Print
E-MailC. Edward Fee is Professor of Finance at Tulane University Freeman School of Business; Zhi Li is Assistant Professor of Finance at Champan University George L. Argyros School of Business and Economics; and Qiyuan Peng is Assistant Professor of Finance at the University of Dayton. This post is based on their recent paper, forthcoming in Journal of Accounting and Economics.
Related research from the Program on Corporate Governance includes Stealth Compensation Via Retirement Benefits by Lucian A. Bebchuk and Jesse M. Fried.
Performance-based stock grant is an increasingly popular form of incentive pay for public firm CEOs in U.S. Under these grants, executives are expected to receive different levels of stock payments (“threshold,” “target,” or “maximum”), contingent on the firm’s meeting pre-specified hurdles by the end of the performance evaluation period. In 2006, the SEC announced new disclosure rules that require firms to report “unearned shares” from outstanding performance-based stock grants in their proxy statement. Unearned shares are the number of shares that executives are expected to receive conditional on whether the firm meets predetermined performance hurdles by the end of the evaluation period.
In our paper, Hidden Gems: Do Market Participants Respond to Performance Expectations Revealed in Compensation Disclosures?, forthcoming in the Journal of Accounting and Economics, we examine whether the disclosed “unearned shares” provide new information about a firm’s future performance. We believe the new disclosure contains forward-looking information for two reasons. First, firms often cite performance expectations over the evaluation period to justify the reported unearned shares. Second, past literature has shown that firms choose specific performance measures that reflect their strategic priorities. These performance criteria, which are often not fully disclosed to the public, may capture information related to CEO actions and firm performance over the long run. However, the disclosure might not be informative. For example, the compensation committee and other corporate insiders might not be able to correctly forecast future firm performance and plan payouts. Or the firm could be unwilling to truthfully reveal inside information. Or the performance-based grants might be poorly designed to be informative, such as the performance hurdles are set at unreasonably high (low) level or the performance measures used are unrelated to firm value. Hence, whether the disclosure of unearned shares is informative is an empirical question.
Using a sample of large U.S. public firms from 2006 to 2013, we identify 5,214 (37.20%) firm-years with reported unearned shares from outstanding performance stock grants. We classify them into three groups based on the reported level: Threshold, Target, and Maximum. There is significant variation in the reported level: 666 (12.77%) firm-years are in the Threshold group; 3,203 (61.4%) in the Target group; and 1,345 (25.80%) in the Maximum group. Firm-years where CEOs do not have outstanding performance stocks (8,803) are classified as “No Signal”, as no information can be inferred from them.
We first examine firm performance after compensation disclosure across the three disclosure groups. Firms in the Maximum group have significantly higher ROA, firm’s Q, sales growth, and profit margin than those in the Threshold group over the next two years after the disclosure. Similarly, firms in the Target group outperform firms in the Threshold group, but by a lower magnitude. Firms in the Maximum (Threshold) disclosure group also have superior (inferior) future operating performance compared with firms that do not have outstanding performance-based stocks (the No Signal group). Thus, it is unlikely that executives expect Maximum (Threshold) level of payouts simply because their firms set performance hurdles too low (high). Further, there is no evidence that the superior future performance of the Maximum (Target) group is driven by earnings management.
It is possible that the forward-looking information conveyed in unearned shares has already been covered by other information sources. We control for five potential information channels based on the literature: post-disclosure management earnings guidance raise; disclosure year positive 10-K sentiment based on textual analysis; net insider selling activities; unexplained CEO cash compensation; and analysts’ earnings forecast errors in the disclosure year. We find that the disclosed level of unearned shares is still highly correlated to firms’ future performances. In summary, unearned shares disclosure contains unique information that is not captured by a firm’s current performance, observable firm and CEO characteristics, or other known information channels.
We next examine whether investors could timely incorporate the information into asset prices. We find that there is no difference in market reaction around proxy filing dates between the Maximum and Threshold firms. However, investors are positively surprised when firms in the Maximum group announce earnings one year after the unearned shares disclosure. Moreover, firms in the Maximum group report earnings that significantly beat analysts’ forecasts over the next four quarters. In contrast, firms in the Threshold group have negative, though not significant, post-disclosure earnings surprises. These findings suggest that investors, including sophisticated analysts, having underestimated the correlation between disclosed unearned shares and future firm performance, are later surprised when actual performance is reported.
Lastly, we examine firms’ long-run abnormal returns following the compensation disclosure. We find that firms in the Maximum disclosure group experience significantly positive abnormal stock returns the following year, while those in the Threshold group experience significantly negative abnormal returns. The results hold after controlling for traditional risk factors, momentum, profitability, post-earnings- announcement drift (PEAD), and information from other public channels. A long–short portfolio that invests in firms in the Maximum group and shorts those in the Threshold group could earn significantly positive post-disclosure abnormal returns.
Given the well documented effect of long-run abnormal return driven by PEAD, we separate the firms into subgroups based on above- or below-median earnings surprises in the disclosure year. Within the Maximum and Target groups, only firms with below-median earnings surprises experience significant positive abnormal returns after disclosure; those with above-median earnings surprises earn risk-adjusted returns on average. Thus, the documented long-run return pattern is not driven by earnings drift. Instead, investors underreact to positive information embedded in compensation disclosure when it differs from the more visible earnings news.
Our paper shows that disclosed unearned shares from performance-based stock grants reveals valid forward-looking information about a firm. This new disclosure was mandated by a 2006 SEC rule change that aimed to raise the standards of compensation disclosure. Our findings suggest that under the enhanced disclosure rule, firms, on average, truthfully reveal new information to the public and that investors could improve market efficiency if they promptly incorporated such information into asset prices.
Our paper provides further evidence on the usefulness of compensation disclosure. As boards use internal signals to evaluate executives, compensation-related disclosure contains incremental information on firm performance that is beyond what is conveyed in financial statements. However, investors have limited ability to process public information that is difficult to process, and their underreaction to this information leads to predictable abnormal stock returns. Due to limited attention, investors might not look for, or find, performance-related information in a place as unlikely as compensation disclosures.
Our paper further contributes to the literature on the importance of corporate voluntary disclosure. Researchers and practitioners have expressed concern that, when the cost of misreporting is low, firms might strategically use disclosure to manipulate the market. As a result, investors are reluctant to take voluntary disclosure at face value. Our results, however, show that firms on average truthfully reveal their internal performance expectations when disclosing complex details of executive incentive pay. This evidence complements earlier findings that when firms voluntarily provide information that is hard for outsiders to verify, the disclosure is generally credible.
The complete paper is available for download here.
