The first years of term limits: an initial analysis of the effects on the Michigan House of Representatives.
In response to complaints about unbeatable incumbents, sixteen states now have some form of term limit legislation. Among the claims of term limit proponents is that by largely removing the power of incumbency and essentially leveling the playing field, elections will be more competitive. Through a comparative analysis of the 1990, 1998, 2000, and 2002 elections for the Michigan House of Representatives, this study provides an initial look at elections within Michigan and shows that the effects of term limits on election competitiveness thus far are ambiguous. Competition was enhanced during the 1998 election due to the large number of vacated term-limited seats, yet in 2000 competition levels returned to normal. In fact, when measured by the number of uncontested elections, competition for office was reduced during the primaries to a level below the 1990 election. The 2002 election muddies the water even more. There the general election figures indicate little overall change in competitiveness, while the primary election shows an increase of competitiveness for open seats coupled with an increase in uncontested elections involving incumbents. In essence, this study suggests that once every six years Michigan citizens can expect to see a competitive race (during the primaries) in their individual House district. Between those years competition will decline as strong challengers wait for the incumbent to step down.
The power of incumbency in American politics is legendary. With congressional re-election rates hovering above the 90% level, this reputation is understandable. Critics decry this situation and offer varying plans to even the playing field. The most popular of these is limiting the number of times an individual can serve in an office. In response to complaints about unbeatable incumbents, sixteen states now have some form of legislative term limit legislation. (1) According to the major organizational supporter of this reform measure, U.S. Term Limits, these laws have "changed the political subculture" of American politics. Among the claims of term limit proponents is that term limits have resulted in a more diverse body of people running for and winning office. In addition, they contend that by largely removing the power of incumbency, elections will be more competitive. This essay addresses the second claim (competitive elections) through a comparative analysis of the elections for the Michigan House of Representatives before and after term limits took effect.
The four election cohorts used in this study are 1990, the last nonredistricting election before Michigan adopted term limits; 1998, the first affected Michigan House election; 2000; and 2002. In order to check the claims of term limit proponents I take a two-track approach. First, to determine if term limits have had an effect on the electoral process, I compare margin of victory statistics for all four general elections. Then, because many of Michigan's House districts are in essence one-party districts, the same comparison is made for the primary elections in the same years. Hopefully, this will provide other scholars with an empirical base on which to begin constructing models of the actual effects of term limits.
Many scholars have noted the electoral advantage of incumbency (Bauer and Hibbing 1989; Cox and Morganstern 1993; Jacobson 1987; Fiorina 1977). When measured by electoral margins, this trend has been increasing over the last thirty years (Breaux and Jewell 1992). Incumbent advantage can have a stifling effect on electoral competition. As competition declines, it undermines the foundation of representative democracy. Term limit proponents contend that by limiting terms of office Americans can recover the lost competitiveness of the electoral system.
Most in the academic community have their doubts about the benefits of term limits (Mondak 1995; McCurley and Mondak 1995; Opheim 1994). Some predict a much higher turnover ratio in legislatures (Francis and kenny 1997); others contend term limits may shift the partisan balance of legislatures (Gilmour and Rothstein 1994). Most of these scholars are worried about a concomitant loss in institutional memory and policy expertise. Another concern is that term limits will cause major changes in the composition and sensitivity of legislative bodies (Crook and Hibbing 1997). Scholars are clearly uneasy about the possible consequences of term limits on the legislative process.
The vast majority of work regarding term limits addresses the probable effects on the U.S. Congress (Reed and Schansberg 1995; Benjamin and Malbin 1992; Crook and Hibbing 1997). (2) Studies of state legislative elections are not nearly as plentiful. One study attempts to gauge the effect state legislative term limits will have on U.S. congressional elections (Powell 2000). Some scholars note that the margins of victory of incumbents have been increasing in state legislature elections (Weber, Tucker, and Brace 1991; Garand 1991). Other studies concentrating on state legislatures have sought to project the possible results of term limits by looking at how they would have affected eralier legislatures (Moncrief et al. 1992; Opheim 1994).
The fact that there is not a huge corpus of empirical work regarding term limits is understandable. Due to their newness, most legislatures have not felt their full effects. Some early studies attempt to predict outcomes through studying previous legislative cohorts (Moncrief et al. 1992; Opheim 1994; Thompson and Moncrief 1993). One ambitious study seeks to gauge the impact of term limits by extrapolating data from a survey of nearly 3,000 state legislators (Carey, Niemi, and Powell 1998). Michael Barnhart (1999) has made an initial empirical assessment of the effects of term limits on Michigan. Barnhart's study, however, only reaches 1998--the first "term-limited" election for the Michigan House. In 2000, Michigan experienced its first post-term limit House election. We can now begin to truly assess the effects of term limits on the competitiveness of elections. This study undertakes that task.
