This work explores the influence of gubernatorial coattails in state legislative elections. Through a district-level analysis conducted in nine states, I measure how party support for the governor affects the percentage of the vote received by candidates running for the legislature. Findings indicate that gubernatorial coattails do influence candidate vote margins, even when factors such as campaign spending, past party performance, and other district-level conditions are controlled. However, the magnitude of this effect is mitigated by candidacy status and the closeness of the gubernatorial election. Specifically, coattail effects are dampened by the presence of an incumbent, while their influence is enhanced in states with competitive gubernatorial elections.
This analysis examines gubernatorial coattail effects in state legislative elections. The major question is: How much of an influence does support for the governor have on the votes received by candidates running for the state legislature? In other words, if a party's nominee for governor does well in a district, does this translate into a larger vote share for that party's state legislative candidate? Previous studies have explored the effects of presidential coattails on congressional elections and demonstrate their impact to be quite limited compared to other factors. The present analysis considers this phenomenon in state legislative elections where gubernatorial coattails are expected to be quite prominent. Given the lower visibility of legislative contests (compared to congressional elections), partisan cues are likely to play a greater role in a voter's calculus, resulting in stronger coattails than observed in presidentialcongressional elections. In addition, the variety of settings in which these elections occur makes it possible to test the influence of several contextual factors that might condition the strength of gubernatorial coattails.
Through the use of a district-level analysis, I examine coattail effects in non-presidential elections in nine states: Alaska, Arkansas, California, Hawaii, Idaho, Illinois, Kansas, Minnesota, and Tennessee. The goal is to determine whether coattails have an influence while controlling for a range of candidate, district, and state-level features. In addition, I am interested in understanding the conditions that might mitigate this relationship. For example, how might factors such as candidacy status affect the impact of coattails? Do we find coattails to be stronger in open seat races where the overwhelming influence of incumbency is absent? Also, what role do institutional features play? Previous aggregate-level studies indicate that legislative professionalism serves to insulate elections as incumbents use the perquisites of office to ward off serious challenge. Are coattails indeed less influential in states that have more professional legislatures? Finally, how might features of the statewide race for governor influence the potency of coattails on the district level? Do more hard-fought gubernatorial races result in greater coattail effects? A multi-state, comparative approach enables me to test several of these important conditional relationships.
STUDYING THE COATTAIL PHENOMENON IN LEGISLATIVE ELECTIONS
A coattail is most often described as a spillover effect whereby an election for an upper-level office influences an election for a lower-level office. Many voters are brought to the polls by a high-profile campaign (such as for president or governor) but end up casting ballots for lower level elections as well (such as for Congress or the state legislature). Strong support in a district for a particular party's candidate for higher office may therefore enhance the vote margin for that same party's candidates running lower on the ballot.
Empirical studies over the years have examined this phenomenon, primarily as presidential coattail effects in congressional elections. Findings from this body of research indicate that presidential coattails do exert an influence (Born 1984; Campbell,1986a; Campbell and Sumners 1990; Flemming 1995; Jacobson 1976, 1997; Mondak 1993), although the magnitude of their effect has diminished over time (Calvert and Ferejohn 1983; Edwards 1979; Ferejohn and Calvert 1984), and their degree of influence varies depending on candidacy status (Flemming 1995; Mondak 1993) and voter attitudes (Mondak 1990; Mondak and McCurley 1994). …