Peer Group Benchmarking for CEOs Fundamentally Flawed According to New University of Delaware Study
September 28, 2012
Companies should move away from peer group benchmarking in favor of relying on internal pay standards and more discretion by the board in setting CEO pay, according to a new University of Delaware study funded by the Investor Responsibility Research Center Institute. The study, co-authored by Professor Charles Elson and corporate governance fellow Craig Ferrere, raises questions about the applicability of peer group benchmarking to determining CEO pay (as opposed to NEO or other executive pay) and blames overuse of peer groups, even properly selected peer groups, for how executive pay has “spiraled” over the years. However, the premise of the study, that boards largely singularly rely on peer group data in setting CEO compensation, may not reflect how compensation committees and boards actually set pay.
The paper argues that there is not a true labor market for CEOs, citing a 2011 study of S&P 1500 firms that showed from 1993 to 2005, less than two percent of approximately 1800 new CEOs had been CEO of a public company before taking on the new role. Noting that “particularly for the large firms comprising the S&P 500, CEOs are rarely traded in any market for their talents,” the study concludes that peer groups, which are applicable only where there is an efficient market determining CEO compensation, should therefore not be over-relied upon to the detriment of the board’s discretion.
Emphasizing that board members are elected by shareholders for their “good and objective judgment,” the authors recommend that “in an invigorated process, benchmarks should be seen as merely a single data point amongst many other necessary factors which must be considered by directors.” While admitting that their suggestions are “not concrete or easily implemented,” the authors nevertheless recommend that boards “develop internally created standards of pay based on the individual nature of the organization concerned, its particular competitive environment and its internal dynamics” and that they not allow peer group data to unduly influence their decisions. The study provides helpful food for thought, but for many companies, peer group data is the starting point upon which other filters are applied after evaluation of the skills, experience and expected achievement of individual CEOs.