New Approaches to Shirking in Major League Baseball

The impact of contract length on player performance has received attention in the sports economics literature. Players are said to shirk following the signing of a guaranteed contract if performance falls below some level of expected performance. While much of the literature uses OLS to identify shirking behavior by players under multi-year contracts, this study employs OLS, fixed effects regression, and two-stage least squares regression to analyze player performance from 2009 to 2017. Using fixed effects specifications, evidence of shirking is found with Wins Above Replacement and On-Base Plus Slugging Percentage - Plus as outcome variables. Using an improved measure of marginal revenue product, evidence of shirking is found using OLS specifications, where shirking is measured as the difference between expected and realized marginal revenue product. Two-stage least squares is then employed to address potential endogeneity in the marginal revenue product specifications. Using 2SLS, economically larger shirking effects are found. The negative impact of contract length on performance persists controlling for additional potential omitted variables