The two sides of CEO pay injustice A power law conceptualization of CEO over and underpayment

Purpose - The purpose of this study was to examine the extent to which chief executive officers (CEOs) deserve the pay they receive both in terms of over and underpayment. Design/methodology/approach - Rather than using the traditional normal distribution view in which CEO performance clusters around the mean with relatively little variance, the authors adopt a novel power law approach. They studied 22 industries and N = 4,158 CEO-firm combinations for analyses based on Tobin's Q and N = 5,091 for analyses based on return on assets. Regarding compensation, they measured the CEO distribution based on total compensation and three components of CEO total pay: salary, bonus, and value of options exercised. Findings - In total, 86 percent of CEO performance and 91 percent of CEO pay distributions fit a power law better than a normal distribution, indicating that a minority of CEOs are producing top value for their firms (i.e. CEO performance) and a minority of CEOs are appropriating top value for themselves (i.e. CEO pay). But, the authors also found little overlap between CEOs who are the top performers and CEOs who are the top earners. Implications - The findings shed new light on CEO pay deservingness by using a novel conceptual and methodological lens that highlights systematic over and underpayment. Results suggest a violation of distributive justice and offer little support for agency theory's efficient contracting hypothesis, which have important implications for agency theory, equity theory, justice theory, and agent risk sharing and agent risk bearing theories. Practical implications - Results highlight erroneous practices when trying to benchmark CEO pay based on average levels of performance in an industry because the typical approach to CEO compensation based on averages significantly underpays stars and overpays average performers. Originality/value - Results offer new insights on the extent of over and underpayment. The findings uncover an extremely large non-overlap between the top earning and top performing CEOs and to an extent far greater in magnitude than previously suggested.