According to Barnhart, the mean turnover percentage (as measured by the number of new members at the beginning of a session) for the Michigan House of Representatives from 1988 to 1996 was 18.6% (Barnhart 1999, 52). Low turnover was a large impetus behind term limit proposals. Table 1 shows whether term limits have arrested this trend.
As expected, the first term-limited election (1998) saw a dramatic increase in turnover as compared to the 1990 election (22% to 58%). Of the 110 House members there were only 46 incumbents in 1998 versus 86 in 1990. Indeed, the percentage turnover in 1998 is almost double the mean for all elections between 1990 and 1996.
On initial inspection it would appear term limits have had the desired effect. Yet, when the 2000 election is added to the mix, the story changes dramatically. After the initial term limits turnover the House seems to have returned to homeostasis. The number of incumbents in 2000 is even greater than 1990. In fact, the 19% turnover rate for 2000 is 8% lower than the mean for all pre-term limits elections of the 1990s. These figures are not surprising given that 64 members had been elected for their first term in 1998.
The 2002 election appears to move turnover dramatically toward the 1998 rates (55 incumbents running with a turnover rate of 50%). A closer look at the circumstances of the election, however, reveals that it may be a tremendous anomaly. First of all, 2002 was a redistricting year in Michigan. This is likely one of the major reasons that two incumbents (Fred Durhal Jr. in the sixth district and Belda Garza in the twelfth district) lost in the primaries and one (Patricia Lockwood in the fifty-first district) was defeated in the general election. It should be noted that none of the three was running against a fellow incumbent.
An additional reason for the 2002 election's trend toward higher turnover rates is related to term limits, but not in the obvious way. The election of 2002 was the first time that the Michigan Senate faced term limits. As a result, 31 representatives abandoned their house seats to pursue one of the 31 open senate positions (21 were elected). Conversely, three term-limited senators ran for and won house seats. Of the three, only one (Glenn Steil in the seventy-second district) faced credible opposition in the primaries. One (Dianne Byrum in the sixty-seventh district) faced true competition in the general election--she won with 58% of the vote. This "churning" of elected officials will undoubtedly be one product of term limits and will affect attempts to measure competitiveness.
Overall, competition for the Michigan House of Representatives had been steadily declining since the 1940s (Barnhart 1999). Another hoped-for consequence of term limits was the stemming of this tide. Like the effects on incumbency, the results here are mixed. As Table 2 demonstrates, the mean margin of victory is reduced in 1998; yet the reduction is fairly small, only 2% lower than for all general elections since 1984. In addition, almost all of this reduction disappears in the 2000 election. The margin of victory in 2002 is below the 1984-1996 mean, but again is only fractionally smaller. Keeping in mind the caveats mentioned earlier, this difference may be even less significant than those of 1998.
Perhaps the use of overall means is not a good measure of competition. The figures can be skewed by the presence of several lopsided victories. This possibility can be countered by identifying the number of truly competitive elections. For this (and Barnhart's) study a competitive election is a contest in which the winning candidate garnered less than 60% of the vote. Here term limits appear to make a substantial difference. In comparison to 1990, competitive elections increase by over 50% in 1998. Even the 2000 election shows greater competitiveness than the 1990 contest. When compared to the mean for all elections from 1988-1996, however, almost half of the gain for 1998 is erased. Additionally, competitive races for the 2000 election move below the mean for the years 1984-1996. The 2002 election continues the trend toward more competitive races but does not reach the levels of 1998. In addition, of the twenty-seven competitive elections, only seven involved incumbents.
There is one additional thing of note about Table 2. The number of competitive districts in 2000 and 2002 barely varies from the 1984-1996 mean (24 in 2000 and 27 in 2002 compared to the mean of 25). This is despite a large difference in the number of incumbents (89-55). Perhaps incumbency is not as big a factor in general elections as was suspected.
It is clear that the early returns on term limits do not show the wholesale changes in competitiveness in Michigan's general elections its supporters
were hoping for. There is still little competition in general elections for the Michigan House. Perhaps this is due to the makeup of the individual House districts. The fact that Michigan has many virtual one party districts may have blunted the effects of term limits on general elections. Often victory in the primary is tantamount to election to office. Therefore, it behooves us to check for the role term limits play in the primaries.
