3ie ~ International Initiative for Impact Evaluation

skip to content

The Power of Political Voice: Women's Political Representation and Crime in India

Publication Source

American Economic Journal: Applied Economics, October 2012, vol.4, iss.4, pp.165-93. Available From:

Link to Source
Author
Lakshmi Iyer, Anandi Mani, Prachi Mishra, Petia Topalova
Country
India
Region
South Asia
Sector
Public Sector Management
Subsector
Decentralization
Equity Focus
Gender
Evaluation design
Difference-in Difference (DID)
Status
Journal Article

Methodology

This study examines the impact of mandatory women's representation in local government bodies in India on reported crimes against women. To remedy the historic underrepresentation of women in government positions in India, the Seventy-Third Amendment to the Indian Constitution, which entered into force in April 1993, reserved for women at least one-third of seats on all newly created local government councils.

Mandatory representation of women may affect reported crime against women in several ways. On the one hand, increased visibility of women in government positions may lead to reduced crime against women by prompting police to take complaints by women more seriously, thereby generating deterrence. On the other hand, increased political representation may lead to an increase in crime against women by prompting retaliation by aggrieved men. Alternately, increased political representation of women may lead to an increase in reported crime against women but not necessarily actual crime if women become more likely to report crimes.

To identify impact, the authors exploit variation across Indian states in the timing of the implementation of mandatory women’s representation. This study uses annual state-level data for 1985 through 2007 on rates of reported crimes targeting women, including rape, kidnapping, murder, dowry deaths, and domestic violence. The implementation of mandatory women’s representation is measured dichotomously using data from the Institute on Social Sciences. Using these data, the authors identify impact by estimating ordinary least squares (OLS) difference-in-differences regressions. To determine whether any observed effects are due to changes in actual crime or changes in reporting, separate regressions are estimated for murders, for which reporting bias is expected to be small. Additionally, data from an individual crime victimisation survey in Rajasthan state is used to estimate similar regressions comparing self-reported crime against women between villages where the local council head position is reserved for women and where it is not.

Main findings

The implementation of mandatory political representation for women was associated with significant increases in reported crimes against women. The preferred state-level OLS specifications including the full battery of controls indicate that mandatory political representation was associated with a 46-percent increase in total reported crimes against women, a 23-percent increase in reported rapes of women, and a 13-percent increase in reported kidnappings of women.

Further analysis suggests that these effects can be attributed to increased reporting of crimes against women, rather than an increase in actual crime. First, OLS estimates for the effect of mandatory representation on reported murders of women indicate no significant differences in reported murders before and after the implementation of mandatory representation; insofar as murders of women should be less likely than other crimes to have gone unreported before mandatory representation, this finding suggests that the observed increase in other crimes was due to increased reporting rather than an increase in actual crime.

Analysis of the Rajasthan survey data provides further support for this interpretation. First, there were no significant differences in the incidence of self-reported actual crime victimisation between women in villages where the council head position was reserved for a woman and non-reserved villages. Second, women in reserved villages were significantly more likely to respond that they had reported crimes to the police, compared with women in non-reserved villages. Collectively, these results imply that the increase in crime indicated in the state-level data is attributable to increased reporting, rather than increased crime against women. This, in turn, suggests that reforms aimed at improving the political representation of women can have significant salutary effects on women's lives that go beyond improving the substantive representation of women's interests in political institutions.