The Timing of Option Repricing

Transcription

1 The Timing of Option Repricing By Sandra Renfro Callaghan Department of Accounting, M.J. Neeley School of Business, Texas Christian University, Fort Worth, Texas P. Jane Saly Department of Accounting, St. Thomas University, St. Paul, Minnesota Chandra Subramaniam* Department of Accounting, M.J. Neeley School of Business, Texas Christian University, Fort Worth, Texas Final copy sent to JOF First Draft: November 10, 2000 Current Draft: December 27, 2002 * Corresponding author: Tel: (817) ; c.subramaniam@tcu.edu The authors would like to thank an anonymous referee, Chris Barry, Bob Vigeland, Don Nichols, Mark Vargus, David Yermack, and participants at Texas Christian University, Université Laval, University of British Columbia, 2001 Annual Meeting of the Accounting Association of Australia and New Zealand, 2001 Annual Meeting of the American Accounting Association and the 2001 Annual Meeting of the Financial Management Association. We also wish to thank Cristian Danciu and Scott Richardson for excellent research assistance. Professors Callaghan and Subramaniam gratefully acknowledge financial support from the Charles Tandy American Enterprise Center. Professor Callaghan also acknowledges support from the Luther King Capital Management Center for Financial Studies at Texas Christian University.

2 The Timing of Option Repricing Abstract We investigate whether CEOs systematically manage the timing of the stock option repricing to coincide with favorable movements in the company s stock price. For a sample of 166 firms that repriced executive stock options in 236 separate events during the period 1992 through 1997, we document that stock price generally rises sharply following the repricing date and continues to increase for the next twenty days. In addition, we provide evidence that CEOs often appear to choose the date of repricing to precede the release of good news or to follow the release of bad news in the quarterly earnings announcements. Since no information about the stock option repricing is released to the public around the repricing date, our findings suggests that CEOs may opportunistically manage the timing of the option repricing date for personal benefit.

3 In this study, we examine the timing of executive stock option repricing. During the past decade, stock options have emerged as the single largest component of compensation for U.S. executives, accounting for over fifty-percent of total compensation (Murphy (1999)). Stock options are granted to employees to align their interests with those of the shareholders. This alignment occurs through a reduction in the employees risk averse behavior that would otherwise accompany their ownership of stock, thus reinforcing the executive s incentive to undertake riskier projects with higher payoff potential (Jensen and Murphy (1990a,b), Mehran (1995), Yermack (1995), Hall and Liebman (1998) among others). However, when the firm s stock price falls significantly below the exercise price of the option, the incentive effect of the option is diminished or lost. One mechanism used to reinstate this incentive is the repricing of executive stock options. Repricing options, however, is highly controversial. Managers maintain that stock option repricing is necessary to restore incentives lost when options are underwater and to retain valued employees. However, shareholders argue that management should not be selectively shielded from declines in stock price since the stock price decline may be a result of their own decisions. Furthermore, they claim repricing undermines the integrity of future stock option plans. A number of papers have studied the decision to reprice stock options. 1 We extend this literature by examining whether a systematic pattern can be documented with respect to the timing of stock option repricing. Using a sample of 236 repricing events occurring during the period 1992 through 1997, we investigate share price movement around the repricing date. Consistent with prior research, repricing firms exhibit negative monthly returns for several months prior to repricing. Using daily returns, we observe significant negative abnormal returns for each of the five days prior to the repricing. For the five days following the repricing, this pattern reverses 1

4 and we observe significant positive abnormal returns. Beyond this five-day period, stock price continues to increase for another twenty days before it stabilizes and approximates market performance. While a positive abnormal return following repricing is consistent with investors viewing the repricing as good news (since incentives are potentially realigned and employee retention issues addressed), it is unlikely to be the case since we find no public announcement of the repricing prior to, or immediately following the event. In fact, repricings appear to become public information only after the release of the subsequent proxy filing, oftentimes several months following the repricing. Therefore, it is difficult to attribute any price increase immediately following the repricing event to public disclosure of the event. Since we cannot attribute the observed abnormal returns to disclosure, we posit that repricing may be timed to occur in proximity to another predictable event. Thus, we focus on the quarterly earnings announcement since managers are likely to have greater private information about the timing and content of these events, affording greater opportunities for management to time the repricing such that it maximizes the value of the repriced options. Consistent with our prediction, we find that repricing event dates tend to either precede favorable earnings announcements, or to follow unfavorable earnings announcements. Therefore, managers holding repriced options realize wealth increases from both the act of repricing, and also from timing the repricing. The act of resetting the exercise price to a new, lower price (usually the market price on the repricing day) increases the Black-Scholes value of the repriced options on the repricing day by an average (median) of $478,720 ($241,586). The observed positive excess return immediately following repricing translates into an average (median) wealth increase, due to timing, of $259,320 ($32,917) when measured five days following repricing, and $558,428 2

