Falling off the Ladder - Earnings Losses from Job Loss

Job loss translates into large and persistent earnings reductions. Using a large administrative dataset from Germany, I document that this reflects both a reduced employment rate and reduced wages, with the recovery in wages being particularly slow. To account for these facts, I build a search model of the labor market in which workers climb an earnings ladder along multiple dimensions. They search for increasingly productive and secure jobs, accumulate general human capital, and extract rents through renegotiation. The estimated model generates earnings and wage reductions from job loss as large and persistent as in the data. 28% of the wage losses from job loss reflect the loss of productive employers, whereas skill loss and the loss of negotiation rents contribute 52% and 20%, respectively. Finally, I show that the setup generates a novel rationale for unemployment benefits since workers overvalue job security. The size of the wedge is directly depends on the earnings losses from job loss, thus highlighting the importance of generating empirically plausible earnings losses in a model of unemployment.