Miller: Regulation of Say on Pay: Engineering Incentives for Executives and Directors SZW/RSDA 2/2011
social costs for taxpayers. 64 This problem is imminent executives, in Switzerland the advisory vote is also if the shareholders of a corporation and the taxpay- being discussed in relation to director compensa- ers are not the same people; e.g., when the taxpayers tion. 65 In terms of form, using an advisory vote as are Swiss, yet it is foreigners who own the majority opposed to a binding vote is probably the better ap- of a systemically relevant corporation (and who must proach. Instead of demanding a mandatory regime therefore foot the bill for any bailout). where the shareholder vote is conclusive, an advisory vote would give the advantage of combining the feed- 3.1.3 Shareholder Approval of Say on Pay back value of the shareholder vote and simultane- § 951 of the Dodd-Frank Act gives shareholders ously preserving the flexibility of the board of direc- of public companies in the U.S. a non-binding advi- tors in matters regarding compensation. According to sory vote, i.e., "say-on-pay", and for the approval of the latest versions of the draft, there is the following golden parachute agreements. Giving an advisory default rule: shareholders at the annual shareholder vote may provide institutional investors with more meeting must approve the compensation of the board power over how board members behave. Without of directors (although they can choose to opt out of the advisory vote, it would be difficult to express this regime). 66 The rationale for the mandatory ap- dissent. Furthermore, it is generally difficult for in- proval vote is the idea that setting their own compen- dependent directors to confront directors about sala- sation is always a matter in which the directors are ries. Therefore, one advantage of feedback from the conflicted; shareholder approval eliminates this con- shareholders is that the independent directors may flict." Even though this is a valid point, shareholders have better leverage resulting from such feedback. may not be the most effective or rational actors when They can rely on the votes of the shareholders to sup- it comes to executive compensation because they port their proposed changes to executive compensa- may not understand the underlying incentive struc- tion. An advisory vote therefore helps to set executive ture or need to sufficiently compensate executives for compensation inside the boardroom and provides a correct business strategies. Therefore, it is better to form of feedback at the annual shareholder meeting. preserve the flexibility of the directors and combine If shareholders do not have this advisory vote, the it with the feedback of an advisory vote about com- next best alternative for them would be to vote the pensation. directors off. But this would generally not be in the interest of the firm. Why should shareholders refuse 3.2 Incentive Alignment to re-elect a director just because his compensation Incentive alignment of principals (i.e., sharehold- is viewed as too high, if he is in fact doing a good ers) and agents (i.e., directors and executives) can job running the company? This would seem unrea- partially be achieved by requiring the latter to hold sonable and would possibly result in a loss of share- equity in the company. In order to achieve the incen- holder value. If the directors disregard the advisory tive alignment, the stock ownership of each director vote of the shareholders and do not deliver justifica- has to be significant. Hence, this would make it ex- tion for high compensation, it will force shareholders pensive to hire directors and executives in the sense to vote directors off. For this reason, the advisory say that one must allocate substantial ownership of the on pay vote creates an escalation process of conflict company to them. A company can address this prob- resolution. The implementation of the advisory vote lem by spreading equity compensation over a number can create an ongoing dialogue between important of years (e.g., five years), and limiting equity compen- investors and the corporation. Providing shareholders of Swiss companies lim- ited by shares with a say on pay vote would create similar possibilities for ongoing dialogue. Moreover, See supra Fn. 4. in contrast to the U.S., where shareholders only have See B//h/er (Fn. 4), 257-267. a non-binding advisory vote for the compensation of See Art. 73 1 k and Art. 7311 CO in the 2010 draft by the Council of States, 13131 2010, 8317-8318; see also von der Crone/Burg (Fn. 27), I1.1.c); see also BGE 127 III 333 336; BGE 126 III 363-364; Meier-Hayoz/Forstrno6e,. See Lucian Bebchu/c/Holger Sparnann, Regulating Bank- (Fn. 25), § 16 N 244-248; von der Crone/Curhonara/M0r ers' Pay, The Georgetown Law Journal 98 (2010) 255-257. c/nez (Fn. 28), 410-412. SZW/RSDA 2/2011 Muller: Regulation of Say on Pay: Engineering Incentives for Executives and Directors 177