5 ($94,063) when measured twenty days following repricing. Relative to annual cash or total compensation levels, this benefit is economically significant to the executives. We further investigate the role of corporate governance in both the decision to reprice, and the timing of the repricing event. We find that the decision to reprice is more likely for firms with weak corporate governance and for firms with greater proportions of equity compensation in their compensation package. While these results suggest that the quality of corporate governance may impact the decision to reprice, we find no evidence that the benefits of timing the repricing event is a function of corporate governance. Several studies have documented opportunistic behavior by management in a variety of settings. 2 We document another setting in which managers are able to extract wealth from shareholders. In particular, the ability to guide the process used to reset the exercise price of executive stock options provides a mechanism by which managers can exploit the existence of asymmetric information for personal benefit. Our study also illustrates a potential downside to increased proportions of equity compensation. Yermack (1995) and Mehran (1995) show that greater proportions of equity compensation results in improved firm performance. We provide evidence that when firms presumably underperform, resulting in underwater options, there is a greater likelihood that firms will reprice these options when managers hold a greater proportion of their total compensation in equity form. The remainder of the paper is organized as follows. Section I discusses institutional background, prior research and the motivation for this study. Section II presents sample selection criteria and provides descriptive data. Section III examines the relationship between repricing and stock price movement. Section IV investigates the relationship between the repricing date, 3

6 earnings announcement date, and content of the earnings announcement. Section V examines the role of corporate governance in repricing, followed by conclusions in Section VI. I. The Repricing of Executive Stock Options A. Institutional background Stock options represent the largest proportion of performance-based compensation received by top management in U.S firms. The basic premise is that options motivate managers to increase the market value of the firm through improved long-term performance rather than by fixating solely on short-term profits. Stock options also reduce risk-averse behavior that would otherwise accompany managers ownership of stock, and encourage managers to undertake riskier projects with higher payoffs. Managers highly value the unlimited upside potential stock options offer, with none of the downside risk since the manager does not own the stock. However, when the market price of the stock falls significantly below the exercise price of the option, the incentive benefit of the option is diminished or lost. Hall and Murphy (2000b) suggest that underwater options have a high probability of expiring out-of-the-money. Therefore, risk-averse executives place little value on these options, resulting in little or no incentive function remaining in the option. In addition, Lambert, Larcker and Verrechia (1991) and Gilson and Vetsuypens (1993) suggest that the greater the options are out-of-the-money, the greater the incentive for managers to engage in high-risk projects despite a low probability of success. Consequently, firms experience significant pressure to restore the incentive attributes of these stock-based programs and to reduce the risk of turnover among key executives. One alternative, is repricing of underwater options. Other alternatives include granting additional options, accelerating new option grants, issuing other forms of equity compensation, changing 4

7 the emphasis on the equity proportion of total compensation, or a combination of these. Balachandran, Carter and Lynch (2001) examines alternatives to repricing and finds evidence that some firms with underwater options choose to increase cash-based compensation in lieu of repricing. The decision to reprice is generally the responsibility of the compensation committee. Once the compensation committee has recommended repricing, the decision is reviewed by the full board of directors. Repricing can be accomplished either through an option exchange, canceling and issuing new options, or by an amendment to change the exercise price. Repricing may also include changes to the vesting period, changes to the expiration period, or replacement of the old options by a reduced number of new options. In most repricings, the exercise price of the option is reset to the market price of the underlying security on the repricing date. There are no rules governing the selection of a repricing date, and disclosure of the date is not required until the subsequent proxy filing. While the compensation committee is generally responsible for the decision to reprice, it appears that the repricing date may be selected independent of the compensation committee decision. For example, Amazon Inc calls for employees with options to exchange them for fewer new options whose strike price would be set at the lowest price the stock trades at from Jan 1 through Feb. 14, 2001, or at 85% of the Feb. 14 price if that is higher (Schroeder and Simon (2001)). Nortel announced on June 5, 2001 that they would reprice employee options stating that the exercise price of the new options will be Nortel s stock price early next year on a date to be set (Wall Street Journal, June 5, 2001). Furthermore, discussions with several executives involved in past repricings lead us to believe that while the recommendation of the compensation committee is central in the decision to reprice, the timing 5

