Should Local Public Employment Services be Merged with the Local Social Benefit Administrations?

mayor that is a board member or a federal state president of the German County Association
are not associated with better labour market conditions.

Table 3: Relationship between Approved Local Providers and the job finding rate

Dependent variable:                            Log long-term unemployed finding a job each month

Standard errors are given in parentheses. Coefficients with * and ** indicate significance at a 5 % and 1 %
level, respectively. Observations are clustered at the level of the 16 German states and weighted accordingly.

The IV estimate and the OLS estimate of the overall effect of being an Approved Local
Provider are shown in Table 3. The coefficients of all estimates are shown in Table A2 and
Table A3 in the Appendix. The IV estimate for the overall effect of being an Approved
Local Provider is not significantly different from zero (it is statistically significant at a 10%
level). But the IV estimate in Table 3 for the overall effect of being an Approved Local
Provider indicates a larger negative correlation between being an Approved Local Providers
and the job finding rate of long-term unemployed than the OLS estimates. Although the
Durbin-Wu Hausman Test reported in Table 3 suggests that the OLS estimate and the IV
estimate are not significantly different from each other, the fact that the IV estimate shows a
larger negative correlation than the OLS estimate is consistent with our expectation that the
OLS estimate is biased upward because of positive self-selection of regions into Approved
Local Providers.

6. The role of organisational characteristics

The OLS-estimates and the IV-estimates for the overall effect of being an Approved
Local Provider are insignificant, but they indicate that despite the evidence of positive self-
selection provided in section 3, regions with Approved Local Providers are not better than
Joint Local Agencies in integrating long-term unemployed into the labour market. This
raises the question why these regions self-selected into being solely responsible for the long-

13

<<   <   12   13   14   15   16   17   18   >   >>