sation to the form of restricted stock. 69 But if manag- high, this may cause a loss of firm-specific expertise ers can only receive restricted stock, how should they if knowledgeable directors are not re-elected. In this pay their mortgages and taxes? Thus, although it may case, freshly appointed directors could have the view be better purely for incentive alignment purposes to that executives know how to run the firm better - this award only equity compensation, it is necessary to is clearly not ideal in a scenario where directors are provide part of the compensation in the form of cash supposed to be a restraint on management. 7 ' Another in order to avoid managers from having liquidity issue is that if the director has only a very short term, problems. Giving directors five-year terms can create he may be motivated to have a very short-term out- the incentive that the directors hope to get re-elected look. He will only be interested in driving the stock for another five years even though this is a very long price up (even if at the long-term expense of the com-
. tenure. Requiring the directors and executives to hold pany) until the point when he is able to sell 71 How- a substantial share of their firm, may make them ever, one possible solution to this problem is to have more motivated. But this measure, combining long the restriction on the stock extend to beyond their tenures in the board and forcing managers to hold a departure (i.e., even if they leave the company) this lot of stock, will also make it harder for sharehold- would reduce the incentives to maximize only short- ers to get rid of bad directors and executives. Thus, it term profits. reduces the incentives for them to do what the share- holders want, particularly since a longer term poten- 3.3 Compensation as a Screening Device tially leads to more enjoyment of "perks" stemming vs. Anchoring from their high position in the company. 69 Another Posner suggests that compensation could work as issue is that when the end of each term approaches, a kind of screening device to evaluate whether poten- directors become more sensitive to managerial influ- tial executives are committed to a firm. A company ence (since the management will choose whether or should basically offer an interesting job which the ex- not to re-elect them), thus, it weakens the bargain- ecutive wants to do simply because he wants to work; ing power of the directors in the pay-setting process i.e., the executive would work because of his intrinsic with the executives. There are possible solutions to motivation. Extrinsic incentives (especially perform- this; for example, the directors could set the date of ance-based compensation elements) could suppress compensation-related negotiations in the middle of a or even replace the intrinsic motivation because the term rather than at the end. Another solution would manager will focus his performance to reach the ex- be to provide for staggered boards so that the pres- trinsically influenced performance goals (even if the sure mentioned above influences only a portion of the goals are not optimal)." The organization can there- directors at any given time. fore use the compensation as a screening element to The Federal Council proposes in his versions of attract people who want to work because of intrinsic the new CO annual elections of the board of direc- rather than extrinsic motivations. The compensation tors as a better solution. 7 However, this is not nec- is therefore a tool to screen all candidates for their essarily the right approach. If directors only have a one year term and the turnover of directors is very Regarding the bias of independent directors, see generally Markus Schimmer/Lisa Hoptinliller/Lukas MfihIc, Mit See Elson (Fn. 26), 983-995. This would also be consis- dern Strategieausschuss zur effektiven LJnternehmensauf- tent with the findings in personal and organizational eco- sicht - Zum Prinzip der symmetrischen Uberwachung, in: nomics which suggest that one should delay payment in Sven Kunisch/Christian Welling/Ramona Schmitt (Eds.), contracts where the performance of an employee is hard to Strategische Fuhrung auf dciii PrOfstand, Berlin 2010, monitor: Robert M. Hutchens, A Test of Lazears Theory 35-64. of Delayed Payment Contracts, The New Economics of He could possibly abstain from investing in research and Personnel 5 (1987) S167. development projects to cut costs. This strategy will boost 61 See Jensen/Meckling (Fn. 7), 318-319 for the formal short-term profits at the cost of long-term shareholder proof of this proposition. value. 70 See Art. 700 Sec. I CO as found in BBI 2010, 8311; see Bruno S. Frey/Margit Osterloh/Matthias Benz, Grenzen also the earlier draft of the revision of company and ac- variabler Leistungsl hne, Die Rolle intrins,scher Motiva- counting legislation in Art. 710 Sec. 1 CO as proposed in tion, in: Peter J. Jost (Ed.), Die Prinzipal-Agenten-Theorie BB1 2008, 1781; see generally BOhler (Fn. 4), 263-264 in der Betriebswirtschaftslehre, Stuttgart 2001, 563-73; and 267-268. see Posner (Fn. 12), 1018-1019.
