Keeping women safe
Since the 2012 Delhi gang rape, numerous policies have been proposed to improve the safety of women. However, perversely, many of the policies designed to protect women may have made them less safe
In December 2012, thousands of protesters flooded the streets of cities across India, demanding a safer environment for women. A 23-year-old female student had died from injuries sustained 13 days earlier, when six men raped and savagely beat her on a Delhi bus. The case gained international attention, and since then South Asian media have reported dozens more horrifying instances of violence against women, several involving tourists: a Danish woman was gang-raped in Delhi after asking for directions back to her hotel, and an American was raped while hitchhiking in the Himalayas.
These cases may reveal an increase in violence — or perhaps they reflect increased reporting by women to police, or heightened media attention. But they have certainly made a big impact on policy discourse in India. In his first Independence Day speech, India’s current Prime Minister, Narendra Modi, chided the entire country, saying, “Today when we hear about these rapes, our heads hang in shame.” And since December 2012, numerous policies have been proposed (and several enacted) to stop this “war on women”. But how many of these short-term protective measures will fail, or even create perverse incentives opposite to those intended — possibly leaving women less safe in the long run?
As an economist, I want to understand, in context, the social, legal, and financial forces that lead individuals, families, and society to undervalue women and unleash violence against them. Contemporary economic scholarship embraces such subjects, employs new tools to evaluate behaviour, and draws on fieldwork for first-hand knowledge of subjects. I draw on all of those elements in my own work, and in that of colleagues’, to understand the ways in which girls and women are harmed: actively, through abuse; or passively, by passing them over and devoting resources to boys. And then I want to apply that understanding to determining which policies can weaken these destructive forces or re-channel them in women’s favour.
Examining differences between South Asia and other parts of the world, or among villages and communities in South Asia, can help illuminate why girls and women receive fewer resources and are often harmed. We can look at trends over time, analyse large sets of data, and in some cases collaborate with governments and non-governmental organisations to conduct rigorous evaluations that directly explore how changes in policies and programmes affect women’s well-being. And we can combine this evidence with insights from economic theory to predict behavioural responses to policies and, therefore, their likely impact: unintended damage, deterrence, or — in some cases — enduring improvements in women’s security. We quickly learn that safe and protected are hardly the same thing, particularly in the long run.
How not to deter rape
The protests following the horrific Delhi bus case, for instance, led to new anti-rape laws in 2013 that criminalised a range of sexual assaults and allowed for the death penalty. In 2014, perhaps scrambling to temper a too-harsh law, Indian courts declared that forced marital sex is not rape. This is an example of how a policy intended to protect women may have made them less safe.
A basic economics insight is that policies that seek to deter crime will succeed only when the expected cost of committing the crime increases. But perpetrators of sexual violence are often known to the victim. If a woman or those close to her believe that a member of her social circle — or even her family — might be put to death, she may face tremendous pressure not to report the rape, and refrain from doing so. If this reaction is widespread, then the law may have the perverse consequence of reducing the expected cost of the crime.
Consider also that if many members of India’s largely male police force believe that women incite rape by how they dress, for example, they may be unwilling to implement the law. Vulnerable women who make it to the police station may be taunted rather than helped. Again, bad news for women.
In addition, in the wake of the Delhi bus case, several of the city’s colleges and working-women’s hostels tightened their curfews for female students, citing the “atmosphere in the city.” But restricting women’s ability to move about the community is likely to have an unintended effect. There is strength in numbers, and for those few women who are out at odd hours, curfews increase the risk. Perhaps Breakthrough, the international women’s rights organisation, had the right idea when it organised mass “Board the Bus” demonstrations to increase visibility: the key to women’s safety is not protecting them indoors, but increasing and even advertising their public presence.
These observations are supported by economic theory, but they remain conjecture until they are tested. So what is the evidence on the forces driving mistreatment of women, and how does that knowledge support or contradict society’s responses?
Work and bargaining power
First, the home remains the most dangerous place for Indian women. Many recent high-profile rape cases in India relate to unmarried women returning late at night from work or entertainment, or teenage girls who go out of the house alone. That makes it appear that a young woman’s decision to continue her studies or to work, rather than marry, is also a decision to risk a much higher incidence of violence. But that is not the case. The first and third leading causes of unnatural deaths for Indian women in 2013 were suicide and fire. A significant fraction of those are likely linked to domestic violence, especially because dowry deaths (revenge perpetrated on young brides for unfulfilled financial promises) often take the form of forced suicide or burning.