Barnhart measures primary election competitiveness for the Michigan House for the years 1984-1998. For his measure he uses the mean number of candidates competing for each seat. He finds that the results of term limits are mixed. The seats of "termed out" districts show increased levels of competition. Conversely, intra-party competition in all other races is at the lowest levels in ten years (Barnhart 1999, 78-9).
While Barnhart's analysis shows ambiguous results, had he used the same methods to measure for primary election competitiveness as he used for the general election he would have discovered a more definite pattern. As Table 3 shows, the 1998 primary election was greatly affected by term limits. Overall the effects are much greater than in the general election. Not only is the mean percentage of winner's vote totals significantly lower (69.76 versus 86.98), but the number of competitive elections is three times higher than they are for 1990 (45 compared to 15).
Barnhart's spurious conclusions regarding primaries are the result of his definition of competitiveness. He uses the number of candidates per contest as his measure. Yet an election with twenty challengers gaining 1% of the vote each is not necessarily a competitive election. Had he continued to use the winning percentage margin as his measure, his results would have reflected the large effects of term limits on primary elections.
Term limits proponents, however, need to temper their joy. The 2000 election shows a return to the old pattern. The percentage of competitive races has virtually returned to the 1990 level. Additionally, the mean winning percentage actually surpasses that of 1990, albeit by only half a percent. Even more ominous is that the number of uncontested primary races in 2000 actually surpasses the number for 1990. The 2002 election arrests the upward trend but fails to match the 1998 numbers.
The 2002 primary election also brings forth one other interesting point. Despite having only nine more incumbents than 1998, the 2002 election had 33 more uncontested primaries (16 versus 49). Of the forty-nine uncontested races, only 11 of those did not involve an incumbent. In addition, of the 11 nonincumbent uncontested races, two were won by term-limited senators (these were the sixty-seventh and seventy-fourth districts). Perhaps term limits stimulate competition for term limited seats, but decrease it in the races where an incumbent is running. By waiting a few years for a seat to open up, viable opponents will avoid the added risk of running against an incumbent.
The early returns on the effects of term limits in Michigan are mixed. Competition was enhanced during the 1998 election due to the large number of term-limited seats. Yet in 2000 competition levels returned to the normal pattern. In fact, when measured by the number of uncontested elections, competition for office was reduced during the primaries. The 2002 general election further demonstrated the lack of effect term limits have on competitiveness. This was not the case in the 2002 primaries. Here the number of uncontested races versus the relatively small number of incumbents demonstrate that the effects on competition will be more ambiguous than proponents proclaim. As time goes on and the number of term-limited seats becomes more dispersed across elections, it is not unreasonable to expect that total competitiveness will increase. This increase will, however, be of an uneven sort.
Essentially, this study suggests that once every six years Michigan citizens can expect to see a competitive race (during the primaries) in their individual House district. Between those years competition will decline as strong challengers wait for the incumbent to step down. The tenuousness of these conclusions should not be understated. This study has provided an initial look at the effects of term limits on elections within Michigan. The full effects will not be known for several more elections.
TABLE 1. Incumbents in the General Election and Turnover Rates Year 1990 1998 2000 2002 Mean 1990-96[dagger] Incumbents 86 46 89 55 80 Turnover % 22 58 19 51[double dagger] 27 [dagger]Source: Barnhart (1999) Michigan Secretary of State; Michigan Manual. [double dagger]Reflects the defeat of one incumbent in the general election. TABLE 2. General Election Competition, Michigan House of Representatives, 1984-2002 Year 1990 1998 2000 2002 Mean 1990-96[dagger] % winning vote 71.25 67.75 69.31 67.33 69.81 # competitive 20 31 24 27 25 elections % competitive 18.18 28.18 21.81 24.54 22.72 [dagger]Source: Barnhart (1999). Michigan Secretary of State; Michigan Manual. TABLE 3. Primary Election Competition: 1990, 1998, 2000, and 2002. Year 1990 1998 2000 2002 % winning vote 86.98 69.76 87.58 74.88 # competitive 15 45 17 38 elections % competitive 3.63 40.90 15.45 34.54 elections # of uncontested 68 16 71 49 elections Mean 1984-96[dagger] Source: Michigan Secretary of State; Michigan Manual. [dagger]Figures not available
The author wishes to thank Timothy Bledsoe, Debra Viles, and Ellen Sawyer for their careful reading of this article. In addition, he wishes to thank the anonymous readers for the Michigan Academician for their helpful suggestions.
