Strategic Retirement In The U.S. House Of Representatives, 1897–1996: “You Can't Fire Me, I Quit”

Abstract

The fact that retirements rather than election defeats are the primary source of turnover in the U.S. House of Representatives has led some to conclude that voters' preferences have little influence on the composition of the House. Jacobson and Kernell's theory of strategic retirement suggests that many members retire to avoid likely defeat, so that one would expect a party's retirement rate to increase during an election cycle expected to be unfavorable to it. The theory is tested by specifying and estimating a multivariate regression model of party retirement rate including variables representing strategic, nonstrategic, and mixed motivations for retirement; the database consists of both parties' retirement rates for the 55th through 104th Congresses. The results suggest that the president's party members are more likely to retire during a midterm year, but this effect depends on model specification. Other strategic variables show no significant effects on party retirement rate. Separating the data by political party, type of retirement and party alignment period yields little evidence of a strategic retirement pattern that might have been obscured in the overall analysis. Additional research on individual‐level forms of strategic retirement and alternative forms of strategic behavior is suggested.