We also know that women who marry later face less violence in the home than those who marry early. A study on the age of marriage in Bangladesh showed that later-married girls received more schooling and more preventive healthcare. IGC research demonstrated that the arrival of garment-sector jobs increased girls’ education and caused women to delay marriage and enter the labour force. Thus, even though, in the short term, unmarried working women may expose themselves to attack, the longer-term perspective must shape policy. A society where daughters and sons are valued equally, and where women face less violence, is a society where women enter the labour force, marry later, and enjoy more bargaining power over household budgets, family decisions, and so on within the home once they do marry.
In countries with wide gender gaps, increasing women’s bargaining power — through employment and decent wages — should be a goal of public policy. (A study in the United States suggests that increasing women’s wages relative to men’s decreases domestic violence. Notably, this effect held true for working and nonworking women alike; a greater expected wage seems to increase female bargaining power, and with it, women’s safety.) Governments have some control over opening the labour market to women, and creating channels for women to gain positions of authority. Clearing the way for women to enter the workplace would thus increase their safety.
Job opportunities can have myriad positive effects in South Asian women’s lives. Researchers studying villages targeted by recruiters for a new woman-friendly industry found that young women in villages visited by a recruiter were less likely to get married, showed delayed fertility, and were more likely to report wanting to work. One of my doctoral students found that women employed in a textile mill who worked longer into their lives delayed marriage without any detrimental effect on eventual spousal quality, and reported desiring fewer children. These effects spilled over, too: marriage age increased for their younger sisters — thereby likely empowering them, too.
Recently, women’s groups and trade unions opposed a move to change India’s factory law to allow women to work the night shift; they were concerned that there were insufficient protections to prevent attacks. But research indicates that this concern, though understandable, is misguided. Call-centre work has created huge benefits for South Asian women, many of whom work at night. Keeping women “protected” from night factory work means holding them back.
The social context
Comparing South Asia to the rest of the world reveals a glaring demographic fact: India has 919 women for every 1,000 men, and in Pakistan and Bangladesh, the deficit of women is almost as severe. By contrast, the global sex ratio is nearly equal. A preference for sons, particularly eldest sons, has led families to abort female foetuses and reallocate resources from girls, decreasing their survival rate.
Digging deeper, I have investigated a statistical mystery: why India has higher rates of malnutrition and stunting than sub-Saharan Africa, despite its lower infant and maternal mortality and, generally, a higher level of economic development. I found that parents hoping for sons invest heavily in the first pregnancy, banking on a male heir, but subsequent children receive fewer resources in India than in Africa, driving the high rate of malnutrition.
A simplistic view of supply and demand may suggest that a shortage of girls will increase their social and economic value. But rather than granting more power to the fewer marriageable women, societies slanted in males’ favour respond by acquiring more brides, to expand the marriage market. A study in India showed that a low female-to-male ratio leads to lowering the age of brides, decreasing their educational attainment and participation in the workforce, and increasing the age gap between them and their husbands — all factors correlated with higher domestic violence.
Communities are starting to show the strain caused by such gender imbalance. In the Jind district in northern India, where there are only 871 women per 1,000 men, bachelors have formed a union to pressure political leaders to supply them with brides. It is also possible that more frequent honour killings in Pakistan and India are linked to an effort to keep women from marrying outside their caste or social group — an effort to ensure that men in the community have enough brides.
Can we devise policies and laws to begin to even the sex ratio? Might explicit cash incentives do the trick? A programme in Haryana state in northern India sought to even out the sex ratio and lower fertility at the same time by giving families cash pay-outs: the largest amount for having only one girl, smaller sums for having two girls or one boy. The payments did reduce the number of children — but through a larger reduction in the number of girls. Families who were encouraged to have one child more often opted for the boy, at least given this incentive structure.
As South Asian governments seek to improve living conditions by encouraging families of smaller size, parents may respond by sacrificing daughters. A resulting dearth of women puts those who remain at greater risk of violence, through channels such as lowered marriage age and increased trafficking of girls. Making a meaningful difference in women’s lives again requires a shift from short-term fixes to addressing underlying causes.
A foundational idea in development economics is inter-linkage: building a factory avails little if the roads leading to it are unusable and the natural resources that feed it are unmined. Successful development is even-handed, informed, and comprehensive. The same applies to enabling safe, empowered women. Improving women’s education and health, increasing their participation in the labour force, and lowering the number of children they have: these effects tend to snowball and can lead to a slow decrease in violence against women.
My own research has suggested a promising starting point along this path. In 1993, India passed a law mandating that one-third of seats on village councils (gram panchayats) be reserved for women. In many Indian states, the choice of councils to which the rule applied was, in effect, random, creating a “natural experiment” Moreover, the law rotated implementation of the reservation rule over election cycles: a village reserving seats for women in one election would not be required to do so in the next.
