When women are in charge


This page in:

In 1993 India adopted gender quotas for local councils. In particular, the position of chief councilor (or Pradhan) was reserved for women in 1/3 of the village councils in any given election – and this 1/3 was selected at random.   As one might expect, this has led to a surge in the number of women holding this post. It also provides a ripe environment for impact evaluation work.  

The papers that have come out show that reserving seats for women results in a change in investment priorities to be more in line with female preferences (infrastructure in particular), that the reporting of crimes against women increased, that attitudes about the competence of female politicians improved, and that women were more likely to be elected after the reservation moved on (and this is not an exhaustive list).  

The latest piece of evidence to come out of this setting is an interesting paper by Beaman and coauthors which looks at the impact of these quotas on the aspirations of girls, and their parents.    And they show the impact of these new female politician role models can be quite powerful.

First, the set-up. Beaman and co. use data from West Benegal, where they did interviews with a “prime-age” male and female per household, as well as all of the adolescents (aged 11-15).   They use the assignment of this reservation in West Bengal, but the way it was rolled out means not only can they look at the effect of a female reservation versus none, but they can also look at dose-response since some communities have had two cycles of female representation (and some had had none by the time they collected their data).  

To start with, they have four measures of adult aspirations for their children: desired educational attainment, desired age of marriage, preferred occupation at age 25 and whether the parent wished for the child to become Pradhan.   With a bit of transformation and combination, they have some results on changes in aspirations.  

First, in villages that have had no reservations, parents have higher aspirations for boys – the aspirations for girls are 0.59 of a standard deviation lower.   But in villages that have had a female leader for two election cycles, the gender gap in aspirations drops by 0.17 standard deviations, with a lot of the action coming in aspirations for children’s occupation. And a lot of the push is coming from mothers – these improve in terms of education and occupation (broadly) while dads show improvements in their wish for their daughters to be a Pradhan.  

Among the kids, the girls aspirations also improve and close the gap in these twice-reserved councils.   Girls are less likely to want to be a housewife, less likely to want their in-laws to determine their occupation, more likely to want to marry after 18 and more likely to want a job that requires more education.  

This is mirrored in some tangible behavior.   In non-reserved areas, boys are significantly more likely to be in school, and are more likely to be able to read and write. Compared to this, in areas with two cycles of reservation, “the gender gap in educational outcomes is completely erased (and even reversed, and girls spend less time on household activities” – 18 minutes less.   And this only starts happening when women are in power for the second time due to the reservation – showing that this second dose matters a lot.  

To solidify that this seems to be driven by a role model effect, they present evidence which undercuts other possible explanations.   First, expectations aren’t at work here: there isn’t an impact of these reservations on the career choice, education or labor markets of young adults (those 16-30 years old). Second, the women Pradhans who are put in place via these reservations don’t invest more in schools, so it’s not an education supply story.   Moreover, you also have the work mentioned above which shows that opinions about the suitability of the women who lead a particular council doesn’t change until the second cycle of reservation.  

So this shows us that the use of political quotas, in this Indian context, not only changes how people view their politicians, but also how they view their children and their chances and how the children view their own aspirations – with attendant changes in behavior.    And this changes within a generation.  



Markus Goldstein

Lead Economist, Africa Gender Innovation Lab and Chief Economists Office

Join the Conversation

February 14, 2012

This is very interesting. I wonder what your thoughts are on confounding the cause and effect though. Perhaps it is because certain places have better gender relations (or more women are empowered) that women Pradhans were voted in the first place? Whereas other places have no women leaders because of higher gender inequality? The 1/3 reservation was for all villages I assume?

Markus Goldstein
February 14, 2012

What you point to makes studies of this type particularly hard. However, in this case because the reservations were randomly assigned to 1/3 of the village councils in any given election, you have random variation that puts a woman in power as head of the village (and importantly, that variation is not correlated with the underlying charachteristics of the village). Hence, this allows for the examination of the effect of women leaders on a range of outcomes and you can be sure that the causality runs from the leader.

Francisco mejia
February 15, 2012

Really powerful post. Two questions/comments

I read recently (don't remember where) in an IE that having a daughter (as opposed to a son) changes gender perception, and discrimination in parents. Was that controlled?

And, related, did live births ratios change as a result of the quotas? If administrative data is available, it can be tested That would make this story even more powerful.

Anita George
February 19, 2012

A very powerful piece of analysis. I wonder if it would be possible to contrast with a state like Kerala which has had a tradition of educating women.

March 14, 2012

Hi Markus,

I love the good job you & your colleagues (I also follow Berk Ozler's posts) are doing. At present, I am living in the US where the political campaign led by the GOP heavily attacks women's rights as regards abortion and sexual & reproductive health rights.

Back in Europe, where I come from, quotas are being disregarded as discriminatory and ineffective, and, worse, as unnecessary. Few impact analysis are conducted there to truly understand the effects of quotes and other policies that actually go beyond positive action (I do not think quotes can be considered affirmative action, by the way).

What I mean is that what passes for research and evaluation on development is a stronghold of democracy for countries where leadership on this issues sits, or should anyway.

Your post has been disseminated through the European Gender Budgeting network and one of my colleagues wrote to me today to say that they will use your post in a upcoming training.

I wanted you to know that what you are doing is not only interesting from the point of view of evaluation but should bolster evidence-based advocacy to make a dent in politics and challenge complacent western culture.

Tremendous thanks from NY,