When a Door Closes, a Window Opens? Long-Term Labor Market Effects of Involuntary Separations

This study estimates the earning losses of workers experiencing an involuntary job separation. We employ, for the first time in the earning losses literature, a Poisson pseudo-maximum-likelihood estimator with fixed effects that has several advantages with respect to conventional fixed effects models. The Poisson estimator allows considering the full set of involuntary separations, including those with zero labor market earnings because of unemployment. By including individuals with zero earnings and by using our new method, the loss in the year of separation becomes larger than in previous studies. The loss starts with roughly 30% and, although it quickly shrinks, it remains at around 15% in the following years. In addition, we find that compared to other reasons for separation, the earning loss pattern is unique for involuntary separations, because no other type of separation implies such permanent scarring. This latter finding makes us confident that the self-reported involuntariness of a separation is a reliable source of information.

7. Fixed-Effects Negative Binomial Regression Models

This paper demonstrates that the conditional negative binomial model for panel data, proposed by Hausman, Hall, and Griliches (1984), is not a true fixed-effects method. This method—which has been implemented in both Stata and LIMDEP—does not in fact control for all stable covariates. Three alternative methods are explored. A negative multinomial model yields the same estimator as the conditional Poisson estimator and hence does not provide any additional leverage for dealing with over-dispersion. On the other hand, a simulation study yields good results from applying an unconditional negative binomial regression estimator with dummy variables to represent the fixed effects. There is no evidence for any incidental parameters bias in the coefficients, and downward bias in the standard error estimates can be easily and effectively corrected using the deviance statistic. Finally, an approximate conditional method is found to perform at about the same level as the unconditional estimator.