My co-authors and I found that women’s entry into these races and electability were persistently higher in villages that had previously been “reserved”. To understand the reasons, we surveyed villagers in depth. Men living in non-quota villages judged male leaders to be significantly more effective than female leaders, but this gap disappeared in villages that had been subjected to the quota provisions. Exposure to flesh-and-blood female leaders had radically altered male villagers’ perceptions of women’s effectiveness.
Separately, we showed that women leaders, newly raised to power through quotas, act as role models for parents and teenagers alike. In our survey of adolescents and their parents in 495 villages, we found that the gender gap in career aspirations closed by 25% for parents and 32% for adolescents in communities that had a female leader for two election cycles. Moreover, the gender gap in adolescent educational attainment was completely erased in these villages, and girls spent less time on household chores.
Another study revealed that women in states that implemented these quotas earlier on were more likely to start businesses. Finally, research by Harvard associate professor of business administration Lakshmi Iyer and others has demonstrated that the increase in female representation led to a 44% rise in the documented number of crimes against women — not from an increase in violence, but from more complete reporting (including a 23% rise in the number of rape cases filed). Exposing citizens to women in positions of power apparently conveys many benefits, including enhancements in women’s long-run safety. Strikingly, there is little evidence of backlash from men.
In The Global Gender Gap Report 2014, recently published by the World Economic Forum, India ranked 114th out of 142 nations surveyed, in terms of gender equality. That is down 13 places from last year, and far below other fast-growing countries such as China and Brazil. It did, however, rank 15th on the political empowerment index, which incorporates ratios of women in parliament and at the ministerial level, and whether a country has had a female head of state. How can India translate its success with female political participation into the advancement of women in other domains?
From economic analysis to effective policy
In June 2014, Anandiben Patel, Chief Minister of the western Indian state of Gujarat, announced that her government would enforce a 33% quota for women in the ranks of the state police. This seems just the sort of policy that I advocate in this article — but it is important to note that the law was already on the books. The problem is implementation.
This example, among others, illustrates that creating a safer society for women requires identifying the relevant policy actors and then providing them with incentives and information that align their interests and those of women. It is maddeningly common for governments to enact laws that sound good and appease the public, but are impossible to implement. Academic studies also often ignore or overlook implementation — or simply relegate action to the footnotes. Implementation challenges should be the subject of research and of careful monitoring and assessment that can determine when some policy goes awry.
Smart, effective responses to violence against women in South Asia (and elsewhere) depend on clear understanding of the social norms and the preferences of the involved parties. In designing responses, we must similarly be cognizant of the shifts in power our policies bring about — and address the consequences. Altering conventions such as inheritance laws, for example, creates winners and losers and threatens power; this may at the very least stir domestic conflict or even provoke a backlash (think of those village honour killings or the ever-younger brides discussed above). While such findings do not justify shying away from policies that enable gender equity, they cannot be ignored. As South Asian countries transfer power and wealth to women, they have the opportunity to test incentives and safety nets — to devise policies to ameliorate any perverse effects resulting from the shift in power. Again, short-run damage control can serve an essential long-run purpose: in the case of property rights, parents will see reason to invest more in their daughters, and these daughters will be better educated and empowered to take advantage of further economic opportunities.
In the meantime, we can take immediate steps in response to the rash of highly visible attacks on women:
- A recent United Nations report listed laws that encourage son preference and disempower women through inheritance restrictions; these should be repealed as deftly as possible, to minimise any backlash from those who lose privilege and power.
- The practice of reserving positions for women should spread across representative bodies in the region — and at higher echelons than the gram panchayat. India’s recent introduction of gender quotas for corporate boards is encouraging. To capitalise on it, companies should be incentivised to hire women at every level, on day shifts and night shifts. Unlike more sensitive reforms involving domestic life, the evidence suggests that such quota-based shifts in power can be affected with little risk of backlash.
- Policies should also incentivise women to form social networks and use them to enable women to overcome traditional barriers to their mobility. My colleagues and I recently examined the gender gap in microbusinesses. We offered business counselling and financial training to women; some were invited alone and others were instructed to bring a friend. The women who trained with friends were more likely take out loans, reported increased business activity and higher household income, and were less likely to report their occupation as housewife. This finding complements other studies that show that women enter the public sphere more successfully if they have a network to draw upon, and that the network itself can have stabilising effects on their lives.
- Finally, local leaders should be encouraged to change antiquated social norms. A council that oversees 42 villages in northern India recently made a public decision to allow inter-caste marriage — a controversial step that reversed centuries of tradition. It was a simple act, made as a direct effort to discourage honour killings that result from lack of brides. If other councils made similar announcements, it might cause marriage markets to open, son preference to weaken, and violence against women to decrease.
Graphic stories of violence will always grab the headlines, and the public demands quick answers and has a short memory. It is up to governments to make steady, evidence-based, interlinked efforts to build a society with greater equity for women, and greater rewards for all.