Relative Performance Evaluation in an Oligopoly

33 Pages Posted: 16 Jan 2012

See all articles by Martin G. H. Wu

Martin G. H. Wu

University of Illinois at Urbana-Champaign

Date Written: January 4, 2012


The rationale of RPE use in executive pay is to filter out common risk, not firm idiosyncratic risk. As common risk is often interpreted casually as non-diversifiable and firms’ idiosyncratic risks as totally independent of one another, we show that these casual interpretations contain three basic flaws and therefore are directly responsible for the RPE puzzle. First, common (non-diversifiable) risk needs filtering, but executives can personally consider it, implying that RPE use in executive pay to filter it out is unnecessary. Second, a firm’s idiosyncratic risk is diversifiable for investors but not for the firm’s executives; therefore it too needs to be reduced through filtering. Third, this latter risk reduction would be impossible, however, if all firms’ idiosyncratic risks were totally independent of each other. This view of total independence among firms’ idiosyncratic risks may be harmless from one firm and duopoly’s standpoint, but it is clearly unrealistic and no longer innocuous, because two firms can share unique shocks not shared by another, or all other firms, in an oligopoly. These unique shocks represent correlated firm idiosyncratic risks which may also be filtered out through RPE-related peer-group selections; yet they do not fit into the casual interpretation of common risk.

Keywords: executive pay, relative performance evaluation, related peer groups

JEL Classification: D8, G3, J33

Suggested Citation

Wu, Martin G. H., Relative Performance Evaluation in an Oligopoly (January 4, 2012). CAAA Annual Conference 2012. Available at SSRN: or

Martin G. H. Wu (Contact Author)

University of Illinois at Urbana-Champaign ( email )

1206 South Sixth Street
Champaign, IL 61820
United States
(217) 333-5957 (Phone)
(217) 244-0902 (Fax)

Register to save articles to
your library


Paper statistics

Abstract Views
PlumX Metrics