Cash: A simple remedy for domestic violence?

Blog Inclusive Growth

In South Asia, 42% of women experience intimate parent violence. In Sub-Saharan Africa, that number ranges from 30% up to 66%. Research from Kenya shows that simple direct cash transfers to households can significantly reduce domestic violence

In South Asia, 42% of women experience intimate partner violence. In Sub-Saharan Africa, that number ranges from 30% up to 66%. Research from Kenya shows that simple direct cash transfers to households can significantly reduce domestic violence

An estimated 30% of women worldwide will experience physical and/or sexual violence by an intimate partner in her lifetime. Domestic violence is the most common type of violence that women experience. In addition to being the leading cause of homicide of women, domestic violence is associated with other adverse health outcomes, including depression, suicidal behaviors, and HIV infection.

The cure for domestic violence is as elusive as its causes are complex; scientists, advocates, and policymakers continue to argue about how to best address the problem. Is a change in individual attitudes required? Are social norms to blame? Is substance abuse the key problem? The potential causes are numerous, and the answers few

However, one clue may lie in the fact that the problem is particularly pervasive in developing countries. In Sub-Saharan Africa, the proportion of women experiencing intimate partner violence ranges from 30% up to 66%, and in South Asia, that proportion is 42% on average. These numbers contrast with significantly lower rates of domestic violence in North America and Western Europe, 21% and 19% respectively. These statistics raise the possibility that domestic violence may not in the least measure be a problem of poverty. Of course the numbers we cite above are merely correlations: it might be that domestic violence is caused by poverty, but it might also be the other way around (e.g. abused individuals may be less productive or unable to work altogether). Alternatively, it may be that population-level factors characteristic of emerging economies, such as discriminatory social norms and laws, are driving the observed association with intimate partner violence.

In a recent study in Kenya, we asked whether the “economic hypothesis” of domestic violence holds water. It makes the clear prediction that when poverty is alleviated, domestic violence should be reduced. We partnered with the NGO GiveDirectly, whose mission is to send unconditional cash transfers to poor households in Kenya. As their name suggests, these transfers are pure “helicopter drops of cash” that households receive simply because they are poor; the transfers do not have to be repaid and can be used in whatever manner the recipients deem appropriate.

In our study, households in Western Kenya were randomly chosen to receive transfers from the NGO GiveDirectly based on a single, readily observable measure of poverty—living in a thatched roof house. Three features of the transfers that households received were randomised as follows: whether the household received a large grant of $1,520 or a small grant of $404; whether the payment came as a single lump sum or in monthly installments; and whether it went to the husband or the wife in the household. The fact that the recipient and control households were randomly chosen enables us to attribute any changes in subsequent outcomes to the cash transfer.

At baseline, the burden of domestic violence in our recipient households was substantial. Approximately one-third of women reported having been physically abused by their husbands in the preceding six months and one-tenth sexually abused.

A year after the cash transfer had been initiated, we found that rates of physical and sexual violence between intimate partners had fallen dramatically in the treatment households relative to the control households, with a somewhat larger effect when the transfer went to the woman. For instance, treatment households experienced a 54% reduction in reports of the husband slapping the wife in the preceding 6 months when females received the transfer, and a 42% reduction when men received the transfer.  Moreover, treatment households experienced a 73% reduction in reports of beatings in female-recipient households, and an 82% reduction in male-recipient households. In female-recipient households, reports of rape dropped by 78%, from being reported by 9% of women to being reported by 2% of women. Interestingly, we also found large effects of transfers on neighbouring households: non-recipient households in villages where other households got transfers showed much lower rates of domestic violence than non-recipient households in villages where no transfers were sent.

Thus, it appears that cash transfers can successfully reduce domestic violence. What might be the mechanism through which this effect occurs?

Bargaining power

One possible explanation for the reduction in domestic violence through cash transfers is that they may have strengthened women’s bargaining power in the relationship. In line with this hypothesis, we find a small increase in threats and emotional abuse of wives by their husbands, which might be expressions of a bargaining process in the household over how the money gets spent. However, the bargaining power story is not consistent with the fact that domestic violence went down even when the husband received the transfer.

Social norms

This bargaining power hypothesis furthermore does not account for the large spillover effects. Physical and sexual violence both fell in control households neighbouring treatment households compared to control households in villages where no transfers were delivered. This suggests that, as rates of violence were falling in their own households, individuals who received transfers may have instigated a change in social norms in their villages.

Psychological wellbeing

An additional explanation for both the treatment and spillover findings may relate to the large psychological impacts of the intervention. Among treatment households, depression and stress fell significantly, while happiness and life satisfaction rose. Interestingly, when females received the cash transfers, males and females alike showed greater reductions in cortisol levels, diminished levels of worries, and heightened self-esteem compared to when males received the transfers. These improvements in psychological wellbeing may have led to fewer violent reactions to intra-marital conflicts.  Notably, both treatment and spillover households experienced an increase in optimism; thus, it may be that the existence of cash transfers – whether one received them or not – provided a reason to hope for a better future, starting with better relationships in the household.

Financial strain

Lastly, cash transfers mitigated a primary source of conflict between domestic partners – financial strain.  We recently began to conduct qualitative interviews in our recipient households, and initial findings suggest that cash empowered men to more effectively fulfill their gendered role as “provider”, as they could better furnish the funds their wives requested to care for their children and themselves. Supporting this hypothesised mechanism is the counterintuitive finding that husbands’ psychological wellbeing improved more when their wives received the transfer compared to when they themselves received the transfer. It is possible that cash transfers enabled households to purchase basic needs so that women no longer needed to ask their husbands for money, which contributed to fewer feelings of inadequacy (and annoyance) by men and, in turn, fewer conflicts. Indeed, when households received monthly transfers, spending on basic needs such as food jumped and the incidence of physical and sexual violence dropped significantly.

Policy implications          

It remains to be seen which of these hypotheses is the best account of the effects of cash transfers on domestic violence. It is also as yet unclear whether some of these findings could be explained by the fact that the survey was based on self-reports, and that treatment households were surveyed more frequently than control households. But the tantalising possibility suggested by these early results is that cash may not only effectively reduce poverty, but also lead to healthier, happier relationships within households, even without directly incentivising behavior change. If these results stand the test of time and further study, cash transfers may become a useful policy tool for female empowerment.