JBNST - Vol. 233/4 - 2013

Using the introduction of fixed unemployment assistance in Germany in 2005 as a unique natural experiment, we find strong evidence that decreased unemployment compensation has an adverse effect on wages. We use micro panel data to identify and estimate the effect of this structural break. In eastern and western Germany, the relative effect is higher for women. In western Germany, the relative effect increases with skill level. In eastern Germany, there is no clear skill-specific pattern.

This paper investigates the role of early life adversity and home resources in terms of competence formation and school achievement based on data from an epidemiological cohort study following 364 children from birth to adolescence. Results indicate that organic and psychosocial risks present in early life as well as the socio-emotional home environment are significant predictors for the formation of competencies. Competencies acquired at preschool age predict achievement at school age. A counterfactual analysis is performed to assess trade-offs in the timing of interventions in the early life cycle.

This paper analyses the impact of cluster structures on employment development in Germany according to the hypothesis of Porter (1998). It develops a new way of measuring the co-location of suppliers and buyers of intermediate goods in a region based on an input-output approach. The resulting indicator is implemented in a shift-share regression in order to analyse the importance of input-output linkages for the employment development in individual industries. One advantage of this approach is that the results can be compared to earlier studies on localization advantages according to Marshall (1890). The results show that the availability of suppliers and customers in the same region was a major engine for job creation in specific industries in the past. In the period 1998 to 2007 this was particularly observed for service sectors, such as Air Transport or Health and SocialWork but also for some manufacturing industries as well as for Agriculture and Construction. It becomes apparent from the comparison with earlier findings that agglomeration advantages are not realizable within a single industry. Positive effects rather result from the right composition of different industries that have the possibility to establish common production chains. However, for a lot of industries the intensity of inter-industrial interdependence did not play a significant role for their employment development.

Explanations for the robust situation of the German labour market in the course of the global recession 2008/09 are the usage of short-time work schemes, the higher internal labour flexibility due to provisions of collective agreements and the reduction of positive balances on working-time accounts. However, another aspect has not been paid much attention to so far. Multinational companies could have reduced employment as a reaction to the global recession not in their home country, but mostly in their foreign affiliates. The possibility ofmultinational companies to choose in which countries they change employment as a reaction on a changed economic situation would have enhanced German economic policy measures. Using the microdatabase direct investment provided by the Deutsche Bundesbank, we analyze the employment changes within German multinational companies during the global recession 2008/9.We find evidence that employment adjustments took place mainly in the foreign affiliates stabilizing employment in Germany.

This study presents a first comparative analysis of Lasso-type (Lasso, adaptive Lasso, elastic net) and heuristic subset selection methods. Although the Lasso has shown success in many situations, it has some limitations. In particular, inconsistent results are obtained for pairwise highly correlated predictors. An alternative to the Lasso is constituted by model selection based on information criteria (IC), which remain consistent in the situation mentioned. However, these criteria are hard to optimize due to a discrete search space. To overcome this problem, an optimization heuristic (Genetic Algorithm) is applied. To this end, results of aMonte-Carlo simulation study together with an application to an actual empirical problem are reported to illustrate the performance of the methods.