How to determine executive compensation remains a hot issue, with a recent Institutional Shareholder Services survey showing executive pay is the top governance issue for institutional investors. Meanwhile, a research paper from the University of Delaware’s John L. Weinberg Center for Corporate Governance has focused debate on the standard practice of benchmarking CEO compensation by looking at a group of similar companies, which it argues leads to escalating pay levels.
Charles Elson, chair of the Weinberg Center and co-author of the study, says selecting the companies to include in a peer group invites manipulation of the results, while the fact that compensation is typically set at the 50th, 70th or 90th percentile level found in the peer group “leads to a ratcheting up of all CEO salaries.”
Todd Lippincott, managing director for executive compensation in the Americas for Towers Watson, agrees that CEO skills are not readily transferrable and says this weakens the case for peer group benchmarking. But he argues that Elson and Ferrere overstate the extent to which corporate boards rely on such benchmarking.
“We find that companies use peer groups as one data point—an important data point for sure, but not the only one,” he says. “In many cases they’ll use multiple peer groups—a narrow one of like-sized competitors and a broader one. They also consider the state of the economy, overall employee compensation in the company and finally, performance.”