Congressional distributive politics and state economic performance

States that were represented by very senior Democratic congressmen grew more quickly during the 1953–1990 period than states that were represented by more junior congressional delegations. States with a large fraction of politically competitive House districts also grew faster than average. The first finding is consistent with traditional legislator-based models of distributive politics, the second with partisan models. We cannot detect any substantively important association between seniority, state political competition, and the geographic distribution of federal funds, so higher district-specific federal spending does not appear to be the source of the link between state economic growth and congressional representation. When Senator Henry “Scoop” Jackson, the ranking Democrat on the Senate Armed Services Committee, died unexpectedly in September 1983, Roberts (1990) reports, the stock market values of defense contractors based in his home state of Washington declined. The share prices of contractors based in Georgia, the home state of the next-most-senior Democratic Senator on the committee, Sam Nunn, increased. When Senator George Mitchell of Maine announced his plan to retire from the Senate at the end of his current term, the New York Times reported that “the most agonizing part of his decision . . . was recognizing that his position enabled him to help his home state in ways that a freshman taking his place could not” (March 6, 1994, p. 11). A vast empirical literature has examined the link between Congressional representation and the distribution of government-controlled economic benefits. Anecdotal evidence suggests that districts represented by some senior congressmen have received disproportionate shares of some types of federal spending. Pearson and Anderson (1968) provide a particularly compelling account of former congressman Mendel Rivers (“Rivers Delivers”), chair of the House Armed Services Committee, and his efforts to channel military ∗ We are grateful to Julie Berry Cullen for outstanding research assistance, to Steve Ansolabehere, Mark Crain, Alan Gerber, John Lott, Jeff Milyo, Sam Peltzman, Kim Rueben, Andrei Shleifer, James Snyder, Charles Stewart, seminar participants at Chicago, Harvard, the NBER, and Stanford, three anonymous referees, and especially to Keith Krehbiel and Barry Weingast for helpful discussions, and to the Center for Advanced Study in Behavioral Sciences and National Science Foundation for research support.