Do Employment Subsidies Work? Evidence from Regionally Targeted Subsidies in Turkey

Gordon Betcherman, World Bank Group
Meltem Daysal, University of Maryland
Carmen Pages, Inter-American Development Bank

This paper studies the effects on registered employment and establishments of two employment subsidy schemes offered by the Government of Turkey to encourage investment in low-income provinces. Using a "difference-in-differences" methodology, we estimate that the registered employment gains for the covered provinces range from 4%-13% for the first program and from 9%-15% for the second program. However, both schemes had substantial deadweight losses. We find that 50% to 80% of the jobs subsidized under the first program would have already been created. Because of better design features, the second subsidy program had lower deadweight losses (25%-50%), and thus appears to have been more cost-effective even though it provided more generous subsidies. Although data limitations constrain our capacity to test whether the dominant effect of the subsidies was to increase formalization of firms and employment rather than boosting economic activity, the evidence we have suggests the former was more important.