Spatially Targeted Government Spending and Heterogeneous Constituent Cost Shares

Document Type


Publication Date

Spring 2009


The law of 1/n (Weingast, Shepsle, and Johnsen, Journal of Political Economy, 1981) posits a positive relationship between the size of an elected body and government spending because the taxpayers in each district bear only 1/nth of the total tax burden. Relying on variation in the number of seats in elective bodies, evidence supportive of the law of 1/n has been found at all levels of representative government. It is possible, however, that these findings suffer from endogeneity bias: polities preferring larger government spending may also prefer to have more elected officials. In this paper, we propose an alternative test of the law of 1/n. In addition to postulating that larger elective bodies will have higher levels of government spending, the law of 1/n also implies that, ceteris paribus, there will be a negative relationship between locally targeted government spending and a jurisdiction's share of the taxation required to finance the spending. We test this prediction by examining the relationship between spatially targeted government spending and the tax burden across states. We find a negative relationship between local tax contributions to the common taxbase and locally targeted government spending for aggregate locally targeted spending and for six of eight subcategories. These findings are robust to the inclusion of other variables thought to influence the distribution of spending on parochial projects.