1. In addition to the 16 states that currently term limit their legislatures, the Idaho and Utah legislatures repealed their limits in 2002 and 2003 respectively. Also, Oregon's term limits law was struck down by its supreme court in 2001. Source US Term Limits at www.termlimits.org/Current_Info/State_TL/index.html.
2. One exception to this is Wayne State University's ongoing massive study of term limits in Michigan.
BARNHART, MICHAEL. 1999. "A Study of Term Limitation and the Michigan House." Ph.D. Diss., Wayne State University.
BAUER, MONICA, AND JOHN R. HIBBING. 1998. "Which Incumbents Lose in House Elections: A Response to Jacobson's 'The Marginals Never Vanished.'" American Journal of Political Science 33:262-71.
BENJAMIN, GERALD, AND MICHAEL MALBIN. 1992. "Term Limits for Law Makers: How to Start Thinking About a Proposal in Process." In Limiting Legislative Terms, eds. Gerald Benjamin and Michael Malbin. Washington, DC: CQ Press.
BREAUX, DAVID, AND MALCOLM JEWELL. 1992. "Winning Big: Incumbency Advantage in State Legislative Races." In Changing Patterns in State Legislative Careers, eds. Gary F. Moncrief and Joel A. Thompson. Ann Arbor: University of Michigan Press.
CAREY, JOHN M., RICHARD NIEMI, AND LYNDA W. POWELL. 1998. "The Effects of Term Limits on State Legislatures." Legislative Studies Quarterly 23:271-300.
COX, GARY W., AND SCOTT MORGANSTERN. 1993. "The Increasing Advantage of Incumbency in the U.S. States." Legislative Studies Quarterly 23:495-514.
CROOK, SARA BRANDES, AND JOHN R. HIBBING. 1997. "A Not So Distant Mirror: The Seventeenth Amendment and Congressional Change." American Political Science Review 91:845-53.
FIORINA, MORRIS. 1997. "The Case of the Vanishing Marginal: The Bureaucracy Did It." American Political Science Review 71:177-84.
FRANCIS, WAYNE L., AND LAWRENCE W. KENNY. 1997. "Equilibrium Projections of the Consequences of Term Limits Upon the Tenure, Institutional Turnover, and Membership Experience." Journal of Politics 59:240-52.
GILMOUR, JOHN B., AND PAUL ROTHSTEIN. 1994. "Term Limitation in a Dynamic Model of Partisan Balance." American Journal of Political Science 38:770-96.
GRAND, JAMES. 1991. "Electoral Marginality in State Legislative Elections, 1968-86." Legislative Studies Quarterly 26:7-28.
JACOBSON, GARY C. 1987. "The Marginals Never Vanished: Incumbency and Competition in Elections to the U.S. House of Representatives." American Journal of Political Science 31:126-41.
MCCURLEY, CARL, AND JEFFERY J. MONDAK. "Inspected by #1184063113: The Influence of Incumbents' Competence and Integrity in U.S. House Elections." American Journal of Political Science 39:864-85.
MONCRIEF, GARY, JOEL A. THOMPSON, AND WILLIAM CASSIE. 1996. "Revisiting the State of U.S. Legislative Research." Legislative Studies Quarterly 21:301-26.
MONDAK, JEFFERY J. "Competence, Integrity, and the Electoral Success of Congressional Incumbents." Journal of Politics 57:1043-69.
OPHEIM, CYNTHIA. 1994. "The Effect of U.S. State Legislative Term Limits Revisited." Legislative Studies Quarterly 19:49-61.
POWELL, RICHARD J. 2000. "The Impact of Term Limits on the Candidacy Decisions of State Legislators in U.S. House Elections." Legislative Studies Quarterly 25:645-61.
REED, W. ROBERT, AND D. ERIC SCHANSBERG. 1995. "The House Under Term Limits: What Would It Look Like?" Social Science Quarterly 76:699-716.
THOMPSON, JOEL, AND GARY MONCRIEF. 1993. "The Implications of Term Limits for Women and Minorities: Some Evidence from the States." Social Sciences Quarterly 74:300-9.
WEBER, RONALD, HARVEY TUCKER, AND PAUL BRACE. 1991. "Vanishing Marginals in State Legislative Elections." Legislative Studies Quarterly 16:29-45.
ENOCH W. BAKER
Wayne State University
|Printer friendly Cite/link Email Feedback|
|Author:||Baker, Enoch W.|
|Date:||Jun 22, 2004|
|Previous Article:||It takes a village: cultural border crossings in Rue Deschambault and Ces enfants de ma vie by Gabrielle Roy.|
|Next Article:||Herpetological assemblages of the Michigan Regional Landscape Ecosystems.|