This paper analyzes the location of manufacturing activities when regional policy is determined by each region's relative propensity to vote. Once voting over government transfers to regions is included in an economic geography framework with size asymmetries, the standard prediction that the larger region becomes the core when trade barriers are reduced no longer holds. The establishment of manufacturing production in the economically smaller region is increasing in the level of regional integration. As trade is increasingly liberalized, the economy eventually features a reversed core-periphery equilibrium where all firms reside in the South. It is further shown that the relative political participation rate increases in the factor scarce region as trade is liberalized. Empirical evidence shows that the model is consistent with qualitative features of the data.

No abstract

K E Y W O R D S capital controls, economic geography, welfare 1 See Grossman (2016) for an overview of the purposes of trade agreements. 2 See, e.g., Schuknecht (1998) for a trade policy perspective on capital controls.

Current research has found ambiguous theoretical and empirical results with respect to the effects of the type of electoral regime on trade policy. The present paper aims to reconcile the different views within a theoretical model. It is shown that the equilibrium level of trade protection can be relatively higher, as well as lower, under a majoritarian electoral rule compared with proportional representation. Trade policy is more (less) protectionist under proportional electoral regimes, as compared with majoritarian institutions, if swing districts are populated by relatively more (less) factor owners with stakes in the exporting sector. It is also shown that politicians optimally apply a lower (higher) level of rent seeking under the majoritarian electoral rule if there are relatively more factor owners in the swing districts with stakes in the exporting (import-competing) sector.

This paper studies location decisions of firms in an economic geography model with endogenous regional policy. Policy is determined by probabilistic voting under proportional and majoritarian elections. Different electoral competition give rise to different location incentives. Under plausible assumptions, the smaller region has a higher fraction of ideologically independent swing voters than the larger region. Majoritarian voting, by focusing electoral competition into swing districts containing the most policy-responsive voters, therefore allocates more subsidies to firms in the smaller region. Compared to proportional voting, this leads to more firms in the region with fewer consumers. Proportional voting thus welfare-dominates majoritarian election.

No abstract