8 may, in fact, be left to management. 3 Once the repricing date is selected, employees are given a period of time, typically 30 days or less, to decide whether to accept the offer to reprice. B. Prior Literature While we are unaware of any study that investigates systematic patterns in the timing of option repricing, there are several studies that examine the timing of other types of equity offerings. In particular, Yermack (1997), Aboody and Kasznik (2000), and Chauvin and Shenoy (2001) investigate the timing of option grants in relation to stock price activity. Yermack (1997) finds that stock price generally increases on the day of, and immediately following stock option grants, and that many option awards are made either one-day prior to, or on the earnings announcement day. Furthermore, the likelihood of a CEO receiving an option grant at a favorable time is associated with the degree of influence the CEO holds over the compensation committee. Yermack concludes that stock option grants are generally timed to precede favorable corporate news announcements. Chauvin and Shenoy (2001) document a period of declining stock price generally preceding an option grant. They conclude that through the release of unfavorable news shortly before the grant date, management attempts to achieve the lowest possible exercise price. Thus, option grants are timed to follow unfavorable news announcements. Using a sample of scheduled option grants, Aboody and Kasznik (2000) find that CEOs receiving options before the earnings announcement are significantly more likely to issue unfavorable forecasts prior to the option grant relative to those receiving their grant after the earnings announcement. Therefore, even when managers are unable to time the option awards, they maximize the value of option awards by timing voluntary disclosures around the grant period. 6

9 In other equity related studies, Seyhun (1986), Lee (1997) and Kahle (2000) find that insiders use their superior information about the future prospects of the firm to time stock purchases to occur prior to abnormal increases in stock price, and time stock sales to occur prior to abnormal declines in stock price. Loughran and Ritter (1995), using initial public offerings, and Jindra (2000), using seasoned equity offerings, provide evidence consistent with firms exploiting transitory windows of opportunity by issuing equity when, on average, they are substantially overvalued. Finally, Chalmers, Dann and Harford (2002) provide similar evidence when examining purchases of directors and officers (D&O) liability insurance occurring near the initial public offering. They find a negative relation between the amount of directors and officers insurance purchased and the future stock performance of the IPO firm suggesting that managers behave opportunistically by timing the offering date to occur when IPO shares are overvalued, while simultaneously insulating themselves from possible negative consequences. Collectively, these studies suggest that because managers possess superior information about the firm, they manage the timing of 1) stock option awards when the awards are unscheduled, 2) voluntary forecasts when the option awards are scheduled, 3) issuances of initial and seasoned equity offerings, and 4) purchases and sales of company s stock for their own portfolio. Given these results and the lack of immediate disclosure of a repricing event, we posit that managers are also likely to manage the timing of a repricing event. II. Sample Selection and Descriptive Data A. Sample selection Our repricing sample consists of 166 firms representing 236 repricing events. We use the Standard and Poor s ExecuComp Database to identify 281 repricing events involving 204 firms 7

10 and occurring over the period 1992 to We focus specifically on repricing events involving the CEO and other high level managers for whom employee-level information is available in the proxy. The sample period starts in 1992 which coincides with the SEC mandate for proxy disclosure of option repricing for named executive officers. No such requirement exists for repricing of non-executive options. We limit our study to repricings that occurred prior to 1998 to avoid confounding our results with the FASB change in the accounting for stock option repricings. 4 In addition to resetting the exercise price, firms sometimes change other features of the option such as the expiration date, the vesting period, or the number of replacement options. While these changes could conceivably affect managers wealth, there are no cases in our sample in which such changes occur independent of option repricing. For the repricing sample, we obtain repricing related information, daily stock price and return data, and financial statement information. In the year of repricing, the SEC requires proxy disclosure of the most recent repricing, as well as any other repricing occurring within the last ten years. From this 10-Year Stock Option Repricing Table, we obtain (i) the number of options repriced, (ii) the repricing date, (iii) the new exercise price, (iv) the market price on the repricing date, and (v) the old exercise price. Other information obtained from the proxy includes shares outstanding, share ownership of officers, directors and institutional investors, management compensation, and composition of the board of directors and compensation committee. We obtain daily stock price and return information from the CRSP database and financial statement data from the Compustat database. Firms are deleted from the sample if (i) the proxy is unavailable or provides insufficient information, (ii) the CRSP database does not contain return information, (iii) the firm reprices in the money options by raising the exercise price, 5 or (iv) repricing occurs for technical 8

11 reasons such as a spin-off or conversion to restricted stock. If a firm reprices more than once in a month (20 trading days), either for the same or for different executives, it is considered a single repricing event effective on the last repricing date with the last exercise price. As presented in Table I, these restrictions result in a Repricing Sample of 166 firms and 236 repricing events. From the remaining 1,663 firms in the ExecuComp database, we construct both a non-repricing matched control sample, and a sample of all remaining non-repricing firms in the database. 6 [INSERT TABLE I] B. Descriptive Analysis of the Repricing and Non-repricing Samples Carter and Lynch (2001) find that repricing firms are smaller, industry specific, and have options that are significantly out-of-the money. Therefore, we construct a matched control sample of non-repricing firms based on these characteristics. For each repricing firm, we select a non-repricing control firm that has the same four-digit SIC code, and that is most similar in size (sales and market value) and return over a one-year and two-year period. The two-year return criterion reflects the median length of time between the option grant and the repricing date for our repricing sample. If a suitable control firm cannot be identified, the potential pool is enlarged to include firms with the same three-digit SIC code. If a firm reprices several times during a single year, the same control firm is assigned for each event. However, the same control firm is never assigned to two different firms with repricings occurring in the same year. The selection criteria result in a sample of 156 control firms (216 event dates). Table II provides comparative information for the repricing sample, the matched (nonrepricing) control sample, and all other non-repricing firms included in the ExecuComp database. Consistent with the control sample selection criteria, there is no significant difference between the repricing and non-repricing control sample with respect to the repricing-year stock return or 9

