The effect of agglomeration size on local taxes

Standard tax competition models predict a 'race-to-the-bottom' of corporate tax rates when firms are mobile. Recent theoretical literature has qualified this view by offering a theoretical explanation why this extreme prediction need not occur: central regions with large clusters of economic activity are able to set positive tax rates without fearing to lose firms to peripheral regions as the firms would forego 'rents' from agglomeration economies. In this paper, we study whether local policy makers effectively tax such agglomeration rents. We test this with panel data from Swiss municipalities between 1985 and 2005. We find that large urban areas set indeed higher tax rates than small ones. This is consistent with the theoretical prediction. Within urban areas, however, municipal tax rates are unrelated to the size of economic activity in and around municipalities while they are positively related to the size of the political jurisdiction. We see this result as evidence that the standard tax competition model for asymmetric jurisdictions is at work in the competition of municipalities within an urban area. Both results are robust to controlling for reverse causality by using instrumental variables. Controlling for fixed effects in a 20 year panel is non-informative and neither supports nor contradicts these findings. As a robustness check we introduce an new measure of cluster intensity which considers the varying intensities in agglomeration economies across sectors.