“Pollution and Labor Supply in a Contest Environment: Evidence from Outdoor Tennis Tournaments” 

Haoming Liu, National University of Singapore

We examine the productivity of professional female athletes during the “China Open,” held outdoors in Beijing. The PM2.5 concentrations range between 1 and 562 µg/m3 in our sample. Professional tournaments provide a window into labor market productivity, given good data availability, i.e., human capital as measured by player rankings, and match outcomes. We document that a player who loses a first set to a similarly ranked opponent is less likely to win the second set as pollution increases. This suggests that the cost of effort increases with PM2.5, or that during the match’s intermediate stages the expected value of winning in the near future is discounted more heavily as the air deteriorates. We construct a three-stage sequential contest model with two players of varying ability. The probability of winning each stage is based on the odds ratio of players’ effort choices during the match. The theoretical model predicts: (i) the second-stage winning probability for the first-stage winner increases with heavier discounting; and (ii) in the asymmetric players’ case, the first-stage winning probability for the lower ranked player increases with heavier discounting. We estimate the structural model via maximum likelihood. We conclude that air pollution does affect such competitive labor supply.