12 the two-year return. Three measures of size (total sales, total assets, and total market value) are estimated using 1992 constant dollars. Using parametric tests, sales and asset are marginally greater for the control sample (p-value 0.10). However, this difference is not significant using non-parametric tests. Similarly, there is no significant difference in market value. In general, these results suggest that the matching procedure resulted in a sample of control firms that are similar to the repricing sample along the identified criteria. [INSERT TABLE II] To further ensure that the repricing and control samples are similar (except with respect to the repricing decision), we determine the extent to which the firms options are out of-themoney. Using methodology similar to Carter and Lynch (2001), we estimate that the repriced options for repricing firms are out-of-the-money an average (median) of 43.4% (42.9%). Since the control firms do not have a repricing date, we use three procedures to assign an event date: 1) the repricing date of the matched repricing firm, 2) the control firm s fiscal year-end which coincides with the end of the return interval used in identifying the control firm, and 3) the control firm s earnings announcement date immediately following the fiscal year-end. Options held by executives in the control firms are out-of-the-money an average (median) of 25.4% (29.1%) using the repricing date of the matched firm, an average (median) of 31.3% (33.0%) using the control firm s fiscal year end, and an average (median) of 28.2% (29.5%) using the earnings announcement date following the fiscal year. For both the repricing and control sample, these estimates are similar to those reported by Carter and Lynch (2001). The repricing and control samples are also compared based on measures of profitability, risk, investment opportunities, and exchange membership. We find marginally significant differences in profitability (EPS and profit margin) using non-parametric tests, but not parametric tests, and 10

13 return volatility is significantly greater (p-value 0.01) for the repricing sample relative to the non-repricing control sample. Finally, both the repricing sample and the non-repricing control sample are significantly different from the remaining ExecuComp firms along most measured dimensions. Brenner, Sundaram and Yermack (2000) similarly document that repricing firms are significantly smaller, less profitable, more risky (return volatility and debt to assets), and have a lower market to book ratio than the remaining ExecuComp firms. C. Characteristics of the Repricing Sample Table III documents the distribution of repricing events by year, as well as the frequency of repricing by sample firms. Relaxing the requirement that proxy and CRSP information be available for inclusion, we find that 204 firms repriced for a total of 281 repricing events. Panel A indicates that 149 sample firms repriced once, forty-one firms repriced twice, and fourteen firms repriced three or more times during the sample period. Consistent with Chance, Kumar, and Todd (2000) and Carter and Lynch (2003), forty-six percent of the firm in the repricing sample are from the technology and pharmaceutical industries, compared to sixteen percent for the remaining ExecuComp sample. Panel B indicates that repricing activity for firms included in the ExecuComp database increased from 1.52 percent in 1992 to 4.18 percent in While Saly (1994) suggests repricing may be optimal during a market or industry downturn, we note that the incidence of repricing actually increased during the six-year sample period over which the market increased by about 150% (e.g., Nasdaq Composite Index increased from 620 to 1565 and Dow Jones Industrial Average increased from 3200 to 8000). Thus, we examine repricing by industry, and, consistent with Brenner, et al. (2000), our untabulated results suggest that repricing is not being used to insulate managers from either market or industry factors. 11

14 [INSERT TABLE III] The mean (median) number of executives having options repriced during a single repricing event is 4.6 (4.0). The mean (median) number of years to expiration for the repriced options is 7.4 (8.1). Since most new option grants during this period were granted with a 10-year maturity, it appears that the expiration period is typically not reset. In addition, twenty-one firms (twenty-two events) condition repricing on executives accepting a reduced number of options. The exchange is usually structured such that the Black-Scholes value of the option grant is the same immediately before and after the exchange. These twenty-one firms reduced outstanding options by an average of 35.8%. Although the expected value of the options is unchanged, the exchange is still beneficial to the manager since it reduces the risk that the options will expire out-of-the-money (Hall and Murphy (2000b)). Finally, more than 85% of the repricing sample selected the current market price on the day of repricing as the new exercise price, approximately thirteen percent reset at a premium (i.e., new exercise price is higher than the market price on the repricing date but lower than the old exercise price), and the remaining reset at an exercise price lower than the current market price. 7 Consistent with Chance et al. (2000), in thirty-seven percent of the repricing events, the stock price returned to the original exercise price in less than 240 trading days (approximately one year) following the repricing event, with fifteen percent reaching the original exercise price within fifty-days. This indicates that even without repricing many of these repriced options would have been in-the-money before expiration. III. Relationship between option repricing and stock price movements 12

