Incentive Equivalence with Fixed Migration Costs
Journal of Public Economics , 95(11/12), December 2011, pp. 1292-1301
Abstract: If migration between communities is costless, and if policy makers in each community anticipate the migration response to policy changes, then the interests of the two communities are perfectly aligned. Decentralization is efficient. This incentive equivalence is well-known. However, the first-order approach used in virtually all the literature may obscure the fact that no convexity assumptions need be made about technology or tastes. This fact underscores the relevance of incentive equivalence, even when there are scale economies in the public sector. Here, the consequences of positive migration costs for this incentive equivalence are considered. In contrast with much of the literature, migration costs are assumed the same for everyone. This formulation differs from the more common "attachment to home" assumption of heterogeneous migration costs. Conditional on the direction of migration, interests of different communities are still perfectly aligned. But natives of different communities may prefer different directions of migration, weakening incentive equivalence if local policy makers have the power to induce large changes in migration flows. However, incentive equivalence fails only if economies of scale in population are important, so that communities may be "under-populated".
[go to paper]

An Index for Capital Tax Competition
International Tax and Public Finance , 16(6), December 2009, pp. 727-752
Abstract: If very specific assumptions are made about the production technology (output per worker is a quadratic function of the capital/labor ratio), people's preferences (identical within any jurisdiction; linear in private good consumption and in public expenditure per capita), and capital supply (fixed), then equilibrium tax rates can be derived in closed form when jurisdictions choose their source-based tax rates on capital noncooperatively. The focus of this paper is how the size distribution of the population of the different jurisdictions affects equilibrium tax rates. It is shown that the (population-weighted) average tax rate is determined by a (relatively) simple index, which must increase as the population distribution becomes more concentrated. The effects of tax harmonization by some subset of the jurisdictions can then be analyzed. Any tax harmonization by a group of jurisdictions must benefit residents of all jurisdictions not in the group. It must also benefit residents of the largest jurisdiction in the group, and must increase the average payoff of all residents of the group of jurisdictions doing the harmonizing.
[go to paper]

How to Avoid Transferring a Valuable Asset
(with A. Glazer)
Public Choice , 138(1/2), January 2009, pp. 3-8
Abstract: Many mechanisms (such as auctions) efficiently allocate an asset to the firm which values it most highly. But sometimes the asset's owner may benefit from the transfer only if the asset is not too valuable to potential buyers. In this setting, we examine the efficiency of mechanisms when the potential buyers have private information about the asset's value. We show that rent seeking, and lobbying, rather than merely wasting resources, can lead to allocations which are close to efficient
[go to paper]

Tax Competition when Firms Choose their Organizational Form: Should Tax Loopholes for Multinationals be Closed?
(with A. Haufler)
Journal of International Economics , 74(1), January 2008, pp. 188-201
Abstract: We analyze a sequential game between two symmetric countries when firms can invest in a multinational structure that confers tax savings. Governments are able to commit to long-run tax discrimination policies before firms' decisions are made and before statutory capital tax rates are chosen non-cooperatively. Whether a coordinated reduction in the tax preferences granted to mobile firms is beneficial or harmful for the competing countries depends critically on the elasticity with which the firms' organizational structure responds to tax discrimination incentives. A model extension with countries of different size shows that small countries are likely to grant more tax preferences than larger ones, along with having lower effective tax rates.
[go to paper]

Preferential Tax Regimes with Asymmetric Countries
(with A. Haufler)
National Tax Journal , 60(4), December 2007, pp. 789-795
Abstract: Current policy initiatives taken by the EU and the OECD aim at abolishing preferential corporate tax regimes. This note extends Keen's (2001) analysis of symmetric capital tax competition under preferential (or discriminatory) and non-discriminatory tax regimes to allow for countries of different size. Even though size asymmetries imply a redistribution of tax revenue from the larger to the smaller country, a non-discrimination policy is found to have similar effects as in the symmetric model: it lowers the average rate of capital taxation and thus makes tax competition more aggressive in both the large and the small country.
[go to paper]

