The impact of performance-based compensation on misreporting

Abstract

This paper examines the effect of CEO compensation contracts on misreporting. We find that the sensitivity of the CEO's option portfolio to stock price is significantly positively related to the propensity to misreport. We do not find that the sensitivity of other components of CEO compensation, i.e., equity, restricted stock, long-term incentive payouts, and salary plus bonus have any significant impact on the propensity to misreport. Relative to other components of compensation, stock options are associated with stronger incentives to misreport because convexity in CEO wealth introduced by stock options limits the downside risk on detection of the misreporting.

Introduction

In September 1998, Arthur Levitt, the then chairman of the Securities and Exchange Commission, remarked that the desire of executives to increase the value of their stock options gave them an incentive to manipulate their accounting numbers (Levitt, 1998). Back in 1997, when large option grants were becoming more prevalent, L. Dennis Kozlowski, former CEO of Tyco, characterized options as a “free ride… a way to earn megabucks in a bull market with a hot company.”¹ Is the earnings management referred to by Levitt motivated by firms that are trying to look like a “hot company” to earn “megabucks” during a bull market? Does the recent increase in the number of accounting restatements coinciding with the increase in the use of stock options serve as evidence of the concerns put forward by these men?

This paper examines whether and how management's incentives, through their compensation contracts, affect the likelihood of engaging in unusual accounting practices that result in a restatement of financial statements. If aggressive accounting practices affect stock prices, then managers with equity-linked compensation might have an incentive to maximize their wealth through accounting choices. Given the increased linkage between a manager's compensation and stock price, it is important to examine the relationship between compensation contracts and accounting choices.

We compare S&P 1,500 firms that announce a restatement of their financial statements over the period 1995 to 2002 with those firms that do not restate. Our sample of restating firms includes firms that restated because their original financial statements were not in accordance with Generally Accepted Accounting Principles (GAAP). We measure the sensitivity of all components of CEO compensation to firm performance and examine the effect of this sensitivity on the incentives to adopt aggressive accounting practices that result in a restatement.

In particular, we are interested in the effect of stock options on the adoption of aggressive accounting practices. Option compensation makes CEO wealth a convex function of stock price. Consequently, the CEO benefits from an increase in the stock price associated with aggressive accounting. However, the loss to CEO wealth in the event of a decline in stock price is limited. Management is rewarded in good times, but not penalized as much in bad times. We measure option sensitivity as the change in the value of stock options held resulting from a 1% change in firm value. In our sample of 266 restated firm-years and approximately 8,000 nonrestated firm-years, we find strong evidence that option sensitivity is positively associated with misreporting. The greater is the sensitivity of CEO wealth to stock price arising from the CEO's option holdings, the greater is the propensity to misreport. Further, we find significant evidence that the greater the convexity of CEO wealth to stock price, the greater is the propensity to misreport. The evidence is consistent with the hypothesis that incentives from options encourage aggressive accounting practices that result in a restatement.

Like stock options, equity and restricted stock also tie CEO wealth to stock price. Equity and restricted stock might also generate incentives to misreport. However, in contrast to options, the payoff from equity and restricted stock has a symmetric relation to stock price, thereby exposing the CEO to price declines associated with the announcement of a restatement unless the CEO unwinds his equity and restricted stock holdings prior to the restatement. The vesting requirements associated with restricted stock and the possible loss of control benefits associated with sale of equity might limit the desire and ability of CEOs to unwind their holdings of equity and restricted stock. This suggests that equity and restricted stock expose the CEO to price declines and therefore may not be associated with a higher propensity to misreport. In line with this conjecture, we find no evidence that incentives from equity and restricted stock are associated with misreporting. Although equity and restricted stock holdings potentially bear the costs of misreporting, there is no evidence that the costs are large enough to mitigate the positive effect of stock options on misreporting.

Long-term incentive plans make CEO wealth a function of longer-term firm value. This reduces the incentives of CEOs to misreport to boost short-term stock prices. Consistent with this expectation, we find no evidence that long-term incentive payouts are associated with a propensity to misreport. The final component of CEO compensation that we examine is the cash component, i.e., salary plus bonus. Increased bonus payments associated with higher earnings are also likely to encourage CEOs to misreport. However, we find no evidence that an increase in salary plus bonus is a significant motivation for misreporting.

We collect data on the magnitude of the restatement, i.e., the effect of the restatement on net income. We find a positive significant association between option sensitivity and the magnitude of the restatement. Higher incentives from stock options are not only associated with a higher propensity to misreport but are also associated with higher magnitudes of misreporting. We continue to find no evidence that incentives from other components of CEO compensation are related to the magnitude of the restatement.

We also examine and control for other firm characteristics that might be associated with misreporting. We find that restating firms are somewhat larger than nonrestating firms. Further, restating firm-years are associated with higher leverage in comparison to nonrestating firm-years. There is no significant difference between restating firm-years and nonrestating firm-years in growth opportunities, research intensity, capital expenditures, and external finance in our multivariate analysis.

The firm's environment might affect the propensity to misreport. Bolton et al. (2003) argue that managers are more likely to use aggressive accounting in speculative periods associated with higher market valuations. We find weak evidence in support of this. Misreporting is more likely in periods in which the mean industry market-to-book ratio is high. However, we do not find that investor optimism, proxied by long-term analyst forecasts, is related to the incidence of misreporting.

Bebchuk et al. (2002) argue that options enable management to extract rents in the form of excessive compensation. This rent extraction is achieved when CEOs taking advantage of information asymmetry are able to pool with CEOs exercising to achieve liquidity or diversification objectives. A consequence of this camouflage is that the market underreacts to the negative information in CEO option exercises during periods of alleged misreporting. Consistent with this argument we find that large option exercises in periods of alleged misreporting are associated with a greater market reaction at the time of the announcement.

The positive effect of incentives from stock options on the propensity to misreport has important implications for the design of executive compensation plans. In the prior literature, Jensen and Meckling (1976) and Smith and Stulz (1985) show theoretically that a greater link between CEO compensation and firm performance is associated with better incentive alignment and higher firm values. This suggests that some option holdings may have a positive effect on firm value. However, the use of stock options beyond some “optimal” level might be associated with an increase in the incentives for misreporting. In other words, “excessive” option usage should be related to misreporting. We find some evidence that restating firm-years are significantly associated with “excessive” option sensitivity.

The rest of the paper is organized as follows: Section 2 reviews related literature, Section 3 develops our hypotheses, Section 4 discusses the data and its characteristics, Section 5 describes the measurement of key variables, Section 6 presents the empirical results, and Section 7 concludes.