When one size does not fit all: Using ex post subjective ratings to provide parity in risk-adjusted compensation

Firms typically use a 'one-size-fits-all' (OSFA) compensation contract that specifies a common formulaic relation between performance and compensation (i.e., a performance bonus) for non-executive managers in similar jobs. However, a contract that is appropriate on average, may be suboptimal for individual managers if heterogeneity in the operating environment creates varying compensation risk. We use field data from a retail firm that introduced an OSFA bonus compensation plan for its store managers. The common bonus formula is based on a weighted sum of objective measures of performance and a subjective rating made by supervisors. The firm intended the supervisors' discretionary subjective rating to evaluate performance on dimensions that are difficult to measure (e.g., store appearance). We test and find that supervisors give uniformly higher subjective ratings to managers whose objective measure of sales performance is measured with greater noise, and to managers who face higher performance target difficulty, the latter assessed both prior to (ex ante) and subsequent to (ex post) the evaluation period. These results obtain after controlling for manager ability and performance, and for alternative mechanisms to mitigate differences in compensation risk (e.g., salary changes, sales target changes, and bonus adjustments). The evidence suggests that supervisors use discretion in subjective ratings to provide manager-specific risk premiums for non-executive managers who are subject to an OSFA contract.