Abstract

While a good deal is known about economic voting in first-order elections, very little is known about it in second-order elections. Classical economic voting theory becomes complicated when applied to second-order elections, because there are two incumbents (national and local) and two economies (national and local). Under these conditions, attribution of responsibility, and therefore, the target of economic sanction, appears unclear. We argue that, at least in unitary states, economic responsibility for local conditions is attributed to the national (rather than the local) incumbent. The implication is that, when the local incumbent and the national incumbent are of the same party, the local incumbent loses votes under bad economic conditions. In contrast, when the local incumbent and the national incumbent are of different parties, the local incumbent gains votes under bad economic conditions. These propositions are tested against a large regional election data-set from France. The main finding, that voters do attribute local economics to the national government, and vote accordingly, would seem generalizable to regional elections in similar systems, such as Italy.