Despite the widespread use of incentive pay, there is limited evidence about what factors influence its organization-wide, broad-based application. This study uses data from three sources and multiple levels, including a unique data set of the total compensation of individual employees in 104 firms over a four-year period (1997)(1998)(1999)(2000), and theoretically and empirically examines the use of bonuses and stock options in organization-wide applications. We examine the efficacy of three main rational theories, principal-agency, positivist agency and contingency theories, which are based on the premise that incentives are related to performance. At the individual level we identify two determinants: type of job and level within the hierarchy and four determinants at the organizational level: performance, risk, size, and strategy. Our results indicate that the factors derived from the three theories provide a limited explanation for the variation in the use of broad-based incentives within and across organizations. Abowd's (1990) longitudinal studies of managerial compensation in Fortune 500 organizations suggested that companies paying a higher proportion of variable pay to base salary, also had higher subsequent financial performance, however, these results varied depending how performance was defined (Abowd, 1990) and whether the company was historically profitable (Gerhart & Milkovich, 1990). Other researchers have found that variable pay's effectiveness depends on the risk of the organization's earnings stream (Garen, 1994;Bloom & Milkovich, 1996;Aggarwal & Samwick, 1999), level of monitoring (Beatty & Zajac, 1996; Gray & Cannella, 1995) or fit with business level strategy (Boyd & Salamin, 2001;Montemayor, 1996;Gomez-Mejia & Balkin, 1992).Both researchers and practitioners have interests in understanding the application and efficacy of using broad-based incentives. This goal, however, is challenging for two reasons.The first is that different theories often offer somewhat different interpretations about how the incentive compensation-organization performance relationship works. Even rationally based theories whose premise is that better compensation practices result in better organizational
Organization-wide Broad-based Incentives: CAHRS WP02-05Page 5 performance differs. For example, agency theory arguably the dominant rational theory with respect to this topic, has developed along two lines: positivist and principal-agent (Eisenhardt, 1988;Beatty & Zajac, 1994). While the two streams share common assumptions about people, organizations and information, they differ in their focus and in their dependent variable (Eisenhardt, 1988). Principal agent-theory focuses on the incentive effects of variable pay and the reduction of agency costs as the path to improved shareholder performance. Positivist agency theory suggests that monitoring rather than incentives may be more cost effective, particularly when the use of incentives shifts too much risk to the individual or when outcomes are not easily...