Is There an Alternative to Peer Group Analysis?
January 22, 2013
Q: Can boards develop an alternative to peer group analysis when setting compensation? -Independent director
A: Long a tool for comp committees in establishing executive pay benchmarks, peer groups have recently been under examination by directors, shareholders, advisory firms, academics and management. All are questioning whether peer group reviews actually create value for companies and shareholders.
To be sure, peer group analysis has flaws, and recognizing them is the first step toward improving the process or putting a new one in place. In most cases, the board is best qualified to customize its company’s peer group. But before we diagnose a solution, we need to assess the flaws at hand.
Breakdowns, for example, can occur when peer group selection processes consider only companies that compete for revenue but not for talent. When a board seeks to determine market-level pay for an executive whose industry background and skill sets have limited transferability, the revenue-oriented approach may work well. However, executives in finance, human resources and law generally have skills that translate across industries, giving them a much larger potential “peer” group.
In other cases, the statistics developed from peer groups do not recognize the tenure, experience, actual job responsibilities, or results that should affect executive pay. We recommend selecting a large number of companies to even out anomalies — to be specific, 12 to 18 businesses.
A bigger challenge occurs when an industry has only a limited number of companies. One complaint about proxy advisors is that they require peer groups to be of a specific size. That forces them to include companies that share neither potential customers nor employees.
Yet it wouldn’t be practical to eliminate peer groups, for all their flaws. Doing so would leave boards and compensation professionals with no market-based starting point for setting executive pay. Additionally, shareholders and others who assess pay-for-performance alignment expect to see in the annual proxy’s Compensation Discussion and Analysis section a list of companies used to determine pay, to help give validity to the process.
So, rather than scrapping the concept of peer groups entirely, boards might consider realistic alternatives for using the data supplied by peer companies. For example, rather than focusing only on the summary statistics, take in the big picture — all of the data.
This would include not only each peer executive’s tenure but an understanding of his or her position. While the company may be an appropriate peer, directors should ask whether the executive is an appropriate counterpart. (Not all CFOs perform the same functions.) Boards might also consider whether to include compensation outliers in the peer group, whether they are high payers or low payers.
For a different approach, directors should consider eliminating pay data for any peer company executive who receives less than half or more than twice the compensation of the executive under review. Another alternative would be to analyze (and disclose separately) compensation data for executives outside the peer group, at companies that are a good industry match but are either too large or too small to meet traditional peer-group criteria.
It is easy to dismiss these suggestions as too much work. However, given the limited number of executive positions that can be used in a proxy analysis, it is clear that for roles other than the CEO and CFO, peer group analysis has limited value. In those cases, boards should rely on supplementary information from published data.
While a wholesale alternative to peer group analysis may be unavailable, boards have many options for how to use other companies’ data, and they can clearly move away from using simple summary statistics.