A Model of Worker Deaths and Firm Adjustment


Here, I (as a component of a paper by Simon J¨ager and J¨org Heining) develop and calibrate a dynamic model of wage-setting based on the model in Kline et al. (2019). Using the model, I estimate firms’ costs of replacing a worker from empirical reactions to worker deaths in German Social Security data. Estimated replacement costs are quite large, on the order of two years of worker salaries. I show analytically that the rise in wages in response to a death is evidence for convex adjustment costs. However, I estimate the convexity of the hiring cost function and find an exponent of 1.09, far from the typical quadratic functional form. I also calibrate the model separately for thick and thin labor markets, finding that replacement costs are almost three times larger in thin markets. I then generalize the model to have two types of workers and estimate the elasticity of substitution between workers of different occupations, finding an elasticity of 6.5. I also estimate that only 30% of the observed earnings response is due to the intensive margin of hours, while the remaining 70% is due to increased hourly wages. Together, these findings imply a substantial degree of imperfect competition in the labor market, and provide evidence for the existence of rents from employment relationships due to costly replacement.

Masters Thesis