Selective firing and lemons

This paper uses the Austrian Social Security Register (ASSD) to explore what information firms infer from the three common types of displacement: individual layoffs, individuals displaced due to a closure and individuals displaced due to a mass layoff. I bring together two strands of the literature, namely signaling and sorting and contribute to it in three ways. First I test whether the individual layoffs are the least productive, second I investigate whether individual layoffs are perceived as lemons (with a specific focus on the high ability individuals) and third I raise the question whether the lemon exists in the resulting matching pattern. Using the Abowd et al. (1999) model I show that the individual layoffs are the least productive measured by the person fixed effect. I confirm the signaling argument of Gibbons and Katz (1991) that individual layoffs are perceived as lemons also for high ability individuals, but I reject the argument of Gibbons and Katz (1991) against the matching model (Becker, 1973). Using three different measures of sorting, I find that the matching changes differentially for the different layoff groups. This leads to the tentative conclusion that both sorting and signaling take place after an individual job loss.