15 This section examines stock performance of repricing firms and provide preliminary evidence that stock options are repriced at times that are favorable to management. Given managers ability to time the repricing event, we also determine the magnitude of the benefit accruing to management from the decision to reprice. A. Stock price changes around the option repricing date We examine the average monthly return for repricing firms and document significant, negative monthly returns starting about ten months prior to the repricing date. Yet, in the repricing month we observe a positive 14% monthly return (significant at p-value 0.01). This suggests that repricing firms, on average, appear to reach their lowest stock price during the repricing month. Given this pattern, and the observation that most options are repriced with a new exercise price equal to the market price on the repricing day, we use daily returns to examine the possibility of opportunistic timing. We hypothesize that option repricing may be timed to occur before the release of favorable news. Thus, we expect to observe significant positive returns following repricing. Following the event study methodology of Dodd and Warner (1983), we estimate abnormal returns around the repricing date for each repricing event. We use the value-weighted index from the NYSE/AMEX/NASDAQ CRSP file as a measure of the market return, and an estimation period for the market model that includes both a pre-event period (days -250 to 121) and a post-event period (days +121 to +250). Thus, the test period includes the stock price decline that generally precedes the repricing event, as well as the immediate increase in stock price that follows repricing. The results are presented in Table IV. [INSERT TABLE IV] We observe significant (p-value 0.10) negative abnormal returns for each of the five days preceding the repricing date. Conversely, daily abnormal returns are positive and 13

16 significant (p-value 0.05) for each of the five days following repricing. This result suggests that good news events generally appear to follow repricing. Table IV also provides the mean daily market-adjusted return for the repricing sample with market defined as the value-weighted market index, the mean industry-adjusted return with industry return defined as the mean return for all firms in the same 4-digit SIC code, and the mean firm return absent any adjustment for the market. In all three specifications, the results are similar with positive and significant returns for each of the five days following the repricing date. 8 These results are consistent with CEOs opportunistically managing the timing of the repricing event to exploit positive price movement subsequent to repricing. 9 To provide perspective on the magnitude of the potential benefit, we accumulate these daily abnormal returns over a five-day and twenty-day window following repricing. For the repricing sample, we observe a CAR of percent for the five (trading) day period, and a CAR of percent for the twenty day period following the repricing. Both are significant (p-value 0.01). After twenty trading days, the average CAR levels off at 8 to 9 percent and is no longer significant. Figure 1 documents CARs for both the repricing and non-repricing control sample over a longer time horizon (beginning day -50 and extending to day +120). Since control firms do not have a repricing date, we again use three specifications to assign an event date: 1) the repricing date of the matched repricing firm, 2) the control firm s fiscal year-end which coincides with the return interval used in identifying the control firm, and 3) the control firm s earnings announcement date immediately following the fiscal year end. The CAR for the repricing sample indicates large decreases prior to, and significant increases immediately following the repricing date, indicating that the repricing date is generally selected as the date the stock price reaches a low. However, we do not observe this same pattern for the non-repricing control sample 14

17 regardless of the specification used. Only the first specification for the non-repricing control sample is reported in Figure 1. In general, the observed pattern for both the daily and cumulative abnormal returns is consistent with managers timing the repricing date to precede good news announcements. [INSERT FIGURE 1] B. Wealth benefits to firm executives from option repricing Since it appears management may opportunistically time repricing, we examine the magnitude of the benefit that accrues to the executives whose options are repriced. We use the Black-Scholes option model for valuing the options. While the model does have limitations for valuing these options (e.g., restrictions often limit executives ability to hedge or arbitrage their option value in the secondary market, the options are not transferable, and executives cannot take short positions in their firms stock), the disclosure requirements promulgated by the SEC (1992) and the FASB (1995) support the use of this model. The total benefit to the executives is calculated as the difference between 1) the option value on day +20 using the new exercise price and the new number of options, and 2) the option value on day 1 using the old exercise price and the number of options outstanding prior to repricing. The mean (median) increase in wealth to all named executives per repricing event is $1,028,657 ($322,646). The total benefit can be further partitioned into the benefit directly derived from resetting the exercise price (referred to as the act of repricing ) and the benefit from timing the repricing. We separately estimate the benefit of each. First, the act of repricing provides an economic benefit to the option holder on the day of repricing. We estimate this benefit as the difference in 1) the option value using the new exercise price and, when applicable, the new reduced number of options and new duration, and 2) the option value on the repricing date using the old exercise 15

