Regional policy impacts of reduced payroll taxes
Cuts in pay-roles taxes as a regional support measure is used throughout Europe. In this study we address some of the shortcomings in previous evaluations. A first shortcoming concerns expected outcomes. Most evaluations focus employment impacts and there are from previous research few indications that this type of support measures have any impacts on employment.
From the theory and business logic we claim that support, such as cuts in pay-role taxes, that is given to a firm can be used to whatever the firm finds most suitable. Therefore, evaluating on employment impacts is a rather narrow focus. We extend the analysis to cover impacts on turn-over, employment, wages, investments, long run debts and profits.
A second short coming is the time horizon. Most evaluations evaluate the impacts of pay-role tax cuts with a few years after the support was launched. In our evaluation the support was launched in 2001 and we follow firms until 2009 which gives us the possibility to monitor the impact development over an 8 year period.
Finally, we propose a method to be used when evaluating regional policy. Since firms are nested within e.g. municipalities or industries a traditional econometric approach using dummy variables to capture regional and industry heterogeneity can produce both biased impact estimates and biased standard errors. To deal with this problem we apply a multilevel model which enables us to not only control for regional heterogeneity but also to some extent measure the importance of these contextual factors. Our results indicate that a substantial part of the variation in our outcome variables can in fact be explained by municipality characteristics, e.g. the composition of population and its characteristics, regional growth, etc.
Furthermore, the results from the evaluation follow a clear business logic: In the short run profits and turnover increase. The increased profits are later used to increase wages (including owners’ wages). Later on in the observation period (after 7 years) we find indication on impacts on investments. As in previous evaluations we do not find any significant impacts on employment. Finally, we are able to identify that the profits increase in the beginning of the evaluation period, but that the effect diminishes over time. From a policy perspective we conclude that the support has some short term impacts that are reduced with time and the long term impacts are uncertain.