Not available in German. We investigate expectations concerning future job loss in former German Democratic Republic (East Germany) shortly before the economic, monetary and social union in July 1990. In order to model these expectations, we take detailed account of individual heterogenieity, the availability and interpretation of information, and the economic and social environment of the individual. Our data base is the Socio-Economic Panel (SOEP) East. We find that, with some exceptions, East Germany hold expectations comparable to those held by individuals having experienced a market economy, which is surprising given the lack of such an economy in the previous German Democratic Republic. Since these expectations are only observed, an adequate estimation method is the ordinal logit model. The corresponding stochastic assumptions are testest extensively using pseudo-Langrange multiplier tests against omittd variables, non-linearity, asymmetry of distribution and heteroscedasticity. Furthermore, we apply Hausman tests to check the validity of the classification of the endogenous variable.
Not available in German. We investigate the plans of individual workers concerning future self-employment in the former German Democratic Republic shortly before the economic, monetary and social union in June/July 1990. Our data base is the Socio-Economic Panel East. We find that the desire to become an entrepreneur is basically determined by individual and household characteristics, including income and asset indicators, and not as much by the current job situation of the individual. The work experience attained in the socialist economy seems to be irrelevant for the decision to become self-employed in a market economy. Furthermore, we find evidence of barriers to entry which may come from capital market constraints and institutional restrictions.We also present an estimation of the determinants of the probability of being self-employed in Summer 1990. Due to institutional restrictions we find only a few characteristics to be important. For estimation, we use the binary and the ordinal logit model. The corresponding stochastic assumptions are tested extensively using pseudo-Lagrange multiplier tests against omitted variables, non-linearity, asymmetry of distribution, and heteroscedasticity.
In an evaluation of a job-training program, the influence of the program on the individual earnings capacity is important, because it reflects the program effect on human capital. Estimating these effects is complicated because earnings are observed for employed individuals only, and employment is itself an outcome of the program. Point identification of these effects can only be achieved by usually implausible assumptions. Therefore, weaker and more credible assumptions are suggested that bound various average and quantile effects. For these bounds, consistent, nonparametric estimators are proposed. In a reevaluation of Germany's training programs of 1993 and 1994, we find that the programs considerably improve the long-run earnings capacity of its participants.
Available Downloads: Discussion paper, appendix, R-code and data
This paper discusses the estimation of binary choice panel data models. We begin with different versions of the static random effects model when the explanatory variables are strictly exogenous. Depending on the autocorrelation structure of the errors, different
estimators are available and we detail their attractiveness in each situation by trading-off their efficiency and robustness with respect to misspecification. Then, we consider the static model when a time invariant unobservable variable is correlated with the time varying
explanatory variables. The non-linearity of binary choice models makes it pretty hard to eliminate individual fixed effects in likelihood functions and moment conditions, because the usual differencing out that works for the linear model does not work here except in special
cases. Imposing quite restrictive assumptions is the price to pay to estimate consistently parameters of dynamics for fixed and random effects, in other words cases when the explanatory variables include lagged endogenous variables or are weakly exogenous only.
(doi: 10.1007/978-3-540-75892-1)
