A Markov Switching Model of Congressional Partisan Regimes

Studies of development and change in partisan fortunes in the US emphasize epochs of partisan stability, separated by critical events or turning points. A major empirical issue that has plagued the study of American political development is the estimation of the critical moments and durations of these partisan regimes. In this paper we introduce a fresh approach to the study of partisan regimes. Our method is based in the method of Markov switching, introduced by James Hamilton. We apply Hamilton’s approach to the size of party coalitions in the US House of Representatives from 1854 to the present. Our model assumes that the political system is either in a state of domination by one party or it is not (in which case the other party dominates). The Markov switching approach also yields estimated state probabilities that allow us to make inferences about periods of empirical party balance. Roughly speaking, when the Republicans constitute the dominant partisan coalition, they can expect to capture 60 percent of House seats in any given election. The Democrats can expect 59 percent when dominant. Our method also allows the estimation of critical transition points between Republican and Democratic partisan coalitions. The periods we identify as governed by a being Republican coalition are roughly 1860 through 1872, 1894 through 1906, and 1918 through 1928.