8 Willer: Regulation of Say on Pay: Engineering Incentives for Executives and Directors SZW/RSDA 2/2011
level of intrinsic motivation. If a candidate passes this is the biased decision-making process where a per-
screen, theoretically it would make them more com- son makes a decision based on initial parameters. 8 mitted to the objectives of the organization. Moreover, people who read about the distribution of Posner assumes that the CEO is, of all the em- compensation tend to selectively hear more about ployees of a company, the most difficult one to con- the peers who earn more money than themselves and
trol because his performance is difficult to monitor. 74 more or less ignore those who earn less. In the end, He contrasts companies with government agencies this could lead to higher negotiated compensations and other non-business organizations. Because of and salaries if the bargaining partners are also an- the lack of competition which product and capital chored by the same cognitive bias because executives markets would provide in the private sector, agency will drive each others' compensation up by their as- costs are usually presumed to be higher in the govern- pirations to outdo their peers. 2 Due to this anchoring mental setting than in private and profit-maximizing mentality, it is even more apparent that a competent organizations. 75 This view is misguided because it director would avoid the company offering a lower ignores the fact that noncommercial organizations salary than its peers. In sum, one can see that offering have tools for limiting agency costs which business low compensation as a "screening device" to try and firms do not have (e.g., there are several supervisors attracted committed individuals is likely to backfire in the government such as commissions, supervisory because in many cases, taking this approach would boards, judges, and the sovereign) .16 Moreover, peo- discourage competent applicants who believe that ple attracted to the highest level of governmental jobs their work would be better rewarded elsewhere. are normally committed individuals. Posner claims that such organizations are able to create a high-com- mitment culture in which the executives work hard, III. Compensation Arrangements in Mergers often "for rather meager pay, because they internal- and Acquisitions ize the goals of their principal."" Offering low pay, as discussed above, could accordingly "be a screen- 1. Say on Golden Parachute Agreements ing device for quality, by eliminating from the ap- There exist a variety of possible compensation plicant pool persons who are not highly committed to agreements in merger and acquisition situations. the employer's mission."" However, it may be risky Golden parachutes are the kind of compensation to adopt the approach of offering low compensation agreements which have caused the most public con- to attract committed people; one could imagine that sternation. 83 A golden parachute is the compensa- many excellent candidates will choose instead to go to the firms offering a higher salary. 79 Moreover, the effect of offering low executive compensation might also be to deter potential investors. They could have See Amos Tversky/Daniel Kahnensan, Judgment under Uncertainty: Heuristics and Biases, Science, New Series the impression that only average directors or execu- 185 (1974) 1128; Richard E. Kopelman/Anne L. Davis, A tives could possibly get attracted by such a low level Demonstration of the Anchoring Effect, Decision Sci- of payment. Finally, the knowledge about how much a ences Journal of Innovative Education 2 (2004) 203; simi- lar Beat M. Barthold/Marc Widnei; Regulierung der van- director or an executive in a similar position earns an- ablen VergOtung?, Eine Bewertung der schweizerischen chors the aspiration level of a candidate. 80 Anchoring Regulierungsbestrebungen im Vergutungsbereich, AJP 18 (2009) 1394-1395. See Scott Highhouse et at., What Makes a Salary Seem Reasonable? Frequency Context Effects on Starting-salary See Posner (Fn. 12), 1018-1019, Expectations, Journal of Occupational and Organizational SeePosner(Fn. 12), 1018-1019. Psychology 76 (2003), 79. 