EFRAIM BENMELECH, Harvard University - Department of Economics, National Bureau of Economic Research (NBER)
Email: effi_benmelech@harvard.edu
EUGENE KANDEL, Hebrew University of Jerusalem - Department of Economics, Centre for Economic Policy Research (CEPR)
Email: mskandel@mscc.huji.ac.il
PIETRO VERONESI, University of Chicago - Graduate School of Business, Centre for Economic Policy Research (CEPR), National Bureau of Economic Research (NBER)
Email: pietro.veronesi@gsb.uchicago.edu

Stock-based compensation is the standard solution to agency problems between shareholders and managers. In a dynamic rational expectations equilibrium model with asymmetric information we show that although stock-based compensation causes managers to work harder, it also induces them to hide any worsening of the firm's investment opportunities by following largely sub-optimal investment policies. This problem is especially severe for growth firms, whose stock prices then become over-valued while managers hide the bad news to shareholders. We find that a firm-specific compensation package based on both stock and earnings performance instead induces a combination of high effort, truth revelation and optimal investments. The model produces numerous predictions that are consistent with the empirical evidence.

MEREDITH R. CONWAY, Texas Wesleyan School of Law, Suffolk University Law School
Email: mconway@law.txwes.edu

In the 1980s and 1990s, the public began to protest the large compensation packages executives were receiving. Average workers were struggling while executives got raises, even as the corporations they worked for failed. This disconnect between executive compensation and executive performance led Congress to attempt to curtail executive compensation. This article will examine the Congress's attempt to temper the amount of compensation through the tax code. These tax code provisions enacted by Congress to restrain excessive executive compensation in fact had the effect of increasing compensation for certain executives at a great cost to stockholders. In 1980, the average CEO made 42 times the average hourly worker's pay. By 1990, the average CEO made 107 times the average hourly worker's pay. In 1993, Congress enacted tax legislation intended to rein in excessive executive compensation. However, in 2000, the average CEO made 525 times the average hourly worker's pay. Compensation amounts that executives receive since the enactment of the tax provisions are increasing dramatically, not decreasing.

Congress believed that linking performance and compensation would adequately address excessive compensation amounts. Stock prices are one indicator of how a corporation is performing, and Congress accordingly enacted tax provisions linking executive compensation to stock performance. The tax provisions sought to convert cash compensation to stock options because it was thought that doing so would align the interests of executives and shareholders by making executives vested owners in the corporation as well. The unintended result was to encourage corporate fraud and accounting misrepresentations intended to inflate earnings and bring executives higher salaries.

The three most important tax code provisions designed to contain or regulate excessive compensation are §§ 162, 162(m) and 280G. None of these sections work as they were intended. On the contrary, they have resulted in even larger amounts of compensation amounts paid to executives at greater cost to shareholders. This article will seek to demonstrate that the effects of these tax code provisions are in direct conflict with their intended purpose.

This draft of the article does not contain completed law review edits.

ALBERT H. CHOI, University of Virginia Law School
Email: AHC4P@VIRGINIA.EDU

We demonstrate how a golden parachute can be used to improve the target shareholders' net return by partially shifting the managerial compensation burden to the buyer through a higher acquisition price. Consistent with the empirical observations, we show that (1) the golden parachute will be contingent on a change-of-control rather than solely on the manager's layoff, (2) the golden parachute will be promised early, for example, at the time of the manager's employment, not just in the face of a takeover or a merger, (3) the shareholders would want to extend its coverage to other employees, and (4) the size of the parachute can be much larger than the manager's annual compensation. We also examine the effect of a golden parachute on the managerial incentive scheme.

