Ambition, Competition, and Electoral Reform: The Politics of Congressional Elections Across Time

3. Institutional Change and Candidate Behavior

3 Institutional Change and Candidate Behavior 3.1 Congressional Candidates in a Partisan Era Congressional elections were very different affairs in the late nineteenth century than they would be in the early 1900s or in the years immediately preceding World War II. The late nineteenth century was the pinnacle of partisan control of congressional elections (Brady, Buckley, and Rivers 1999a). Party machines exercised enormous control over nearly every aspect of the nomination and ballot process from the selection of candidates to stand for each office to the printing and distribution of the ballots that voters used to choose candidates. Party control over the machinery of elections loosened considerably with the adoption of the direct primary and various forms of the Australian ballot in the early decades of the twentieth century. However, the congressional elections literature suggests that a strong candidate-centered system did not emerge in the United States until after the conclusion of World War II. Our argument in this chapter, and more broadly throughout the book, is that many of the important dynamics of congressional elections across time remain unexplored. For example, how did the weakening party control over the machinery of elections affect the behavior of strategic politicians and election outcomes. Did the adoption of the direct primary affect the dynamics of challenger emergence in congressional races? More broadly, how did the changes in electoral institutions affect electoral competition in the United States? This chapter builds the theoretical foundation that we use to provide the evidence for how electoral institutions shaped the behavior of candidates, parties, and voters from 1872–1944. 27 28 Ambition, Competition, and Electoral Reform Due to the almost complete control by parties over the machinery of elections during this era, one would imagine that it would have been difficult—if not impossible—for incumbents or other candidates to develop anything akin to a “personal vote” (Cain, Ferejohn, and Fiorina 1987) since citizens were voting for a party slate and could not easily make office-by-office choices. Swenson (1982, 20) argues that party machines purposefully recruited “docile” candidates rather than those with large personal followings. His account suggests that candidates were selected not for their vote getting abilities, but rather for their loyalty to the party organization. Unlike today’s congressional elections, decisions about whether to seek a place on the ballot were typically not left up to the individual candidates. As Ostrogorski (1964, 196) notes, “Being aware of the fact that the [party] machine holds the keys of the electoral situation ... The candidates of the party are the first to realize that they are not at liberty to attain their object independently of the Machine and still less in opposition to it.” Few would argue that consistently putting party above the interests of constituents is a useful strategy if one is trying to cultivate a longterm relationship with one’s constituency. However, parties from this era were more concerned with maintaining the collective reputation of the party than with recruiting candidates with large personal followings. As Swenson (1982) notes, scholars who have written about this era point out that the electoral institutions in place allowed parties to pursue their collective interests while retarding the individual candidates to leverage their personal reputation. Given these institutional arrangements, it is not surprising that the late nineteenth century is characterized as one of the most polarized as reflected by high degrees of party unity in congressional voting (Poole and Rosenthal 1997). These arguments are certainly not without merit; nevertheless, our own prior research on this subject suggests that this account is far too simplistic and does not adequately explain variation in observed outcomes. We have demonstrated, for instance, that party organizations often sought to recruit experienced candidates to run in competitive congressional districts. Indeed, Carson and Roberts (2005) find that candidates with prior electoral experience were more likely to be nominated for office when national and local conditions favored their candidacy. National economic conditions, an incumbent with a small prior margin of victory, and the absence of an incumbent in a particular district were all directly related to experienced candidates’ entry decisions. Similarly, we have shown that incumbents and experienced Institutional Change and Candidate Behavior 29 candidates during this era responded to changes in district boundaries brought on by congressional redistricting. Incumbents and strategic politicians ran when House districts were drawn favorably and bailed out when facing an adverse gerrymander (Carson et al. 2006...