How Market Feedback Affects CEO Pay

Stock prices affect various corporate decisions such as the amount of CEO compensation, as emphasized in traditional pay-for-performance studies, and capital investment. However, an unanswered question is whether changes in firms’ stock price in a short window around events related to their corporate governance matter for the compensation of their CEOs. In a new paper, we analyze whether these short-term price changes factor into compensation policies more than suggested by the traditional pay-for-performance argument (which is normally based on stock returns over one year or longer). We refer to decision makers’ learning from changes in stock prices as the market feedback effect and examine whether this effect applies to boards when they decide on CEO compensation.

Using stock announcement returns (derived from the stock price changes in three days around governance events) as a proxy for shareholder opinions allows us to evaluate whether boards actively infer shareholders’ attitudes toward CEO pay from stock prices. If such learning exists, we posit that it may cause boards to incorporate shareholder attitudes into the subsequent setting of CEO pay. In so doing, directors may build a reputation for being shareholder-friendly and have better career outcomes. This prediction is consistent with the survey evidence that some directors offered CEOs less pay to avoid the risk of investor opposition, highlighting the importance of shareholder opinion to directors’ decision making.

The governance events that we focus on are say-on-pay (SOP) voting failures announced by firms’ peers in compensation. These failures suggest that peer firms’ CEO pay packages are inappropriate. More importantly, the failures indicate that focal firms’ CEO pay design may also be problematic (e.g., focal firms opportunistically choose firms with overpaid CEOs as their peers in compensation or adopt similarly inappropriate pay policies). Hence, a peer firm’s SOP voting failure is likely to invite the focal firm’s investors to question whether their own CEO is overpaid and to reevaluate the CEO’s compensation plan.

We focus on compensation peers’ SOP voting for two reasons. First, directors typically play a role in CEO-pay benchmarking, identifying the firms to use as benchmarks. Therefore, directors care about investors’ opinion on the chosen compensation peers. Second, we focus on SOP voting failures, rather than other types of governance events, because they are salient governance events that investors have paid close attention to over the past decade. The SEC introduced non-binding SOP voting to provide shareholders with greater power to express their views regarding executive-compensation issues with implementation of the Dodd-Frank Act in 2011.

Directors’ process for making pay decisions is a “black box” to researchers because it tends to consider multiple pieces of information. We shed light on this black box and propose that shareholders express their opinion in focal firms’ prices following compensation peers’ SOP events, and directors respond to such price movement by revising CEO pay (i.e., creating a price-feedback effect for the board’s compensation decisions).

Our empirical analyses have two steps. First, to verify that compensation peers’ SOP failures indeed catch attention of focal firms’ investors, we show that the focal firms’ prices respond negatively to compensation peers’ announcements of SOP failures. This result suggests that focal firms’ shareholders are attentive to compensation peer firms’ inappropriate pay-setting and express dissatisfaction with focal firms’ choice of an inappropriate benchmark and the resulting executive pay. In particular, we extract compensation peers from the Incentive Lab database and each firm’s SOP votes from Semler Brossy’s annual say-on-pay reports. We calculate the focal firm’s three-day cumulative abnormal market-adjusted return (CAR) around its SOP-failed peer’s annual shareholders’ meeting date. We find that focal firms are more likely to experience a lower CAR when their compensation peers’ SOP failure is severe. Importantly, we find that the R2 of the prediction model (regressing focal firms’ CAR on peers’ voting outcomes and other control variables) is low (13.5 percent), suggesting that investors may have information about the focal firm’s compensation that compensation peers’ voting failure or other characteristics do not fully convey. Hence, boards may learn additional information from CAR.

The second and primary research question is whether focal firms’ directors recognize that a price drop in the three-day window (i.e., around peers’ SOP failure announcements) indicates investor dissatisfaction, consequently leading them to cut CEO compensation. We find that focal firms that experience a more negative CAR subsequently show more significant CEO pay decreases than firms with a less negative CAR. This result is consistent with a feedback effect. Specifically, we find that focal firms experiencing a negative market reaction in the event of their peers’ voting failure reduce CEO pay by more than 12.46 percent (around $1.07 million) in the subsequent year, relative to those with no adverse market-spillover effect. The effect on CEO pay is stronger when the board of directors is more powerful, when proxy advisers hold a negative view of the CEO’s pay, and when the hired compensation consultant is less reputable.

Investors should reward (punish) directors who learn information about investors’ opinions of CEO pay from stock prices. Among firms that have experienced adverse price reactions around peers’ bad governance event, the focal firm board’s decision to cut CEO pay in the following year could increase shareholders’ satisfaction and hence lead to more favorable shareholder votes in elections for incumbent directors. We find that directors who cut their CEO’s pay following the price drop receive more voting support from investors than other directors, supporting the notion that investors reward directors who demonstrate shareholder-friendly behavior and monitor firm management.

Overall, our study documents a market-spillover effect within the SOP voting scenario. We further provide evidence in support of the learning-from-prices channel. The findings of our study contribute to the literature on the impact of the feedback effect in the stock market and are useful to regulators in evaluating the effectiveness of the rule that mandates periodic SOP voting.

This post comes from C.S. Agnes Cheng at the University of Oklahoma’s Price College of Business, Iftekhar Hasan at Fordham University, Feng Tang at the University of Macau, and Jing Xie at the University of Macau and Hong Kong Polytechnic University’s School of Accounting and Finance. It is based on their recent paper “Market Feedback Effect on CEO Pay: Evidence from Peers’ Say-on-Pay Voting Failures” available here.