76 See Posner (Fn. 12), 1018-1019. Sec also Rost etal., The See generally Daniel Daeniker/AleXander N/k/tine, Corporate Governance of Benedictine Abbeys, What Can Golden Handshakes, Golden Parachutes und hnliche Ver- Stock Corporations Learn From Monasteries?, Journal of einbarungen bei M&A-Transaktionefl, in: Rudolf Tschbnj Management History 16 (2010) 97-103. (Ed.), "Mergers & Acquisitions IX", Vol. 79, Zurich 2007; Posner (Fn. 12), 1019. see also Bbckli (Fn. 6), § 13 N 246a-259; Peter Nobel, Posner (Fn. 12), 1019; see also Inauen etal., management Finanzmarktrecht und internationale Standards, 31d Ed., revue 21 (2010) 48-54. Bern 2010, § 10 N 112 and N 534; BSK BEFIG-Tschjjni/ See supra Part 11.3.1.2 at the end. Iffland/Diem, Art. 29 N 24-29; Arie Gerszt, "Goldene But cf. von der crone/Burg (Fn. 27), III. Fallschirme" im Schweizerischefl Recht, Nr. 103 der SZW/RSDA 2/2011 Muller: Regulation of Say on Pay: Engineering Incentives for Executives and Directors 179
tion for an employee in the case that his employment transactions will occur more easily without the resist- contract is terminated after the change of control in ance of executives who may lose their jobs. 89 If this a company. 14 According to § 951 of the Dodd-Frank assumption is true, shareholders should have an advi- Act and the rules promulgated by the Securities and sory approval vote about golden parachutes since it is Exchange Commission ("SEC"), companies will essential for the future development of the shareholder now be required to provide the shareholders with ad- value of the company. However, gratuity payment par- ditional information about golden parachute agree- achutes do not facilitate possible takeover transaction ments.85 The Dodd-Frank Act distinguishes mainly because the managers do not know whether they will two types of golden parachutes: (1) payments which receive the golden parachute ex ante. 90 are negotiated ex ante in an employment contract of The shareholders of the target company want an executive and (2) payments which are not negoti- the executives to agree and to work towards a share- ated in advance to a takeover ("gratuity payments"). holder-value-increasing takeover despite the fact that In the latter case, the managers do not know if they they are risking the loss of their jobs. Golden para- will receive a payment in the takeover until after the chute compensations which facilitate a takeover and takeover has occurred. 86 provide the CEO of the target a severance payment The justification for granting golden parachutes are made because they are part of the executive's com- originates from the idea of encouraging the manager pensation contract. 9 ' This type of payment is the sub- to engage in an efficient corporate transaction. 87 Man- ject of the shareholder advisory approval in § 14A(a) agers of the target would, according to this theory, re- (1) of the amended SEA. The new Rule 14a-21(c) sist selling the firm or the division which they control does not require a separate shareholder advisory vote if they do not get compensated for the possible loss of to approve golden parachute compensations if each their job. For instance, the executives have to give up golden parachute compensation arrangement has al- the private benefits of control and the future income ready been disclosed in a manner complying with the they would have if they must leave the company. Giv- new Item 402(t) 92 and if the disclosures of the golden ing up these benefits can be an obstacle to the takeover parachutes were subject to a previous say on pay ad- because the executives might use their discretionary visory vote. 93 power to work against a merger. The CEO and the The second type of golden parachute - the gratu- board could adopt a variety of takeover defense meas- ity payment - shall be approved by the shareholders ures which would raise the costs for a possible takeo- according to § 14A(b)(2) of the amended SEA and ver. Thus, golden parachutes do probably facilitate the SEC-promulgated regulation .’ In a takeover situa- market for corporate control 88 because useful takeover
26. This idea does not seem to be a part of the Swiss doc- Reihe Diskussionspapiere des Forschungsinstituts f r Ar - trine. The Swiss doctrine views golden parachutes as an beit und Arbeitsrecht an der Umversit t St. Gallen (June illegal defensive measures in takeover battles; see e.g. 2004), available at:
Iliiller: Regulation of Say on Pay: Engineering Incentives for Executives and Directors SZW/RSDA 2/2011 180
