Incumbency is not a barrier to women”s entry into politics, according to research by Quentin Lippmann of the Paris School of Economics, to be presented at the Royal Economic Society”s annual conference at the University of Sussex in Brighton in March 2018.
Why? Because when incumbents cannot run for re-election, a small share of male incumbents are replaced by women but an overwhelming share of female incumbents are replaced by men. This is due to a backlash effect against female candidates only in places where the incumbent is a woman.
These results are obtained in the context of small municipalities in France where over 80% of incumbents are men. The unique election setting of the mayors makes it possible to disentangle the impact of incumbency from alternative factors.
The new study provides important insights for public policy on the impact of term limits on the election of women in politics. It shows that term limits would not increase the share of women in politics, but that they would decrease the persistence of individuals of either sex in positions of power over time.
More…
Incumbency is often assumed to be one of the main obstacles to women”s entry in politics. As incumbents are predominantly men who enjoy an electoral advantage, when they run for re-election, the entry of new candidates, including women, is limited. On the basis of this argument, it is argued that term limits could accelerate the entry of women in politics.
Yet we know very little on the validity of this argument. Previous research has indeed shown that in places where incumbents” turnover rate is high, the share of women in politics is also high but this positive correlation could be due to many alternative factors than incumbency.
For example, if gender discrimination decreased, women would increasingly win against men and this would cause both a higher turnover rate of incumbents and a higher share of women in politics.
The objective of this research is thus to analyse the causal impact of incumbency on the entry of women in politics. The research focuses on access to the position of mayor in small municipalities in France where 80% of incumbents are men. The unique election setting in these municipalities makes it possible to disentangle the effect of incumbency from alternative factors.
Using data from the 2014 French municipal elections, the author concludes that:
• In places where the incumbent is a man, incumbency does impede women from accessing the position of mayor and accounts for 10% of the persistence of male leaders.
• But on the other hand, the persistence of female mayors hinges almost entirely on incumbency. When female mayors are not eligible for their succession, the probability that a woman becomes mayor drops from 70 to 20 percentage points, which accounts for 80% of the persistence.
• As a consequence, for the entire population, incumbency does not act as a barrier to women”s entry in politics because, although a small share of incumbents are women, it protects them a lot more than men.
The fact that incumbency does not act as a barrier to women”s entry in politics on the entire population is surprising and raises several questions. In particular, why are so few male incumbents replaced by women? And why are female incumbents overwhelmingly replaced by men?
Investigating the mechanisms behind these findings, this research finds two channels that drive the results. When a male incumbent is not eligible to his succession, the election of a woman depends almost entirely on the pool of experienced female politicians.
But this factor has little impact on the replacement rate of female incumbents by women. Indeed, when a female incumbent is not eligible for her succession, experienced female politicians suffer from a strong penalty undermining their electoral prospects. This does not seem to be due to a lower ”quality” of female candidates but is consistent with the existence of a backlash effect against women after a female incumbent has been removed.
From a public policy perspective, these findings have two important implications regarding the impact of term limits:
• First, they establish that policies aiming at removing incumbents such as term limits may not increase the share of women in politics.
• Second, while removing incumbents may not necessarily improve descriptive representation through space, it does when descriptive representation is considered through time. In this case the persistence of individuals of same sex at positions of power is key and it should be as low as possible. As incumbency impedes women from replacing male incumbents but also enhances the creation of female bastions, removing incumbents would unambiguously decrease the persistence.
Quentin Lippmann
Paris School of Economics | +33670047253 | qlippmann@gmail.com
