I analyze the effects associated with the introduction of the earnings test on older Czech males' labor supply in 1996. Using data from the Labor Force Survey, I apply a difference-in-differences estimator to measure the effect of the policy change in the Czech pension scheme using a sample of retired males aged 60 to 64 in 1995 and 1996. I use an age group not affected by the earnings test to separate the effect associated with the implementation of the earnings test. Correcting for sample selection, results indicate that the earnings test decreases the labor supply of retired males affected by the policy change. Their participation reduces by 7.7 percentage points, while hours of work decline by 4.8 hours per week. The response to the earnings test is moderately lower after controlling for the announcement of the policy change. The pecuniary value of the labor supply drop was negligible for the state in 1996, but increased significantly between 1996 and 2000.