18 price, old number of options and old duration. The repricing decision has a mean (median) benefit of $478,720 ($241,586). Second, if the repricing date is selected to precede anticipated stock price increases, managers accrues a further benefit from timing the repricing. We estimate this timing benefit as the difference between 1) the option value at day +20 using the new exercise price, and 2) the option value on the repricing day using the new exercise price and, when applicable, the new reduced number of options and new duration. We estimate a mean (median) benefit to the manager from this timing decision of $558,428 ($94,063). Re-estimating the timing benefit at day +5 and day +40, rather than day +20, results in a mean (median) benefit of $259,320 ($32,917) and $728,987 ($176,401), respectively. All estimates are statistically significant (p-value 0.01). The results are presented in Table V. [INSERT TABLE V] To provide some perspective with respect to the magnitude of the benefit, we compare these estimates with both the level of cash and the level of total compensation for the top five executives. We estimate that the mean (median) change in Black-Scholes value from resetting the exercise price is 25.88% (13.76%) of cash compensation, and 13.14% (5.45%) of total compensation, while the mean (median) timing benefit is 27.35% (8.08%) of cash compensation and 18.82% (3.52%) of total compensation. These estimates are statistically significant (p-value 0.01). The results suggest that both the act of repricing and the timing of the repricing date have a significant economic effect on executives wealth. The magnitude of this benefit is largely determined by the fact that management tends to reprice options as the stock price reaches a minimum, and just prior to a period with abnormal positive returns. It seems unlikely that this pattern consistently occurs by chance. Therefore, in the next section we specifically address the manner in which managers may time repricing around news releases for personal benefit. 16

19 IV. Stock Option Repricing around Quarterly Earnings Announcements We examine two potential strategies for timing repricing relative to news announcements; repricing before good news, or repricing following bad news. While the results in Section III suggest managers may time repricing to precede good news, they are silent regarding the possibility of timing repricing relative to bad news announcements. Therefore, this section examines the choice of the repricing date and its relationship to the mandated quarterly earnings announcement. Using PR newswires and the Dow Jones News Retrieval Service, we examine news reports that precede and those that follow the repricing date. These announcements primarily relate to new product introductions and innovations, new alliances, downsizing and/or restructuring, management and analyst forecasts, patent approvals and denials, as well as quarterly earnings announcements and other indicators of firm performance. Given the significant potential for error in classifying the markets expectation of the news contained in these announcements, we focus solely on the quarterly earnings announcements to further test the hypothesis that option repricing is timed relative to corporate news announcements. Earnings announcements not only provide a disclosure event that is required for all firms, but a disclosure in which management is presumed to have more private information about both the date and the content of the earnings announcement. We predict that repricing is more likely to precede a positive earnings announcement, providing managers the economic benefits associated with the related stock price increase. In the case of a negative earnings announcement, management may delay repricing to follow the announcement in order to obtain a lower exercise price on resetting. Since, observing stock price 17

20 changes around the earnings announcements allows us to capture the market s perception regarding the quality of the news (good or bad), we can test the relationship between the timing of the repricing event and the quality of the news. For each repricing event, we identify the earnings announcement date occurring immediately prior to and immediately following the repricing date. Earnings announcements that occur during the weekend or on a holiday are classified as occurring on the next business day. Figure 2 presents the frequency distribution of the repricing dates relative to the nearest quarterly earnings announcement date. Repricing dates are distributed normally around the earnings announcement date. The most frequent repricing date is the second day following the earnings announcement (5.19%), while the second most frequent day (4.76%) occurs two days prior to the earnings announcement. We further partition the repricing sample based on whether the repricing occurs prior to the earnings announcement (pre-announcement repricers) or following the earnings announcement (post-announcement repricers). Three repricing events occurring on the earnings announcement date are not included in either group. [INSERT FIGURE 2] A. Earnings announcement CAR For both pre-announcement and post-announcement repricers, we estimate the earnings announcement CAR as the three-day cumulative abnormal return centered on the earnings announcement date (i.e. days 1, 0, and +1 with earnings announcement date as day 0). We use the estimation period and the value-weight market index used in the previous section except day 0 is defined as the earnings announcement date rather than the repricing date. For pre-announcement repricers, we observe a mean earnings announcement CAR of percent (p-value 0.01) when options are repriced in the two-day window occurring before the 18