JAEYOUNG SUNG, University of Illinois at Chicago - Department of Finance
Email: U42022@UICVM.UIC.EDU
PETER L. SWAN, UNSW
Email: peter.swan@unsw.edu.au

In this paper we provide a simple agency model of executive pay as it relates to both firm size and executive career concerns as managers are recruited by firms of different sizes over their careers. Unlike the matching literature which assumes that the talent of managers is common knowledge, we explicitly model the hiring decisions of firms of different sizes when managerial talent is unknown but reflected imperfectly in firm performance in the manager's early career. We assume that agents' reservation utility levels are solely determined by their labor market opportunities, and we endogenously compute their reservation utility levels taking their labor market opportunities into account. Our empirical results show that a doubling of firm size raises CEO effort productivity by about 62% and ability productivity by over 80%, whereas pay increases by only 50%.

JIE CAI, Drexel University
Email: jc468@drexel.edu
RALPH A. WALKLING, Drexel University - Lebow College of Business
Email: rw@Drexel.edu

The post Sarbanes-Oxley Act period is associated with several initiatives designed to give shareholders a greater voice in the boardroom. The latest of these initiatives is the Say-on-Pay Bill (H.R. 1257) which passed the House of Representatives on April 20, 2007 by a 2 to 1 margin. This bill does not limit CEO pay but requires an advisory shareholder vote on executive compensation packages. Using the abnormal return of 1,245 firms surrounding the House passage of this bill, we examine whether the market interprets shareholders‘ say on executive pay as adding or subtracting firm value. Stocks of firms with positive abnormal CEO compensation react in a significant, positive manner to the Say-on-Pay Bill. The positive market reaction is stronger among the firms with weaker, but not the weakest governance. In addition, abnormal returns are higher in the subset of firms more likely to receive higher disapproval votes from shareholders and firms more likely to implement changes under the pressure of shareholder votes. Thus, the bill has the greatest impact among the subset of firms most likely to benefit and implement changes. Given the uncertainty surrounding passage, implementation and efficacy of this proposed advisory vote, the results are likely to understate the actual impact of Say on Pay legislation. Our findings suggest that the market views this legislation as value-creating for the companies where it is likely to have the most impact. These results provide important evidence for the current debate regarding the Say-on-Pay legislation in Congress and shareholder access to proxy. Our results also shed light on the role of activist investors.

JEAN CANIL, University of Adelaide - Business School
Email: jean.canil@adelaide.edu.au

Schaefer (1998) and Baker and Hall (2004) posit a firm size effect for regular executive compensation but not specifically for executive stock option grants. They propose an inverse relation between pay-performance sensitivity and firm size along with a positive relation between the marginal productivity of executive effort and firm size. The product of pay-performance sensitivity and executive productivity is 'incentive strength'. They find a weakly positive association between incentive strength and firm size. We substitute Hall and Murphy's (2002) pay-performance sensitivity metric to detect a firm size effect in CEO stock option grants. After adjusting for small-firm risk aversion and private diversification 'clienteles', we document evidence of a residual small-firm effect impacting on incentive strength principally through grant size. Given lower small-firm deltas, grant size appears to have been increased by compensation committees to ensure small-firm CEOs are not under-compensated relative to their large-firm counterparts. We also find that firm complexity influences pay-performance sensitivity as well, but not labor productivity (proxying for CEO productivity). No evidence is found that firm smallness and complexity impact on labor productivity. However, we empirically confirm a negative relation between pay-performance sensitivity and firm smallness and, by implication, firm complexity.

OLIVER G. SPALT, University of Mannheim - Department of Business Administration and Finance
Email: spalt@bwl.uni-mannheim.de

This paper documents that riskier firms grant more options to non-executive employees using a large panel of US firms from 1992 to 2005. These results are not explained by factors on the industry level. A simple model in which a risk-neutral firm and an employee with cumulative prospect theory preferences bargain over the employee's pay package can provide an explanation for this otherwise puzzling behavior. The key feature which can make options attractive is the well-established tendency of individuals to overweight small probabilities of large gains. I calibrate the model using standard parameters from the experimental literature and find that it fits the data remarkably well. The model provides a unified framework for thinking about both employee stock option grant and exercise decisions. This paper provides further evidence that individual biases in risky choice can be exploited by firms and that economic profits from exploiting such biases can be large.