Union University

Union University Department of Political Science

Department of Political Science


Evans Presents Paper on Strategic Retirement in Congress

Jan 14, 2008

This past week, Dr. Sean Evans, Associate Professor and Chair of the Department of Political Science, presented a paper entitled "Strategic Retirement in the Republican House of Representatives, 1995-2006" at the Southern Political Science Association, in New Orleans, LA.

This paper examines whether members of congress retire strategically. That is, politicians leave their office when it is best for them, whether to avoid electoral defeat or to move up to a more prestigious or powerful office. He conduced his research with two colleagues (John Swain of Governors State University and Brian Reed of the University of Montana at Billings).

To test their theory, the authors collected data on every member's decision to retire from the 104th to the 109th Congress (1995-2006). For each member, the authors categorized the member's decision to run for a higher office, run for reelection, or retire from politics altogether. The authors also collected data on three kinds of variables that might explain this decision.

First, aggregate strategic variable affect all members or all members in a party. From this, we expect serving as a member of the president’s party in a good year (popular president and strong economy)is likely to reduce the number of retirements because members of the president's party expect to ride the president's popularity. We also expect that members of the president’s party in a midterm election are more likely to retire. The vote surge that helped elect a member in the presidential election year is likely to decline making vulnerable members of the president's party to retire rather than suffer defeat.

Second, we identify six individual level strategic variables relate to the member's own person electoral situation. First, members from small states are likely to run for higher office because they are more qualified and politically successful than other candidates in the state. Second, when there is an open governor or Senate seat, members are more likely to run for higher office because it is easier to win an open seat than defeat an incumbent. Third, members who have their districts altered by redistricting are likely to retire. Fourth, members involved in a scandal are more likely to retire because they are likely to lose. Fifth, members in marginal districts (won election by a small percentage) are more likely to retire.

Third, we identify eight nonstrategic variables that relate to the benefits and satisfaction derived from House service. First, older members are more likely to retire for health reasons, wanting to do something else, or the strain of politics/campaigning. Second, members in party and committee leadership positions are less likely to retire because of the influence they exert. Third, committee term limits may lead committee chairs to retire due to their loss of influence. Fourth, the lower the salary, the more likely a member is to retire. Fifth, service in the majority increases the probability that one runs for reelection because one can better achieve their policy goals. Sixth, seniors members of the party are likely to retire when their members go from majority to minority status due to loss of power. Seventh, members who are further from the ideological mean of their party are more likely to retire because they are out of step with their fellow party members and thus less effective. Eighth, members who take a term limits pledge are more likely to retire due to their commitment to a citizen legislature.

Our analysis finds that for those who retire from the House, individual strategic and nonstrategic reasons have the most impact while aggregate level strategic factors have little to no impact. All individual level strategic factors are strongly related to the decision to retire while the individual level nonstrategic factors of age, financial stresses, the switch to Republican control, and committee and self-imposed term limits alter the benefits calculation enough that members choose to retire from the House.

In examining the decision to run for higher office, we find support for the progressive ambition thesis as members run for higher office when they believe they will win. As expected, individual level strategic variables explain the decision the most with serving from a small state and open Senate races having a strong, positive relationships with the decision while age has a negative relationship with running for higher office.

In our third model that examines all three decisions at one time, we find that individual level strategic and nonstrategic variables are most important. Serving in a small state and open seats increase the chance of running for higher office while age, redistricting, scandal, taking a term limits pledge, and moving from majority to minority status increase the chance of retiring from congress and politics altogether.

Our results tend to confirm strategic retirement theory in general though we note the low explanatory power of strategic aggregate variables and the importance of context as some variables will explain decisions now (term limits pledge, committee term limits) that will not explain decisions at other times. Second, we find that most members retire voluntarily which contradicts the idea that politics is not enjoyable anymore because most run for reelection or run for higher office. Third, we find that congress tends to reflect social movements provings its responsiveness to the public (e.g., term limits) and that people run for politics for reason than just power (e.g. term limits members run and then return home). Fourth, we see that institutional rules (e.g., committee term limits) affect members decisions. Fifth, the change from majority party to minority party status changes the benefits that members gain from serving, especially for older members who were able to exert more power and influence.