Although considerable research has addressed the relationship between performance ratings and pay increases, this literature has been criticized for 2 reasons. First, most researchers have ignored the fact that performance ratings may have not only a direct effect on pay but also an indirect effect through promotions and market adjustments. Second, almost all of the research has been cross-sectional. We studied the relationships between performance ratings and salary raises by using longitudinal data and found that performance ratings correlated more with total pay increases than with merit pay increases. Compared with the average cross-sectional correlation, the use of aggregated data (over time) resulted in a larger performance rating–merit pay increase relationship.