tion, the shareholders must be provided with all dis- The Swiss doctrine sees golden parachutes as dia- closure of the proposed gratuity payments according metrically opposed to the interests of the corporation to the new Item 402(t) of Regulation S-K. 15 If a pre- and thereby violating the duty of care and loyalty. 102 viously approved golden parachute agreement will However, this interpretation could create a conflict be amended or changed, a separate vote will also be with the maximization of shareholder value if one is necessary." The separation between the approval of more persuaded by the American view on the util- the severance payment disclosure and the vote about ity of golden parachutes discussed in the paragraph the "acquisition, merger, consolidation, or proposed above. Based on the conclusions drawn from the U.S. sale or other disposition of all or substantially all the doctrine, the prohibition of golden parachute com- assets of an issuer" 7 makes sense. Indeed, there could pensation agreements as proposed in the draft of the be situations where the shareholders want to approve Committee for Legal Affairs of the Council of States the takeover transaction but they also want to have the in Switzerland is troublesome as long as there is no possibility to say no to the disclosed golden parachute. research about the actual impact of golden parachute The doctrine in Switzerland takes an entirely dif- agreements on the shareholder value of Swiss pub- ferent view of golden parachutes. The Swiss doctrine lic corporations." 3 Although broadly a prohibition, sees severance payments as a defensive measure." this draft makes it possible to grant golden para- The incentive to facilitate takeovers is neither the chutes in exceptional circumstances as long as these subject of any discussion nor mention in the doctrine. agreements are in the company's interest."" A man- Golden parachutes are taboo, especially in takeover ager who wants such an agreement may use argu- situations.9' Such golden parachutes will be disal- ments stemming from the U.S. doctrine, although it lowed if they have the effect of making a takeover is unclear whether golden parachutes in Switzerland more difficult, even if the subjective intention behind would work in the same way as in the U.S., consid- the golden parachute is not to impede the takeover." ering the institutional setting unique to each coun- A golden parachute agreement is also impermissible try. It could be that there is similar beneficial effect if it is an obvious violation of the corporation law." in Switzerland when taking into consideration the Swiss market for corporate control and the circum- stances of the Swiss corporation law and economy. amended SEA was ambiguous because of the "unless" If one takes the position that Switzerland is relatively clause at the end of § 14A(b)(2). similar to the U.S., then the prohibition of golden These elements are: the names of the CEO, the CFO and the three highest paid executives and all the cash, equity, parachutes could reduce the force of the market for pension, perquisites, tax reimbursements, other sources of corporate control and therefore decrease incentives income and the total amount in US$; see the new Item for the management to engage in efficient takeovers. 402(t)(2) of the Regulation S-K; see also § 14A(b)(1) of Thus, this reduces the alignment of the incentives the amended SEA. 96 See § 14A(b)(2) of the amended SEA and the SEC Rule of the shareholders and the management according l4a-21(c). to the principal agent theory. 105 The better solution § 14A(b)(1) of the amended SEA. would be to require a say on all gratuity golden par- ' See Bock/i (Fn. 6), § 8 N 546-550. achute agreements in the takeover situation, and to See e.g. BOck/i (Fn. 6), § 7 N 205-205a; BSK BEHG- TschOni/IfJland/Diem, Art. 29 N 19-24; Ruc/o!fTschOni, require disclosure of material elements of the golden M&A-Transaktionen nach Schweizer Recht, Zurich/Ba- parachute agreements if they have not been subject se/Geneva 2003, 9. Kapitel N 26; Dieter Zobl/Stefiin Kra- to a prior vote during an annual shareholder meeting. met; Schweizerisches Kapitalmarktrecht, Zurich/Basel/ Geneva 2004, N 322-324. 