50th Congress of the European Regional Science Association: "Sustainable Regional Growth and Development in the Creative Knowledge Economy", 19-23 August 2010, Jönköping, Sweden
The paper asks whether subsidies aiming to redistribute economic activity across regions can be justified with the welfare argument. Moreover, different tax systems are compared with respect to the size of the subsidy needed for achieving a certain spatial distribution of economic activity, and achievable welfare level. The constructed capital model with two regions is chosen as the underlying model, with the regions interpreted as parts of one country. Until now only a little attention has been paid to policy and welfare issues in the NEG literature. It is known that the results of a political intervention meant for increasing welfare could instead decrease it. The paper shows that with an appropriate tax system the adverse effects of subsidies can at least be alleviated. A proportional subsidy to capital increases the overall capital stock, enabling to draw utility from a wider variety of goods. However, the regional effects are different: compared to the no-subsidy case, the subsidized region can enjoy a larger share of locally produced and thus, cheaper goods, while the non-subsidized region has to import more. Moreover, when the subsidy is financed by a uniform income tax, all economic agents bear the tax burden and might face a welfare decrease. If the subsidy is financed by collecting taxes only from the subsidised region, the residents in the non-subsidised region are not hurt by the income loss, but they still lose due to higher price index. The residents of the subsidised region have to give up more of their income for financing the subsidy. Thus, in both cases the net effects are unclear, depending on the values of parameters and the level of trade costs. The Rawlsian and utilitarian welfare function, and the compensated Pareto criterion are applied for welfare analysis. According to the first two concepts the policy can be welfare increasing. The analysis of a potential Pareto improvement through a compensation shows that mostly it is not possible to set a preference order over the policy and no-policy case. However, over some range of trade costs a welfare improvement is possible.