21 earnings announcement. We also examine several other windows and regardless of the window selected, we document a positive earnings announcement CAR for pre-announcement repricers. These results, reported in Table VI, provide further support that managers choose a repricing date prior to the release of favorable earnings news. [INSERT TABLE VI] Examining the earnings announcement CAR for post-announcement repricers provides our first evidence that firms with negative earnings surprises also time repricing. For postannouncement repricers, the mean earnings announcement CAR is 7.76 percent (p-value 0.01) when options are repriced in the two-day period following the earnings announcement. Again, regardless of the window selected, we observe a significant negative earnings announcement CAR for post-announcement repricers. This result suggests that managers, possessing superior information regarding the earnings announcement, systematically delay repricing to follow the release of unfavorable earnings news. Overall, tests of pre- and postannouncement repricers indicate that managers anticipating favorable earnings reports, reprice prior to the expected stock price increase to increase their benefit from the decision to reprice, while managers anticipating bad news reprice following the expected price decline to obtain a lower exercise price. B. Post-repricing CAR The observation that repricing is timed differently depending on the content of the earnings announcement also suggests systematic differences in stock price movement around the repricing date for pre- and post-announcement repricers. Therefore, we separately estimate the post-repricing CAR for pre- and post-announcement repricers. We hypothesize a significant positive post-repricing CAR when repricing occurs prior to the earnings announcement (pre- 19

22 announcement repricers) implying that managers time the repricing to precede the release of good news. On the other hand, when repricing follows the earnings announcement (postannouncement repricers), we do not hypothesize abnormal price activity since there is no expectation of a systematic release of good or bad news following the repricing event. Combining these predictions, we further predict that the post-repricing CAR for preannouncement repricers will be significantly greater than that of post-announcement repricers. This comparison is made using both a sample of 63 firms that repriced within five days prior to or following the earnings announcement, and a sample of 105 firms that repriced within twelve days prior to or following the earnings announcement. Testing only those repricings in close proximity to the earnings announcement date reduces the effect of other confounding news events. Using the first sample (five-day window), in Table VII we document a significant positive post-repricing CAR for pre-announcement repricers with a mean (median) of (+0.100). For post-announcement repricers, we do not observe a significant post-repricing CAR. Furthermore, as hypothesized the post-repricing CAR for pre-announcement repricers is significantly greater than that of post-announcement repricers (p-value < 0.01). For the second sample (twelve-day window), the post-repricing CAR is positive and significant for both the preand post-announcement repricers. However, despite the significant CAR for post-announcement repricers, the post-repricing CAR for pre-announcement repricers is significantly greater than that of post-announcement repricers. These results are generally consistent with the prediction that repricing is timed to precede good news, and to follow bad news announcement. [INSERT TABLE VII] C. Stock returns and other news announcements 20

23 To this point, we have focused on the earnings announcement as a specific news release that firms can time repricings around. However, it is reasonable to assume that they also time in concert with other anticipated announcements by the firm. Therefore, we delete all observations where repricing occurs within five days of the earnings announcements, resulting in a sample of 127 repricing events (55 pre-announcement repricers and 72 post-announcement repricers). This restriction reduces the effect of the earnings announcement on the returns associated with repricing. Using the methodology in Figure 1, we re-estimate daily CAR over the longer time horizon for the partitioned sample. The results are presented in Figure 3. For pre-announcement repricers, we observe a V-shaped curve, similar to that documented in Figure 1 for all repricing firms suggesting that pre-announcement repricers likely benefit from good news events other than the earnings announcements. The post-announcement repricers, on the other hand, exhibit a more severe decline prior to the repricing date, and we do not observe the same positive CAR following repricing. Thus, post-announcement repricers, in order to obtain a lower exercise price, simply impound the negative stock reaction associated with negative announcements that occur prior to the repricing date. V. Additional tests Corporate Governance In this section, we investigate determinants of both the decision to reprice, as well as the timing of the event. Similar to previous repricing studies that focus on the repricing decision, we examine the relationship between corporate governance and the structure of managerial compensation. We further investigate whether the benefit that accrues to management from timing repricing is related to corporate governance. 21

24 A. Influence of Top Management and Executive Option Repricing The board of directors is responsible for the decision (not necessarily the timing) to reprice executive stock options. Board membership generally includes insiders who are officers and executives of the firm, and those non-employee directors who are affiliated through a significant business relationship or interlocking directorship. Consistent with prior literature, those directors on the board not classified as insiders are classified as outsiders (Newman (2000), Newman and Mozes (1999), Byrd and Hickman (1992), and Baysinger and Butler (1985)). If management is influential in the decision to reprice, or if repricing is opportunistically timed to benefit management, then we expect inside members of the board and/or the CEO to have significant influence over the Board of Directors. The most obvious opportunities are when (i) the CEO is also the Chairman of the Board, (ii) the CEO is a member of the compensation committee, (iii) an insider is a member of the compensation committee, or (iv) there is a high proportion of insiders on the board of directors. Chance, Kumar and Todd (2000) find that the proportion of insiders on the board increases the likelihood of repricing, and Brenner, Sundaram and Yermack (2000) document that the presence of an insider on the compensation committee increases the likelihood of repricing. However, Carter and Lynch (2001) find neither of these factors significant in explaining the likelihood of repricing. We extend these analyses by examining situations in which we expect insiders to have reduced influence on the compensation committee. Similar to Yermack (1997), we examine situations in which the compensation committee includes a non-executive chairman of the board, and when the compensation committee includes an outside director who is a major (>5%) stockholder. In addition, we investigate the role of institutional investors since they have been among the most vocal critics of option repricing (Schism and Lublin 1998) and prior research 22

