Why We Need to Rethink CEO Compensation Practices

A 2018 report by The Wall Street Journal showed that in 2017 a quarter of CEOs among the 133 largest U.S. firms saw their pay go up by approximately 25%. The average increase for the entire sample was around 10%. In sharp contrast, wages for employees at the same firms barely showed an increase. This obvious disparity in wage increases prompts a re-examination of that age old, much-debated question:  do CEOs deserve the pay they receive? According to a recent study by Herman Aguinis, Luis R. Gomez-Mejia, Geoffrey P. Martin, Harry Joo, and Ernest O’Boyle, many don’t.

Many people argue that CEOs are overpaid and their high salaries are not justified, especially as CEO pay continues to increase in comparison with wage increases for everyone else. How does this position compare with the economic reality? To understand if CEOs really deserve their compensation packages, we need to understand how CEO pay and performance are linked and analyse CEO pay and CEO performance data.  Are some CEOs paid vastly more than the rest?  Do some CEOs vastly outperform the rest? And how much do these two groups overlap?

To get to the heart of these questions, we used a novel and precise methodological approach to understand the distribution of CEO pay and CEO performance across our CEO sample. We analysed the pay and performance of over 4,000 CEOs from publicly traded firms across 22 industries, including agriculture, forestry, mining, air travel, banking; and basic manufacturing. Our results on both CEO pay and performance looked like the “power law” distribution shown in Figure 1, with many CEOs all the way to the right of the distribution – to the right of mean pay and mean performance. However, to the left of the distribution, some CEOs exhibit extreme pay and extreme performance values. This is in comparison to a “normal” bell curve distribution, where most scores cluster around the middle. This demonstrates that the highest paid CEOs are paid multiple times more than the rest of the CEO cohort, and the highest performing CEOs outperform the rest by a large margin.

Figure 1: the distribution of CEO pay and performance

But here is our most important finding. When we examined the CEO pay and CEO performance distributions side by side, we found little overlap between top earners and top performers. None of the top 1% performing CEOs are among the top 1% paid CEOs regarding salary or bonus. Only 5% of the top 5% performing CEOs are also among the top 5% paid CEOs and only 20% of the top 5% performing CEOs are also among the top 5% paid CEOs.  These results indicate that (1) CEOs at the top of the performance distribution create vastly more value than those at succeedingly lower levels of the performance distribution; (2) CEOs at the top of the pay distribution are remunerated far more than those at succeedingly lower levels of the pay distribution; and (3) there is minimal overlap between the CEO pay and CEO performance power law distributions; because of the power law distributions shown above, the extent of the mismatch is amplified.

In other words, the CEOs who create disproportionately large amounts of value are in a different group from the CEOs receiving disproportionately large amounts of compensation, and vice versa. These results hold across all pay components (base salary, bonuses, and long term income) and almost all industries. 

What implications does our research have for executive compensation, governance, and HR practices? A typical approach many companies follow for negotiating and designing CEO compensation contracts is for the board’s pay committee (usually with the help of compensation consultants) to ascertain the average CEO pay for a particular performance cohort in a given industry. The committee then uses that figure as a reference to estimate CEO capability relative to their peer group and sets CEO pay accordingly. However, as we have shown, CEO pay does not have a normal distribution. There is a huge range in CEO performance, with those in the top quartile delivering results many multiples better than the average. This approach to CEO compensation based on averages results in significantly underpaid CEO stars and overpaid average performers. No wonder CEO poaching is a frequent phenomenon and the typical CEO lasts just a few years on the job. After all, information on CEO performance is readily accessible to recruiters and headhunters, making it easy to identify the individuals producing top results.

Professional thinking needs to shift towards the power law view.  Headhunting firms that adopt the power law view, rather than a normal distribution mindset, will be able to identify CEOs who are not the highest earners but are among the top performers. Headhunters will be able to proactively approach these CEOs and tempt them with an offer to move to another firm. Similarly, firms should examine CEO pay and CEO performance based on a power law perspective to prevent dysfunctional turnover (i.e., the departure of a high performing CEO). From a public relations perspective, our results suggest that firms who embrace a power law mindset can better anticipate and prevent potential shareholder disapproval and unwanted media attention. Just as important as identifying top performing CEOs is identifying top earning CEOs who are not creating commensurate results.

This blog post is based on the authors' research, which originally appeared in Management Research  and is published under a CC BY-NC-ND 2.0 license.

If you would like to explore this issue further, the following authors have taken the time to write commentaries published within the same issue analyzing this new research and discussing its impact: Martin J. Conyon, Donald C. Hambrick, Michael A. Hitt, Katalin Takacs Haynes, Adam J. Wowak, Michael J. Mannor, Patrick M. Wright, Anthony J. Nyberg, Robert M. Wiseman, Hadi Faqihi, Albert A. Cannella, Jr., Valerie Sy, Gerald E. Ledford, Edward E. Lawler, James P. Walsh, B. Joseph White, and Jeffrey R. Edwards. The issue is available as free access until June 10^th. 2018.

About the authors:

Herman Aguinis is the Avram Tucker Distinguished Scholar and Professor of Management at George Washington University School of Business. His interdisciplinary research addresses human capital acquisition, development, deployment, and research methods and analysis. He has published over 140 journal articles, five books and delivered over 300 presentations and keynote addresses on all continents except Antarctica. He has been elected to serve on the track for the Presidency of the Academy of Management, the preeminent professional association for over 21,000 management and organization scholars including about 21,000 spanning more than 120 countries. Find Herman on twitter @HermanAguinis, on LinkedIn or via e-mail.

Luis R. Gomez-Mejia is Regents Professor and Weatherup/Overby Chair of Management at Arizona State University. He has been named in several studies as one of the top management scholars in the world and published over 200 articles and 15 books in management. His recent research interests concern strategic decision making with a particular focus on family firms.  In 2015, he was listed in Thomson Reuters Highly Cited Researchers, ranking in the top 1% of researchers in economics and business worldwide.

Geoffrey P. Martin is an associate professor of strategy at the Melbourne Business School. Prior to completing his PhD at IE Business School in Madrid and his MBA at Melbourne Business School, Geoff held a number of senior strategy and operational risk positions at Deloitte, GT Global, TXU, IAMCO, Egg Bank and Credit Suisse, where he was Vice President. His research interests cover strategic decision making, executive compensation, executive risk taking and corporate governance.

Ernest O’Boyle is an associate professor of management and organizations at the Tippie College Business, University of Iowa. His research interests include star performance, counterproductive work behaviour, research methods, and ethical issues surrounding publication practices. He is the recipient of the Academy of Management Early Career Awards for both the Research Methods Division and Human Resources Division.

Harry Joo is an assistant professor at the School of Business Administration, University of Dayton. His research interests include star/high performers, research methods, and bridging the science-practice divide.

With world-leading experts contributing to our Real World Impact blog, you can be sure to find thought-provoking content at www.emeraldgrouppublishing.com/realworldimpact.htm