The Efficiency Consequences of Local Revenue Equalization: Tax Competition and Tax Distortions
(with M. Smart)
Journal of Public Economic Theory , 8(1), January 2006, pp. 119-144
Abstract: This paper shows how a popular system of federal revenue equalization grants can limit tax competition among subnational governments, correct fiscal externalities, and increase government spending. Remarkably, an equalization grant can implement efficient policy choices by regional governments, even in the presence of differences in regional tax capacity, tastes for public spending, and population. If aggregate tax bases are elastic, however, equalization leads to excessive taxation. Efficiency can be achieved by a modified formula that equalizes a fraction of local revenue deficiencies equal to the fraction of taxes that are shifted backward to factor suppliers.
[go to paper]

Public Input Competition
Journal of Public Economics , 89(9/10), September 2005, pp. 1763-1788
Abstract: Public investment in infrastructure may help create agglomeration economies, by attracting mobile factors such as skilled labour. Competition among regions in public investment can then be destructive. This paper analyzes the Nash equilibria to a simple model of public input competition. Even though the regions are assumed identical, the equilibrium may not be symmetric. The problem with non-cooperative behaviour is not (only) that regions invest too much, but that too many regions may choose to invest. Depending on the parameter values, the Nash equilibrium may be efficient, may be inefficient or may not exist (at least not in pure strategies). Better mobility among regions leads to more aggressive competition. The analysis suggests that rents from public investment may be dissipated by governments' competition to attract mobile factors.
[go to paper]

Efficient Migration and Income Tax Competition
Journal of Public Economic Theory , 5(2), April 2003, pp. 249-278
Abstract: This paper examines the consequence of the brain drain for the income tax systems of the source and destination countries for the migration, if the two countries' policies are set noncooperatively by self-interested voters. It is assumed that the brain drain does increase the value of world output: workers with the highest income-earning ability are assumed to be more productive in one country than in another. There are costs to migration of these high-ability workers, costs that are less than the gain in the value of their production. However, for lower-ability workers, the gains in production in moving from the low-productivity country to the high-productivity country are assumed to be less than the migration costs. Voters in the high-productivity country want to capture rents from migrants. These voters are aware of the influence their tax policy has on people's migration decisions. Voters in the low-productivity country also behave strategically. I solve for the Nash equilibrium income tax rates. Increased mobility of highly skilled workers cannot decrease, and may increase, progressivity in the income tax system of the destination country, if migration actually occurs. Finally, the effects of transfers between countries on their income tax systems are examined. Redistribution between countries tends to lead to less redistribution within countries. If transfers between countries are set by a vote of all residents of both countries, then the transfer chosen will be the one that leads to the least progressive tax possible being chosen in each country.
[go to paper]

The Optimal Majority with an Endogenous Status Quo
Social Choice and Welfare , 21(1), August 2003, pp. 131-148
Abstract: Using the voting procedure proposed by Baron (1996), the consequences are examined of changing the majority required to change legislation. When the majority required is greater than fifty percent, and when voters behave strategically, the first policy proposed (on a stationary equilibrium path) is never defeated subsequently. Which policy gets proposed first depends on which voter gets to make the first proposal. But increasing the required majority induces a mean-preserving spread on the distribution of these policies, if voters' types are distributed symmetrically. Thus before the voting procedure begins, voters would prefer unanimously to see the required majority reduced.

Efficient Migration and Redistribution
Journal of Public Economics , 87(11), October 2003, pp. 2459-2474
Abstract: It has often been noted that the brain drain may be abetted by highly progressive tax systems in the regions which are likely sources for emigration by those with the most income-earning potential. This paper presents the case in favour of such policies. The underlying assumption here is that there are productivity differences between regions, and that emigration of the most skilled workers from less productive regions increases the overall value of national output. The problem lies in sharing these gains with the less-skilled workers left behind in the low-productivity region. If national governments can transfer income among regions, but if the transfers cannot be targeted to particular individuals, then there is a trade-off between equity and efficiency. Generous transfers redistribute the gains, but tend to impede efficient migration, since prospective migrants will only emigrate if their earnings, net of all fiscal transfers, and net of migration costs, are higher in the destination region. Greater tax progressivity in the source region helps relax a constraint, and make the trade-off less difficult.