Women’s Political Participation and Development

On July 29th, 2019, Brookings India hosted Sonia Bhalotra for a Development Seminar on Women’s Political Participation and Development. She presented two co-authored papers, “Women Legislators and Economic Performance” (2018)^[1] and “Maternal Mortality and Women’s Political Participation” (2018)^[2]. The two discussants for the event were Atishi, leader of Aam Aadmi Party and a noted figure in the education sector, and Mudit Kapoor, Associate Professor of Economics at Indian Statistical Institute, Delhi Centre.

The first paper discussed how there are no costs, in terms of loss of economic efficiency attached to greater female political representation in a competitive electoral system. The second paper highlighted allocated efficiency associated with women representatives in countries that have gender-based quotas in their legislatures, and the reduction of incidence of Maternal Mortality.

The presentation of these papers was followed by Atishi’s insights as a political leader. She pointed out that a female politician cannot rely on intrinsic motivation alone. A more holistic view will entail an understanding of the underlying power structures in political parties. She also noted how strategic motivation (re-election) is an essential part of a democratic system. In her experience, she found that the underlying reason for rent-seeking while in office was the need for cash for re-election.

She summarised that if the ultimate objective of strategic motivation was re-election, then it would be important to understand voter preferences. Does the electorate respond to high quality governance and delivery of social and public goods such as schools and hospitals, or do these matters not find traction in the electorate and mostly responds to cash pay-outs? It is essential to understand this demand side mechanism to forge any changes in the political landscape.

The second discussant, Mudit Kapoor, pointed out that the subjugation of women has been pervasive over time. He argued that the worst form of slavery is not when it is explicit but when it is implicit, that is, one manipulates the other into complying. He quoted esteemed philosophers Francois Fendon (1687) & Adam Smith (1776)- to show how even enlightenment era thinkers thought a woman’s place was strictly within the domestic sphere and should not have access to too much knowledge. Kapoor concluded that in terms of gender quotas, certain human rights are  ends in themselves.

Summary details and key insights of the two papers presented by Bhalotra are presented below:

Women legislators raise economic performance

Raising the share of women in India’s state legislative assemblies is not only likely to lead to better representation of women’s and children’s concerns in policy-making, it is also likely to lead to higher economic growth.

Research suggests that women favour redistributive policies and thus, have tolerance for higher taxes. This could make it plausible that, at least in the short to medium term, women politicians are less effective than men at promoting economic growth. Using comprehensive data for 4265 state assembly constituencies for the period of 1992 to 2012, the study shows that the opposite is  true women legislators not only clock in greater growth, they also lead to more efficient development outcomes.

This study overcomes two challenges that arise in answering the question of whether women politicians are good or bad for growth. These are:

1. a) Constituency-level data on economic activity not being available. To resolve this, the authors use satellite imagery of night luminosity which, previous work shows, is a proxy for economic performance. It was found that it is positively associated with GDP at the state level.
2. b)The challenge for research seeking to estimate causal effects of electing women is that constituencies where women win elections tend to be systematically different from those that elect men in ways that may be correlated with economic performance. In other words, differences between male and female legislators may spuriously reflect differences in voter preferences.

To address this challenge the authors use a previously ratified statistical approach that involves comparing male and female legislators who win against the other gender by a narrow vote margin. The results emerge from a thought experiment that asks how economic growth in a constituency would change if a male leader were replaced by a female leader, all other factors being the same.

Key insights from the paper:

1. Women legislators in India raise economic performance in their constituencies by about 1.8 percentage points per year more than male legislators.  This means that when average growth is seven percentage points, the growth premium associated with female legislators is about 25%. To understand the mechanisms underlying this striking finding, the study explored differences between male and female legislators in:

• Corruption
• Efficiency
• Motivation

Each of the listed factors has been associated with economic growth in developing countries. The study found evidence in favour of women in each case.

2. Male legislators are about three times as likely as female legislators to have criminal charges pending against them when they stand for election. The study estimated that this can explain about one fourth of the difference in growth between male and female-led constituencies.
3. When the authors buttress this result with estimates of actual corruption in office, measured as the rate at which legislators accumulate assets while in office, we find that this is ten percentage points lower per year among women than among men. These findings line up with experimental evidence that women are more fair, risk-averse, and less likely to engage in criminal and other risky behaviour than men.
4. Since economic infrastructure is an important input to growth in developing countries, the study analysed MLA performance in implementation of PMGSY, the massive federally-funded village road construction program. Authors find that male and female politicians are equally likely to negotiate federal projectsfor road building in their constituencies. However, women are more likely to oversee completion of these projects. The share of incomplete road projects is 22 percentage points lower in female-led constituencies. 
5. Finally, separating the sample into swing and non-swing constituencies, we find that women legislators only perform better than men in non-swing constituencies.In swing constituencies, where electoral uncertainty is greater, elected men appear to exert more effort to improve economic growth. One interpretation of this is that men exhibit political opportunism while women display more intrinsic motivation. 
6. Even if gender differences in tendencies towards corruption are intrinsic, if opportunities for corruption decline with development, it may be that women are especially effective relative to men at producing growth in less developed countries. On the other hand, to the extent that women are intrinsically more motivated in public-facing occupations such as politics, they may out-perform men in many environments.

Raising women’s political participation leads to faster maternal mortality decline

Using event study design in a cross-country dataset, the authors examine the impact of gender quotas (instead of competitive elections in the first paper) on Maternal Mortality Rate (MMR). The hypothesis is that MMR is stubbornly high because it is a low policy priority. In terms of the Millennium Development Goals, the global community failed to meet the goal on MMR. The Sustainable Development Goals are more ambitious, and hence, there is a need for focused policy interest in the area.

Interestingly, the decline in MMR, beginning in the 1990s, is coincided by an increase in women’s representation in Parliaments. After the Beijing Convention of 1995, 21 developing countries made reservations for women in their national legislatures. This abrupt change of quotas is used to capture the causal effect on MMR.

Key insights from the paper:  

1. The authors estimate that the introduction of quotas for women in parliament, results in a 9% to 12% decline in maternal mortality. For lower MMR we require trained birth assistance, pre-natal care, and lower teenage pregnancies. As soon as gender quotas are implemented, the authors find a structural break in the following indicators:
   1. Increase in skilled birth attendance (8% to 10%)
   2. Increase in pre-natal care utilisation (6% to11%)
   3. Decline in birth rate
   4. Increase in girls’ education
      It’s critical to point out that these indicators show improvement without a parallel change in GDP, health expenditure or developmental assistance for health.

      2. Without any changes in the allocated resources, either one of the two mechanisms can be at play allocative efficiency (effective targeting of women with no loss in targeting men) versus a substitution effect (more resources for women at the cost of health expenditure on men). With further analysis, substitution effect is ruled out as the authors find that there is no variation in male mortality or that of targeting gender neutral diseases. The authors conclude that women policy makers are more effective in targeting women’s health, that is, allocative efficiency is at play.

3. The authors find reinforcing evidence from the period in which the United States experienced rapid declines in maternal mortality. The historical decline made feasible by the introduction of antibiotics was significantly greater in states that had longer exposure to women’s suffrage.

Hence, gender quotas can be used as an effective policy tool to meet SDG goals on gender equality and MMR.

^[1] Baskaran, T., S. Bhalotra, B. Min, Y. Uppal. 2018. Women Legislators and Economic Performance. IZA Discussion Paper 11596, Bonn.

^[2] Bhalotra, S.R., Clarke, D., Gomes, J. and Venkataramani, A., 2018. Maternal mortality and women’s political participation.