99 According to the Swiss Takeover Board, it is not necessary that a board makes a defensive measure for the subjective cember 20, 2007, E. 2.1.1.4; BSK BEHG-TSChdfli/tflunc// purpose of defense; it is sufficient that a measure is from Diem, Art. 29 N 27. an objective point of view adequate to aggravate or prevent See Art. 717 CO; BSK BEHG-TS'ChiiOI/Itt/dlfld/Diem, a defensive measure; see Swiss Takeover Board, Empfeh- Art. 29 N 27. lung 0249/05, Saia-Burgess Electronics AG, August 23, See e.g. Art. 731m Sec. I CO as published in BBI 2010, 2005, E. 1.1.2; BSK BEHG-Tschc'ini/Jf//and/Diem, Art. 29 8318-8319; see generally BOO/er (Fn. 4), 274. N 19. 114 See Art. 731m Sec. 2 CO as cited in BBI 2010, 8139. See Swiss Takeover Board, Empfehlung 0249/05, Saia- 105 But cf. Lipton/Savitt ( Fn. 8), 749-752; see Swiss Takeover Burgess Electronics AG, August 23, 2005, E. 1.3.2; Swiss Board, Empfehlung 0249/05, Sam-Burgess Electronics Takeover Board, Empfehlung 0343/03, Implenia AG, Dc- AG, August 23, 2005, E. 13. SZW/RSDA 2/2011 Muller: Regulation of Say on Pay: Engineering Incentives for Executives and Directors 181
This additional transparency would improve monitor- of this, disclosure of golden parachutes should be ex- ing and it would lessen the undesirable incentives for tended in a reasonable and understandable manner to the directors and executives to obstruct takeovers. a broader group of individuals. However, there are many mid-level executives, and extensive disclosure 2. Extend Disclosure to a Broader Group could create information overkill. Disclosure regard- of Individuals ing this issue should therefore be succinct, in a way that explains whether the mid-level employees have Even if one agrees that golden parachutes should the same model of contracts as the top-level execu- not be banned entirely, it is not clear which golden tives, or in which ways the contracts are materially parachute agreements should be disclosed. Should different from those of the top-level executives. it only be those of the top five executives or should there also be a broader group of individuals whose parachute agreements ought to be disclosed? Pro- IV. Unintended Incentive Effects of Taxation viding the figures for more people would certainly and Salary Caps help shareholders obtain the information necessary to exercise their voting rights if the company had The Congress of the U.S. intended to reduce ex- to disclose whether mid-level executives have the cessive compensation in public companies with high same type of golden parachutes, or how they are taxation of such payments. Congress therefore en- designed and triggered. There are at least two types acted § 162(m) of the Internal Revenue Code (I.R.C.). of clauses in an employee's contract which make an According to its title, "Certain Excessive Employee agreement a golden parachute; (I) change of control Remuneration", it imposes taxes on executive pay. clause, (2) the condition that the manager loses his Congress introduced a cap of one million dollars; if job (or that he gets demoted). If only one of these executive pay exceeds one million dollars, the com- two types of clauses is included, the golden parachute pany will lose the tax deduction on the amount exceed- is single triggered; the contract is double triggered ing one million U.S. dollars."' But there is an excep- if both conditions have to be met in order to receive tion: performance-based compensation, as opposed to the compensation."" The idea behind including the cash compensation, can always be tax deductable. 119 mid-level executives in the disclosure is that the In order to make such expenses deductible, compa- shareholders know whether the upper-level execu- nies must satisfy the conditions in § 162(m) I.R.C. tives have the same incentives as the mid-level ex- This Section requires that there be objective perform- ecutives have. According to Segal, employees often ance goals, that the compensation committee consist have double triggered parachutes and top executives entirely of outside directors, that there be shareholder have single triggered parachutes. Such a situation is approval of the compensation, and that the outside not always ideal because it creates conflicting incen- directors certify that the executives have reached the tives which will result in inefficiencies within these performance goals."' However, I.R.C. § 162(m) has hierarchies; 107 shareholders should have the opportu- led to unintended consequences