Term limits and women’s representation: a Democratic opportunity and a Republican dead-end

ABSTRACT

When legislative term limits were proposed, one expected impact was an increase in women’s legislative representation. However, researchers have found that this logical solution is not as effective as many had hoped. Because of findings of a partisan gap in the pipeline of potential female candidates, we argue that studies must consider the partisanship of the female legislators when analyzing the effect of term limits on increasing the number of female state legislators. This study utilizes a cross-sectional time-series analysis of 49 states from 1990 to 2014. Our analyses suggest that Democratic women benefit from term limits while Republican women do not.

KEYWORDS:

• State politics
• women and gender
• legislative
• term limits

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 Due to our interest in partisan differences, we exclude Nebraska from the analysis since it has a non-partisan state legislature.

2 The 14 states and the year term limits went into effect are AR (1999), AZ (2001), CA (1997), CO (1999), FL (2001), LA (2009), ME (1997), MI (1999), MO (2003), MT (2001), NV (2010), OH (2001), OK (2005) and SD (2001). The date of effect is taken from the National Conference of State Legislatures (www.ncsl.org).

3 We use the xtreg command in STATA to estimate cross-sectional time-series regression results with random effects.

4 This information is from the publically available information on the website of the Center for American Women and Politics (www.cawp.rutgers.edu).

5 Because the Termlimited variable incorporates term-limited turnover relative to normal turnover, we believe this measure is superior to using the raw percentage of seats actually term-limited out in a given state. The advantage of using the actual raw percentages is that the values can be measured for each particular year. To be thorough, we estimated the equations with the actual percentage of seats term-limited out in a given year instead of the Termlimitedness measure. There was no substantive difference in the results. The only slight difference was found in the equation used to explain the amount of Democratic women as a percentage of the upper chamber (see Table 2). 

6 These figures were taken from the Census Bureau which releases the numbers every five years. The figures for the year closest to a given election were used.

7 Although the Elazar typology is a crude measure, it is a well-established and accepted one. We also estimated the equations using the citizen ideology scores developed by Berry et al. (2007). No substantive difference in the findings for term limits resulted from the use of the different measure for state ideology. The results for citizen ideology suggest that the percentage of Democratic women as a percentage of both the Democratic caucus and of the entire legislature is higher in states with more liberal citizen ideologies. Conversely, states with more conservative citizen ideologies have fewer Republican women as a percentage of the overall legislature with no effect found for the percentage of women in the Republican caucus.