The paper analyzes 'subsidy games' between countries in order to attract foreign direct investment (FDI) from a third country. The winner of this game results from the interaction of two factors, relative country size and employment gains from FDI: a large (or 'central') country is more likely to attract FDI, and so is a country with high unemployment. The subsidy equilibrium is compared with two alternative solutions: zero subsidies and first-best subsidies. It is shown that total welfare may be greater under subsidy competition than under zero subsidies: the gains from efficient location implied by subsidy competition may more than outweigh the losses from higher subsidies. Moreover, departing from subsidy competition to zero subsidies or to first-best subsidies (without side payments) implies a gain to one country and a loss to the other. This suggests that it may be difficult to reach a consensus to move away from the status quo of subsidy competition.