25 indicates that institutions provide a monitoring role in corporate governance (Schleifer and Vishny (1997)). However, neither Chidambaran and Prabhala (2000) nor Carter and Lynch (2001) find evidence that institutional ownership is related to the decision to reprice. Finally, we include a proxy for compensation structure. We define compensation structure as the cash proportion of total compensation for the top five executives identified in the proxy. Cash compensation includes salary, bonus, and other cash payments, while total compensation is measured as the sum of cash compensation, the total value of restricted stock, the Black-Scholes value of stock options granted, long-term incentives, as well as any other payouts occurring in the repricing year. Mehran (1995), and Jensen and Murphy (1990b), among others, document that the percentage of equity compensation is correlated with firm performance. However, in a period during which the options are underwater, managers with greater proportions of equity compensation have greater risk due to the increased probability that these options will expire out-of-the-money (Hall and Murphy (2000b)), and therefore have a greater incentive to undertake risky projects to raise the stock price (Gilson and Vetsuypens (1993) and Lambert, Larcker and Verrecchia (1991)). One mechanism to reduce this compensation risk is repricing underwater options. Hence, we evaluate whether greater proportions of equity in managerial compensation is positively related to the decision to reprice. B. Decision to Reprice Univariate Comparison To provide insight into the type of firms that reprice, Table VIII documents corporate governance characteristics for the repricing and non-repricing control sample. These summary statistics relate to institutional and insider ownership, compensation structure, board membership, and compensation committee membership. Institutional ownership is significantly lower (p-value 0.05) for repricing firms relative to non-repricing firms. The cash proportion 23

26 of total pay is also significantly lower for repricing firms relative to non-repricing firms with a mean of 60.0% and 70.2%, respectively. 10 However, there is no significant difference with respect to insider ownership. [INSERT TABLE VIII] Repricing firms also exhibit differences with respect to board membership. With the exception of one non-repricing control firm, the boards of all sample firms include at least one insider. Insiders comprise 44.8% of the total board membership for the repricing sample, compared to 36.8% for the control sample. This difference is statistically significant (p-values 0.01). Corporate governance is also presumed to be weak when the CEO also serves as the chairman of the board. However, we are unable to document differences between the samples along this dimension. Since the compensation committee is directly responsible for the decision to reprice, we also examine compensation committee membership. In 44.6% of the repricing observations, at least one insider served on the compensation committee compared to 25.2% for the control sample. This difference is statistically significant (p-values 0.01). However, among firms with at least one insider on the compensation committee, the mean ratio of the number insiders to the total number of compensation committee members is 45.6% for the repricing sample compared to 39.5% for the non-repricing sample. This is not significantly different at conventional levels. Together, this suggests that, in the decision to reprice, the presence of an insider on the compensation committee is a relatively more important governance characteristic than the number of insiders represented on the committee. Furthermore, a CEO serving as a member of the compensation committee does not appear to increase the probability of repricing. Conversely, corporate governance is expected to be strong, or managerial influence weak, when an outside Chairman of the Board is a member of the compensation committee, or 24

27 when members of compensation committee are significant shareholders in the company. We do not document significant differences between repricing firms and non-repricing firms along these two dimensions. Since one quarter of the repricing firms experienced multiple repricings during the sample period, we also compare the governance characteristics of firms with multiple repricings to those having a single repricing event. We find no significant difference between these groups. Overall, these univariate tests suggest that the decision to reprice is positively related to the percentage of board seats held by insiders and the presence of at least one insider on the compensation committee. It is negatively related to the percentage of shares held by institutions and the proportion of cash compensation to total compensation. C. Decision to Reprice Multivariate Analysis To further examine the role of corporate governance in the repricing decision, we combine the repricing and control samples and estimate a logit regression. The results are presented in Table IX. In Model 1, we regress variables representing the determinants of corporate governance on an indicator variable (0,1) that identifies whether the firm repriced. The results indicate that while the percentage of insiders on the compensation committee is a significant (p-values 0.01) indicator of repricing, the percentage of insiders on the board is not. Brenner et al. (2000) and Chance et al. (2000) test similar models, however each study includes only one of these governance variables and each finds the included variable to be a significant predictor of repricing. Therefore, for comparison purposes, we re-estimate the logit model including, along with the remaining governance variables, only the variable representing the percentage of insiders on the board in Model 2 (similar to Brenner, et al. (2000)), and only the percentage of insiders on the compensation committee in Model 3 (similar to Chance et al. 25