Job creation through public start-up assistance?

Abstract

This study examines the impact of German public start-up assistance programmes administered by the Deutsche Ausgleichsbank (DtA) on the performance of young firms. The empirical analysis is based on firms from the ZEW Entrepreneurship Study that either received start-up loans in one or more DtA schemes or did not receive any funding from the DtA at all. The paper applies a non-parametric matching approach often applied by labour market economists. The interesting success measure is the average annual employment growth rate over a six year period and the resulting causal effect is the difference of this measure between the group of subsidized firms and the selected control group firms that did not receive any DtA start-up loans. The empirical analysis shows that DtA start-up loans significantly improve the average employment growth rate.

Acknowledgements

Financial support from the Deutsche Ausgleichsbank (DtA) and the German National Research Council (DFG) grant LE 1147/1-1 is gratefully acknowledged. I thank Susanne Prantl, Georg Licht, Konrad Stahl, Daniel Skambracks and Jochen Struck for helpful comments. All errors that remain are, of course, the responsibility of the author alone.

Notes

The questionnaire is part of a project co-financed by the German National Science Foundation (DFG) under the grant LE1147/1-1.

CREDITREFORM is the largest German credit rating agency.

Several firms refused to answer all questions but at least gave information as to whether they had exited the market or not. For analyses dealing with the survival of firms, a statement regarding the survival status is possible for an additional 2234 firms.

The date of birth is not available if firms apply for public assistance. In these cases the maximum score of conformity only can reach 75%. Further search variables (legal form, economic sector classification) were not available in the DtA database at the time the merging took place.

Unfortunately, there is no statistical test to check the validity (see Almus et al., 1999).

See also the literature cited in this paper.

Here, Epanechnikov kernel density estimates instead of histograms serve as tool to show the similarity in the relative frequencies (probability density) since both groups contain the same number of observations after the matching process (c.f. Silverman, 1986).

These numbers are calculated by simply summing up the number of employees at start-up in each exiting firm.