The main idea of this thesis is to deepen our understanding of the relationship between tax policy and heterogeneous capital. The first chapter revisits the question of whether preferential tax regimes are desirable in a context where some jurisdictions have leadership advantages in their choice of tax policy. It is argued that if regions or countries involved in tax competition act sequentially as Stackelberg competitors, they will prefer to limit the use of preferential tax policy. If firms located in small regions face higher mobility costs on average than those located in large regions, small regions want to ban preferential tax regimes while large regions will tend to support them. If jurisdictions are populated mainly by firms with low mobility costs, they will prefer preferential tax treatments. On the other hand, if they are populated mostly by firms with high mobility costs, small regions want to restrict preferential tax policies while large regions will favour them. The second chapter embraces the neoclassical theory of investment to model the rate of investment in physical and intangible capital. It uses data from the EU KLEMS database, the Oxford University Centre for Business Taxation and the Tax Foundation. It concludes that the equations for the rate of investment in physical and intangible capital are distinct. Corporate tax incentives affect the rates of investment in physical and intangible capital, but differently. The higher rate of depreciation of intangible capital relative to physical capital seems to explain the increasing ratio of investment in intangible to physical capital. The third chapter examines heterogeneity by type of capital within the relationship between capital and its user cost, for five types of physical capital asset and two types of intangible capital asset. The dataset is almost similar to that of chapter two. The results show that, in the short-run dynamics, both the dynamic fixed-effects and GMM results seem to agree on the role of changes in the user cost of capital on the accumulation of the stock of capital. Overall, dynamic fixed-effects estimation seems to yield results that are more consistent with the theoretical conclusions on investment behaviour and empirical results for physical capital already established in the literature.

No description available.