In this paper, I investigate the relationship between methods of pay, including piece rates and bonuses, and the level and variance of wages using longitudinal data from the NLSY (1988-1990). Results using OLS and fixed-effects show that piece rate workers earn a premium compared to other workers while the positive effect of bonuses measured with OLS disappears when I use fixed-effects. Also, it is shown that the effect of piece rates is negatively related to the level of tenure, which suggests that rates may be cut as workers accumulate seniority. Finally, using minimum distance procedures to estimate a covariance model of wages, I am able to show that most of the variance for piece rate workers can be attributed to unobserved worker productivity, which is not the case for workers having pay for performance schemes.