Redistribution and the Alignment Effect: Theory and Evidence

Abstract: This paper studies the impact of political alignment on the redistribution of grants between central and local governments. A model, where the central government uses its discretionary power to further the prospects of its constituent parties at all levels of government, is constructed and used to make a prediction regarding the allocation of grants. Furthermore, the prediction is tested using a regression discontinuity design on a large data set from Swedish municipalities. In addition, to account for the formation of party coalitions at both the central and local level, the model features a novel conception of political alignment where alignment exist on a scale as opposed to the ordinary, binary view. These coalitions also force a novel construction of the running variable in the regression discontinuity design based on municipal mandates. These mandates are weighted depending on whether they are inside or outside the local ruling coalition. Among other things, this novel perception of the running variable of political alignment provides a connection between the core supporter hypothesis that has been widely discussed in the literature within the politics of redistribution field. Since the weight is exogenous to the empirical model, the empirical results depends upon the choice of weight. Ultimately, it is found that aligned municipalities receive more grants than unaligned conditioned on the choice of weight being approximately valid