despite the aspira- nity to verify whether such a situation exists. Because tions of the drafters; executive pay has actually risen since the enactment of the aforementioned Section." One explanation for this result is the possible psycho- 00 See Shareholder Approval of Executive Compensation and Golden Parachute Compensation, Proposed Rule, Release Nos. 33-9153, 34-63124, 75 Federal Register 66590, See I.R.C. § 162(m)(1); William A. Klein et al., Federal 66600 (October 28, 2010); see also the final rule with Income Taxation, 150 Ed., New York, 2009, 558. modifications and amendments of the proposed rules: 109 See I.R.C. § 162(m)(4)(C). Shareholder Approval of Executive Compensation and 110 See I.R.C. § 162(m) and Staff of the Joint Committee on Golden Parachute, Compensation; Final Rule, Release Taxation, 109th Congress, Present Law and Background Nos. 33-9178; 34-63768, 76 Federal Register, 6010-6047 Relating to Executive Compensation (2006), 4. (February 2,2011); Daeniker/Nikiiine (Fn. 83); 122-123. ' See Staff of the Joint Committee on Taxation (Fn. 110), 107 See Michael Segal, Partner in the Executive Compensa- 1-9; Todd Perry/Marc Zenne,; Pay for Performance? Gov- tion and Benefits practice, Wachtell, Lipton, Rosen & ernment Regulation and the Structure of Compensation Katz, Address at the Evening Speaker Series of the Co- Contracts, Journal of Financial Economics 62(2001)457- lumbia Business and Law Association, Columbia Law 459 and 486-487; James S. Wallace/Kenneth R. Ferris, School, New York, N.Y. (October 26, 2010). IRC Section 162(m) and the Law of Unintended Conse- or
182 M Iler Regulation of Say on Pay: Engineering Incentives for Executives and Directors SZW/RSDA 2/2011
logical effect of the fact that Congress chose to set amount of dividends and therefore cannot pay com- the compensation cap at one million US$. Because pensations exceeding three million Swiss Francs.' 16 Congress chose this figure, executives who received Tax deduction is less relevant if the company does lower sums may have felt empowered to demand not generate any profits. Quite apart from the issue 2 compensation of one million dollars.' Furthermore, that the CO should not contain taxation provisions,' 7 although I.R.C. § 162(m) did lead to increased stock it is questionable whether this is the right initiative. It option based compensation, for a long time, it was is conceivable that this initiative may lead to a analo- permissible to issue stock options without recogniz- gous outcome to that resulting from the similarly ing any expense in the income statement if the issuer constructed I.R.C. § 162(m); namely, incentivizing followed the Generally Accepted Accounting Princi- managers to ask for a higher fixed payment because ples. ' 'I This led to investors not having a clear picture the parliament has "legitimated" payments of up to of the actual value of the compensation being paid three million Swiss Francs. (one can obviously see that it is harder to value stock Another possible solution has been proposed by compensation than cash compensation). In addition, Schweiger et al. in the form of Art. 731In CO ("Alter- if a risk-averse executive is to be paid in stock rather nativrnodell").''8 This provision only applies to public than cash compensation, he would tend to demand corporations and it allows slightly more flexibility in higher compensation since stock payments are riskier granting "very high compensations."' 9 There is a three to him than cash payments. 114 million Swiss Francs salary cap if the company either The Swiss parliamentary initiative No. 10.460, cannot generate profits or if the share capital and the which was submitted to the Swiss Council of States general reserves are impaired.' 2 Though, under this on June 22, 2010, seeks to disallow tax deduction of model, it would be possible to make an exception of compensation exceeding three million Swiss Francs this salary cap if the shareholders approve such "very in a newly proposed Art. 677 CO ("Tantiemen- high compensation.""' However, it is doubtful whether Modell")." 5 This provision also seeks to ban pay- rationally apathetic voting shareholders are competent ments exceeding three million Swiss Francs if the enough to assess whether the salary cap in the pro- company will not pay certain dividends as specified posed Art. 731n CO according to the "Alternativmod- in the law. Note that the part of the proposed rule dis- ell" is an efficient solution.'22 The directors could eas- cussed here only comes into play if a company gen- ily represent to a rationally apathetic shareholder that erates profits. If the company suffers losses or does the corporation has to pay higher salaries in order to not have retained earnings, it cannot pay the specified prevent employees from leaving to work for the corpo- ration's competitors. It is hard for such shareholders to verify whether the directors are asking for a truly le- quences (November 2006), available at SSRN:
approve very high compensation in all cases (whether surprises one with the following statement: "In the or not the company generates profits). The third wake of the financial crisis there is almost certainly model should arguably not be implemented because going to be some regulation of executive compensa- the same deficiencies as in the "Alternativmodell" tion - it has begun in the form of conditions in the occur. In addition, this model also contains the ad- recent bailouts of insolvent financial firms. The ques- ditional problem that the shareholder vote would be tion is not whether, but how best, to limit executive conclusive rather than advisory. As discussed above, compensation.""' The regulation of executive com- this would be an undesirable result if one assumes that pensation should empower shareholders, according shareholders are rationally apathetic. to the legal and economic theories discussed in this In all of the three models above,' 24 there is a tax article. Providing shareholders with a say on pay does burden. Excessive compensation is not tax deductible not solve all problems regarding excessive executive because it is regarded as share of profits instead of compensation. However, this article provides a few regular salary payment.' 25 Therefore, the additional hints for future regulation. In essence, the following tax burden makes the incentive setting more costly reform measures ought to be considered: First, the for shareholders. It is illogical to impose a tax on pos- full disclosure of compensation for all upper-level sibly social welfare creating incentives. This leads to and mid-level executives and directors is required the question, implicitly discussed at various points in order to reduce agency costs. Second, public cor - in this Article, whether the government should pun- porations should be required to pay executives and ish executive compensation in this manner at all. As directors a substantial part of the compensation in a matter of fact, high levels of CEO compensation the form of long-term restricted stock in the corpora- could be socially valuable, to such an extent that it tion. The executives should also be prohibited from should not be "discouraged" by tax penalties. CEOs hedging or selling the stock for a specified number of can create jobs which will lead to further income tax years.'28 Third and on a related point, executives and revenues. Why should the government disallow the directors may receive a substantial payment of their deduction of socially valuable corporate expendi- compensation only at the end of their term, and such tures? In advance of enacting these kinds of capital compensation should be linked to the future perform- retention and taxation laws, it should be determined ance of the company.2S Reforms should aim to set if the demand for harsh taxation of excessive execu- incentives for long-term planning of company opera- tive payments is socially desirable. It will be hard to tions. Finally, shareholder monitoring should become undo changes in the Code as soon as they have been more affordable and should be made more effective enacted; thus, it will take time to correct provisions and easily exercisable. Proper regulation of executive that turn out to be socially costly.'26 compensation would balance two key aims: (1) al- lowing shareholders to delegate business decisions to management and to incentivize maximization of the V. Conclusion and Ideas for Future long-term value of the corporation, and (2) provid- Regulation ing shareholders with the tools to restrain managerial behavior when necessary. Posner, a staunch believer in the free market and a main protagonist of the law and economics discipline,
24 See