The CEO pay‐performance relationship: pooled vs. industry models

Abstract

Summarizes previous research on the links between chief executive officer (CEO) compensation, firm performance and industry; and compares pay‐performance relationships calculated by pooling data with those based on industry segmentation. Develops a model incorporating six factors which may affect CEO cash compensation (tenure, net company income, income variance, net sales, returns to shareholders and beta) and uses 1986‐1992 data from a sample of large US firms covering eight industries to test it. Shows, using Andrew’s Sine Technique regression, that there is a wide variation between individual industries which is obscured when data is pooled. Discusses the methodology used, consistency with other research, the limitations of the study and the underlying reasons for the findings.