Economic development and crime against women

With economic development, technology becomes more skill-intensive than brawn-intensive, increasing the earning potential of women. Analysing Indian data from 2004-2012, this article shows that the decreasing gender gap is associated with more rapes and indecent assaults against women, particularly in areas where gender bias is high. The key mechanism behind this is male backlash against female empowerment, in settings where social institutions have traditionally favoured men.

The World Bank (2018) identifies violence against women and girls as a global pandemic that is not just devastating for victims but also has significant economic costs¹. Continued violence against women and girls in India is an increasingly salient societal and policy concern. The murder of a young female doctor in Kolkata in August this year has sparked nationwide grief and outrage. The tragedy begs the question of whether growing educational and career aspirations among women (who are responding to higher returns to education and greater earning potential) spark backlash from men, who see this phenomenon as women going against their socially allocated place at home as per patriarchal norms.

The economic literature investigating drivers of crime against women has largely focused on domestic violence (Sekhri and Storeygard 2014, Cools et al. 2015, Angelucci 2008, Hidrobo et al. 2016, Amaral 2017, Amaral and Bhalotra 2017) In a recent study (Bandyopadhyay, Jones and Sundaram 2020), we take a unified approach to analyse two distinct types of crime – domestic violence and crime against women outside the home – and propose female empowerment as a driver of both types of crimes. We highlight that female empowerment can impact these two very different types of crime differently.

Crime against women in society

Our argument is that in a gender-biased society, female empowerment can engender a hostile response, giving rise to crime against women outside the home. This is known as a ‘backlash effect’. In our framework, two groups in society that we refer to as “patriarchs” and “feminists” have opposing preferences over the degree/extent of women’s rights (or gender bias in social institutions). A shock to technology that makes it more skill-intensive (relative to brawn-intensive) increases the relative productivity and earning potential of women (Galor and Weil 1996, Rendall 2017). This decrease in the gender gap empowers feminists to invest in a political contest for greater women’s rights, causing alienation among patriarchs. The patriarchs respond by also investing more in the political contest, resulting in increased hostility between the two groups, which in turn raises the likelihood of crime against women in society. This narrative is consistent with the idea that shifts in power, or changes in the status quo in a gender-biased society can provoke male backlash (Pande 2015).

Domestic violence

The impact of a decrease in the gender gap on domestic violence is not necessarily the same across contexts. While a decrease in the gender gap increases women’s bargaining power in the home, the increase in crime outside the home/in society may limit their ability to exercise the expanded outside options, mitigating this effect. Also, women may respond to an increase in their relative earning potential by exercising more managerial control vis-à-vis their husbands, which might trigger more domestic violence. We argue that the ultimate impact of a decrease in the gender gap on domestic violence depends on the gender bias in social institutions that determine both crime outside the home and the deterrents to domestic violence.

Institutional gender bias

Our study utilises district-level data from 2004-2012 on crime against women in India – sourced from the National Crime Records Bureau (NCRB), combined with information on labour market outcomes for individuals aggregated to the district level from the National Sample Survey Organisation’s (NSSO) Employment and Unemployment surveys – to study the relationship between the gender gap in earning potential² on the one hand, and domestic violence, rapes and indecent assaults (the latter two categorised as crimes outside the home) on the other.

We harness the wide variation in culture and institutions across Indian states to delve into the role of gender bias in social institutions in driving the relationship between gender gaps and crime against women. We use the proportion of “missing women”³ in each district (Anderson and Ray 2010, 2012), the percentage of females with access to Essential Services and Opportunities (ESO)⁴, sourced from McKinsey, and the percentage of households with women reporting that they practice “purdah” or other forms of religious face coverings⁵, sourced from the India Human Development Survey (IHDS) for 2005. These measures capture a wide range of social, cultural and institutional gender bias across India.

Male backlash

We show that a narrowing of the gender gap is associated with more rapes and indecent assaults against women, particularly in areas where gender bias is high. We are able to show that this backlash effect is not driven by a decrease in under-reporting of crime against women. We do this in three ways. First, we demonstrate that the backlash effect is stronger in areas of high gender bias where institutions are traditionally in favour of men. Second, we show that the effect comes from a narrowing of the gender gap among low-educated and not high-educated individuals, and finally, we find that qualitative results are borne out in the IHDS (2005, 2011) data, wherein reporting is a less serious issue relative to administrative data (since individuals are interviewed in their own homes about their experience with crime, which is less intimidating and stigmatising than filing a report at a police station). In the context of crime inside the home, we find that a narrower gender gap lowers domestic violence, consistent with an increase in the woman’s bargaining power. However, this relationship holds only in regions of relatively low gender bias, as captured by fewer “missing women”, a higher ESO index, and fewer households practicing purdah.

Conclusion

Our study relates to the macroeconomic literature exploring the relationship between female empowerment and economic development (Doepke and Tertilt 2019). We provide empirical support for the idea that as economic development propels technological change that increases the relative productivity of women, persistent societal gender bias can generate hostility towards female empowerment and in turn drive an increase in crime against women.

Notes:

The World Bank estimates the economic cost of violence against women in Latin America as 3.7% of gross domestic product (GDP).
Earning potential is the wage that an individual of a particular gender can expect to earn on average in the district. For each district, it is calculated as the gender-specific weighted average national wage across industries, where the weights are industry shares in the first period (2004). Constructing a measure of earning potential in this manner ensures that it is arguably unaffected by factors related to crime in the district. We refer to the ratio of earning potential of men to women as the gender gap.
“Missing women” is an indicator of son preference, which is a cultural measure of gender bias.
The ESO measure captures institutional barriers faced by women relative to men.
The prevalence of purdah picks up the extent to which women are restricted from public spaces, pointing to the presence of gender norms that prescribe spaces and roles to women.