Abstract
A puzzle exists over why voters continually re-elect their congressional and state representatives although these same voters often support initiatives to limit the number of terms that their legislators may serve. Residents of California, Colorado and Oklahoma voted in 1990 to limit the terms of their state legislators to tenures ranging from six to twelve years.1 In California these same voters, however, re-elected 87 of 91 (or 96 percent) of all incumbent state representatives and senators whose terms they sought to limit.2 Among re-elected incumbents, 67 (or 77 percent) had served as long as or longer than the maximum term prescribed under their state’s term limit initiative. Voters defeated only two incumbents who would have been removed under the terms of California’s tenure restrictions had they been in effect. Similar voter behavior characterized recent state elections in Colorado and Oklahoma.3 More generally, these seemingly contradictory voter preferences also are apparent in public opinion polls. At the Federal level, while 62 percent of voters immediately prior to the 1990 general elections indicated their intentions to reelect their congressman, that same poll showed that only 24 percent said they “approve of the way Congress is handling its job” (Gallup and Newport, 1990, pp. 36–37). Support for term limit initiatives and the low public esteem of legislators as a group both appear to be inconsistent with voters’ strong preferences for their own representative and the regular reelection of incumbents.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Endnotes
In California, two legislative term limit initiatives were considered by voters in November, 1990. (Under state law, should conflicting ballot initiatives both win, the one receiving the larger number of votes prevails.) Proposition 131, limiting lawmakers to twelve consecutive years service (six terms in the State Assembly and three terms in the State Senate), was rejected by voters 62% to 38%. Proposition 140, limiting members of the Assembly to three terms (a maximum of six years) and State Senators to two terms (a maximum of eight years), was approved 52% to 48% [California Journal (December 1990, pp. 578–79)]. Unlike the defeated initiative, Proposition 140 also bars legislators from ever again holding the same office after having served their maximum term. In Colorado, voters approved an eight-year term limit for all state office holders and a 12-year term limit for federal legislators by a 71% to 29% margin in November, 1990 [Denver Post (November 8, 1990) and Gibbs (1990, p. 39)]. Voters in Oklahoma passed a twelve-year limit on state legislators’ terms in September 1990 [Gibbs (1990, p. 39)]. Another ballot initiative in Washington State would limit politicians in the state legislature to 10 consecutive years, in the Congress to 12 consecutive years, and to two terms as governor or lieutenant governor recently qualified for the ballot [Schnellinger (July 3, 1991, p. B1)]. This last initiative is unique in that it allows sitting politicians who have exceeded the term limits only one more term after the present one. Finally, restrictions were imposed last year on how long municipal officeholders can remain in office in San Jose, CA; Wichita, KS; and Kansas City, MO. [Perry (July 17, 1991, p. A10)].
California Journal (December 1990, pp. 590–606).
Colorado voters removed only one incumbent in 1990 who had served longer than the tenure limits. At the same time, those same voters reelected 12 incumbents who had already served the maximum tenure that would have been allowed under the tenure limits. [State of Colorado Abstract of Votes Cast, various issues and the Denver Post (November 8, 1990).]
Some political commentors have argued that voters would have preferred “genuine campaign-finance reform” but chose term limits as a second best solution [Gibbs (1990, p. 39)]. The problem with this argument is that campaign finance reform initiatives which were on the ballot in California were defeated at the same time the term limits were passed. One explanation for this apparent puzzle is that campaign finance reform actually lessen competition faced by incumbents [see e.g., Lott (1987a) and (1989)].
See also Lott and Fremling (1989) for another example that analyzes changing the composition and level of transfers.
For simplicity, we assume that the number of constituents in these three different groups is fixed. In support of this assumption, Stigler (1971) suggests that politicians may find it optimal to differentiate among constituents using certain characteristics that are relatively costly for voters to change (age, sex, etc.). In such cases, the constituent groups will remain relatively impervious to changes in the structure of taxes and transfers in the political market. [See also Lott (1987b) and Lott and Fremling (1989).]
Entry barriers in political markets [e.g., Crain (1977) and Lott (1986 and 1987a)] provide yet additional explanations for why incumbents may become insulated from political competition.
Voters may be able to effectively sort into office those politicians who intrinsically value the same outcomes that the voters do and thus prevent shirking in the last period [e.g., Lott (1987b); Lott and Reed (1989); Lott and Fremling (1989); Lott (1990); Wright (1989); and Lott and Davis (1992)]. There is also ample empirical evidence that politicians do not deviate from how they have voted in the past when they are in their last term of political office [e.g., Lott (1987b) and Van Beek (1991)]. Other empirical studies suggest that there is little shirking by politicians between the ideal level of output desired by the voters and the politicians [e.g., Bender (1990 and 1991); Goff and Grier (1990); and Peltzman (1984 and 1990)]. However, even if politicians who value the same mix of outputs as their constituents are sorted into office quickly, the interests of constituencies may change over time. This fact combined with the marginal cost differential faced by incumbents and challengers in supplying transfers can make it difficult to remove incumbents when voter preferences change.
This last discussion has testable implications. While tenure creates entry barriers and is thus related to the extent of opportunistic behavior, this does not necessarily imply that politicians will deviate from the interests of their constituents’. The second would measure each politician’s expected remaining tenure. If voters are represented by a politician who is deviating from their interests, their opposition will increase with the length of time that the politician is expected to remain in office. We would thus expect the percentage of a representative’s district that voted for a term limit initiative to be positively related to the degree of shirking and to the expected remaining tenure of those who are shirking. Unfortunately, however, while estimates of expected remaining tenure are theoretically possible, the necessary tests that would have to be done to determine the level of shirking [e.g., Lott and Davis (1992)] are not possible with the data available for state house or senate seats. Alternative tests such as regressing the percentage vote received by a term-limit initiative on such variables as the length of tenure of existing politicians face potentially important pitfalls. For example, a minor concern is that all politicians do not deviate more from their voter’s interests as their tenure increases [Lott and Davis (1992)]. A more important concern is that voters have rational expectations and that if politicians are going to deviate a lot from their constituents’ interests at a later date, voters may be able to infer their representative’s behavior after only a short period in office [Lott and Reed (1989)]. These two considerations suggest that current tenure may be either positively or negatively correlated with support for the term limits.
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 1996 Kluwer Academic Publishers
About this chapter
Cite this chapter
Dick, A.R., Lott, J.R. (1996). Reconciling Voters’ Behavior with Legislative Term Limits. In: Grofman, B. (eds) Legislative Term Limits: Public Choice Perspectives. Studies in Public Choice, vol 10. Springer, Dordrecht. https://doi.org/10.1007/978-94-009-1812-2_14
Download citation
DOI: https://doi.org/10.1007/978-94-009-1812-2_14
Publisher Name: Springer, Dordrecht
Print ISBN: 978-94-010-7307-3
Online ISBN: 978-94-009-1812-2
eBook Packages: Springer Book Archive