How do you boost women’s representation in government, ask Jay Euijung Lee and Martina Zanella? Simple gender quotas risk being gamed, for instance by parties fielding women in unwinnable constituencies. The key, they suggest, involves distinguishing between two different kinds of discrimination faced by women in political careers.

In the pursuit of women’s representation in government, gender quotas of various kinds have been adopted by over 130 countries. However, simply implementing quotas is not enough to guarantee meaningful change. Political parties and voters may resist quotas, especially when biases against women’s competence in politics are deeply entrenched. Our recent study shows that even in such settings, however, quotas can gradually reduce these biases if they are designed appropriately.

Our case study is municipal council elections in South Korea. Since their inception in the 1990s, these councils have been overwhelmingly male-dominated, with women making up a mere 2% of elected officials. On top of this, 60% of Korean respondents in the World Values Survey of 2005 agreed that men make better political leaders than women. To combat this gender imbalance, a quota was introduced in 2006 along with wider reforms to the electoral rules for these elections…

An initial backlash but subsequent reversal

The immediate response did not align with the intentions of the gender quota policy as political parties initially sought to circumvent the quota’s impact.

The context is that candidates could be elected through two alternative arms: 80-90% of councillors are elected via a multi-member plurality-vote arm in constituent wards (“ward arm”) while the remainder come from a party-list proportional representation arm (“PR arm”). The quota required that every odd-numbered candidate on the party list for the PR election arm be female. Since most councillors elected through the PR arm were the number-1 candidates from their parties – therefore female – the quota ensured a greater share of women got elected through the PR arm.

However, while parties complied with the quota in the PR arm, they strategically nominated fewer women in the ward arm, where candidate gender was unregulated. The withdrawal of women was especially severe in candidate positions high up on the ballot, where the probability of election was higher. This strategic counteraction allowed parties to comply with the quota while minimizing the actual increase in elected female politicians.

Critically, however, this resistance did not last. Over time, parties began nominating more women in the ward arm, reversing the initial response. Specifically, in municipalities exposed to higher quota intensity, on average 0.41 fewer women were nominated in the ward arm immediately following the quota’s introduction, but 1.10 more women were nominated in the ward arm three elections (12 years) later.

Distinguishing between two types of gender discrimination

We show in the paper that this shift in candidate nomination strategies towards females is due to parties gradually learning that their initial assessment of women as less competent than men was wrong, as the quota forced them to experience more women in action.

As they were forced to experience more women in action, parties gradually learnt that their initial assessment of women as less competent than men was wrong

Central to our study is a distinction between two types of discrimination.

Taste-based discrimination occurs when individuals or parties hold a general dislike against women, such that political parties knowingly compromise the average quality of candidates by selecting less qualified men over more qualified women.

Statistical discrimination, by contrast, happens when fewer women are nominated than men because women are thought to be less competent or because there is more uncertainty (for instance, due to less information available) around their competence.

Distinguishing between these two sources of discrimination is important because it influences what kind of policy is needed to address them. If taste-based discrimination is dominant, efforts should target attitudes and cultural norms, perhaps through awareness campaigns or education initiatives. However, if statistical discrimination is the main issue, interventions can focus on providing concrete information about the true competence levels of men and women, such as performance measures of past male and female councillors.

In our paper, we set out a model that gives contrasting predictions for the two types of discrimination, allowing us to tell the two apart. It predicts that, if quotas dismantled statistical discrimination, political parties would increasingly place women in competitive electoral wards where candidate competence is crucial for success. By gaining new information about women’s abilities, parties would realize that female candidates are more competent than previously thought.

By contrast, if quotas had primarily reduced taste-based discrimination, we would expect to see more women running in less competitive electoral wards. In this scenario, quotas would not alter perceptions of women’s competence but instead foster a recognition that women are simply “not so bad to work with”. As a result, parties would nominate more women overall, but the new women would be less competent than the already-selected “top” women and thus placed in lower-stakes races.

Our empirical analysis finds evidence for the former, leading us to conclude that statistical discrimination to be the main source of discrimination. Moreover, consistent with a lack of information about women being a key condition for statistical discrimination, we observe that the increase in female candidates in response to quotas is more pronounced in municipalities where information was particularly scarce – specifically, municipalities where no women had been elected before the quota.

We also find that the increase in ward female candidates is seen among rookies with zero councillor experience. This demonstrates that experiencing quota women in action triggered parties to reassess the abilities not only of those individual women, but also the abilities of women as a whole.

How gender quotas should be designed – in politics and beyond

From a policy perspective, the South Korean case highlights the importance of ensuring that quotas bring in new women and competent women. Quotas that simply regulate the female share of candidates can lead to women being placed far down on the ballot or in unwinnable constituencies, as seen in Spain and France, limiting gains in female representation among elected councillors.

In other countries, quotas have been misused to bring unqualified female candidates into office, often through nepotism or patronage. For example, in Argentina, India, and Mexico, female family members of male politicians have sometimes filled quota seats, perpetuating existing power structures rather than challenging them. In such cases, biased beliefs about women’s competence may remain unchanged or even be reinforced, weakening the long-term effectiveness of the policy.

In some countries, female family members of male politicians have filled quota seats, perpetuating existing power structures rather than challenging them

The South Korean experience offers valuable insights into how gender quotas can gradually erode biases and lead to a sustainable increase in female representation that, despite initial resistance, could endure even if the quota were later withdrawn. These lessons extend beyond politics to other male-dominated environments, such as corporate boards, where the need for diversity is also becoming more apparent. To quote the European Commissioner for Equality, Helena Dalli, “entrenched selection patterns of corporate board members continue to largely overlook women candidates” everyday.

Sign up here to receive a monthly summary of blog posts from LSE Inequalities delivered direct to your inbox.

All articles posted on this blog give the views of the author(s). They do not represent the position of LSE Inequalities, nor of the London School of Economics and Political Science. 

Image credits: KIM JAE-HWAN via Shutterstock.