In our paper, One Dollar CEO Salaries: An Empirical Examination of the Determinants and Consequences, which was recently made publicly available on SSRN, we examine a sample of 278 CEO-firm-years (87 firms and 88 CEOs) between 1995 and 2009 where the CEO’s salary is $1. First, we analyze the determinants of a firm’s (or CEO’s) decision to reduce annual salary to $1. We explore characteristics related to the firm, stock, market following, CEO and board of directors, and find that larger firms, higher growth firms, more heavily-traded firms, firms with greater leverage, and firms with lower lagged return-on-assets are more likely to have $1 CEO salaries. We also find positive associations with retail investor ownership in the firm, CEO ownership in the firm, and when a CEO is also the chairperson of the board. Overall, these findings suggest that relatively more powerful CEOs at visible firms with high retail investor ownership are more likely to set a $1 salary, particularly following poor firm performance.
Second, we compare all the components of CEOs’ total annual compensation before and after the decision to reduce salary to $1 to provide evidence on whether the salary reduction is part of a substantive effort to reduce overall compensation. Along with a reduction in salary, there is a statistically significant decrease in the mean and median values of CEO bonus. However, we do not find a significant decline in total annual compensation; instead, total compensation appears to increase for roughly half of the CEOs, due primarily to increases in stock option compensation. When we partition the cases by whether the CEO is a continuing or new CEO, we find similar results, again suggesting that only for half of the cases is a reduction in salary to $1 accompanied by a substantive reduction in total compensation.
Third, we examine the short-window market reaction to a firm’s announcement of a CEO’s $1 salary. Using an event study methodology, we find significant abnormal market reactions (i.e., abnormal return variance and trading volume) on the announcement date, consistent with the news having information content for the stock market. We also find that the reaction is stronger for firms making the announcement in a press release issued prior to the proxy statement. When we examine signed abnormal returns, we find that the mean reaction is close to zero because the proportion of announcements considered positive or negative is fairly symmetrically distributed. Upon further examination, we find that the abnormal returns tend to be negative when the CEO taking the $1 salary has longer tenure and greater ownership in the firm, suggesting that the market perceives the announcement negatively for the relatively more powerful CEOs.
Finally, we examine long-window changes in future firm performance, market following, and stock characteristics associated with a firm’s decision to set a $1 CEO salary. We first conduct a difference-in-differences test between our entire sample of $1 CEO salary firm-years and a propensity-score (p-score) matched control sample (Rosenbaum, 2005; Rosenbaum and Rubin, 1983) of non-$1 CEO salary firm-years. This procedure enables us to match treatment ($1 salary) and control firms on 14 observable firm, stock, market following, CEO and board characteristics. We find some evidence that $1 CEO salary firms, on average, tend to: 1) reduce capital spending; 2) experience a rebound in return-on-assets; 3) gain one additional analyst; and 4) have higher trading volume in the following year, as compared to the control group of firms.
We then examine differences just within the sample of $1 CEO salary firms by partitioning firms on whether the CEO’s total annual compensation decreases or increases in conjunction with the new $1 salary. We do so because a subset of the CEOs in our sample not only reduced their salaries to $1, but also reduced all components of their compensation as a sign of their commitment to stay with the firm until performance improves. We find that for the subset of firms where the CEO’s total annual compensation decreases, capital expenditures decrease, the percentage of retail investors increases, and size-adjusted-returns (SARs) are significantly higher in the following year, relative to the subset of firms where the CEO’s total annual compensation increases. These findings suggest that there are several favorable long-window outcomes associated with a $1 CEO salary decision, and when coupled with substantive action (i.e., real reduction in total CEO compensation and firm spending), the market also responds favorably over the next year.
Our study contributes to the executive compensation literature by systematically investigating a distinctive, high-profile compensation practice that captures the attention of managers, investors, regulators, and the business press. We provide evidence on the types of firms and CEOs who are likely to initiate a massive reduction in salary, both with and without a corresponding reduction in total annual compensation. We also contribute to the managerial symbolism literature by showing that a $1 (or less) CEO salary is another potential symbolic tool used by management that may or may not be coupled with substantive effects on total compensation. Finally, our analyses for the short- and long-window market reactions to the announcement should be of interest to the many stakeholders debating the merits of setting a CEO’s salary to $1. Our findings suggest that the market does not react to all announcements of a $1 CEO salary the same; positive reactions and outcomes tend to occur for newer, less powerful CEOs and when the $1 salary is accompanied by real reductions in total compensation.
The full paper is available for download here.