The last period problem in politics: Do congressional representatives not subject to a reelection constraint alter their voting behavior?
✍ Scribed by Mark A. Zupan
- Book ID
- 104630972
- Publisher
- Springer US
- Year
- 1990
- Tongue
- English
- Weight
- 703 KB
- Volume
- 65
- Category
- Article
- ISSN
- 0048-5829
No coin nor oath required. For personal study only.
✦ Synopsis
In a recent cross-sectional study of Senate roll call voting, Joe Kalt and I (1990) find that the extent to which legislators faithfully represent the interests of their constituents is negatively related to the legislators' security in office. Where security in office, as measured by tenure or most recent margin of electoral victory, is higher, that is, legislators' voting behavior less closely reflects underlying constituents' economic interests and ideological preferences. Contrary to the capture theory of politics (Peltzman, 1976; and Stigler, 1972), our findings suggest that policing of policymakers by constituents is imperfect. In specific instances of policymaking as well as on the broad bundle of issues addressed during their terms in office, policymakers appear to have some leeway to pursue their own independent ideological preferences.
John Lott (1988), among others, has correctly pointed out that our crosssectional results are subject to an important simultaneity bias. While greater security in office may lead to more ideological "shirking" (Alchian and Demsetz, 1972;and Jensen and Meckling, 1976) by elected representatives, more ideological shirking also diminishes a representative's security in office.
As Joe Kalt and I (1990) note, the presence of a simultaneity bias works against finding the results that we report. Indeed, it is perhaps because of such a simultaneity bias that Lott (1987) finds legislators' voting to be unrelated to the presence of a reelection constraint. For both retiring and nonretiring congressional representatives, Lott examines the (absolute value of the) changes