It’s never too encouraging when, in the course of your research, you find that most scholars writing on your topic have taken the opposite position.
After the landmark results of the 2010 Mid-term Congressional Election, it appeared as though congressional contests had gotten more competitive. High-stake districts all over the country were popping up in the national news media, the Tea Party was commanding attention in vulnerable districts, and campaign spending was at an all-time high. While observing this dynamic election, I revisited David Mayhew’s seminal work on congressional electoral competitiveness: “Congressional Elections: The Case of Vanishing Marginals.” In his article, Mayhew claims that “marginal” House elections, those that yield a narrow vote margin, will occur less and less as time goes on. Using measurements of congressional elections from 1956 to 1972, Mayhew argues that this decline in “close” House races makes any resulting seat changeovers less significant.
A reassessment of this argument was in order, especially given the magnitude and rapidity of shifting party control that began in 2006. The American electorate watched party control of the House switch hands twice in only four years; a trend that had not been observed since the 1950-54 period.
When I moved beyond Mayhew to find more recent studies, I thought I would find a good number of scholars that would support my assumption that Mayhew had prematurely declared the death of congressional competitiveness. What I actually found was that most had written in support of Mayhew. Some focused on the effect of redistricting, others on the role of incumbency advantage, but all concluded that House elections were going to remain stagnate.
There were, however, a few studies that offered hope in mounting a successful challenge to the Mayhew camp. One in particular by Gary Jacobson provided a useful framework for the refutation. In his article “The Marginals Never Vanished: Incumbency and Competition in Elections to the U.S. House of Representatives, 1952-82,” Jacobson offers an effective counter to the Mayhew-born definitions of “competitive” and “marginal” and challenges the degree to which competition has declined, if at all.
Common throughout all of the works that I encountered in my research, both the supporters and the dissenters, there lacked a study that applied the commonly used metrics of analysis to the most recent elections. In fact, most of the significant papers only offered assessments up until the late 1980s or early 1990s. It was clear that, regardless of the high number of Mayhew supporters, the most recent congressional contests needed to be factored into the discussion.
Predominately, using the path forged by Jacobson, I evaluated congressional elections from 1950 to 2010. My evaluation included measuring trends in incumbent vulnerability, vote marginality, and seat changeover sensitivity (seat swing ratio). Through all of this I found that Mayhew did indeed identify a break point for congressional elections in the middle of the 1960s. Before 1966, the percentage of “marginal” elections (defined as those elections where the winner received 50 to 60 percent of the vote) fell between 50 to 55 percent of all House elections. Since 1966, however, the amount of “marginal” House elections has hovered between 45 and 20 percent of all elections. While the electoral behavior observed after this break point operates in a lower range, the results tend to present greater variability.
When the percentages of “marginal” elections are presented graphically over time, the results seem to resemble a sine curve after the 1966 break point. The peaks and valleys show high variability and more or less coincide with each decade. At the beginning of the decade, the amount of marginal races is low; whereas, towards the end of the decade, the amount of marginal elections tends to increase. This is especially interesting given the discussion on redistricting and its effects on the results of the 2012 election.
I argue in my paper that Mayhew identified a decline in the overall percentage of competitive races in a given House election, but that does not necessarily mean that the fight for party control is any less competitive or any less significant. There may be a smaller percentage of competitive races, but this small percentage can still greatly impact who controls the House.
While Mayhew’s work commands a great deal of authority and following in the subject of congressional electoral competition, I don’t feel as uncomfortable as I did before. A reevaluation, regardless of the outcome, was in order.