Combatting domestic violence
A new economic model helps us understand the economic consequences of gender-based violence
‘Given the scourge of Gender-Based violence in the country, which is declared a national address (sic), the government is implementing the Emergency Response Action Plan on Gender-Based Violence and Femicide, which was announced by President Cyril in September 2019.’
So begins the press release by the South African government to announce the 16 Days of Activism for No Violence Against Women and Children, a United Nations campaign held annually from 25 November to 10 December. It continued: ‘During the 16 Days period, Government together with civil society and the private sector will host a series of community and sector dialogues and activities to foster a collaborative effort in dealing with GBVF.’
The sloppy writing does a disservice to the severity of domestic abuse globally and in South Africa. The World Health Organisation estimates that roughly one in three women globally have experienced physical or sexual violence at some point in their lives. Most of these cases involve intimate partner violence. Almost 27% of women aged 15 to 49 who have been in relationships report experiencing physical or sexual abuse by their partner.
South African women experience some of the highest rates globally. In 2023, 42,631 women reported being raped in South Africa, the most extreme form of sexual violence, or 70 women for every 100,000 people. This ratio is the third-highest in the world, after Botswana and Lesotho. What is more, most of these women reported being raped by someone they knew: two-thirds were assaulted inside either their own home or the home of the perpetrator. It is clear that addressing gender-based violence must start with confronting violence within intimate relationships.
While the scale of the problem has certainly attracted more activist attention in recent years – and, to be fair, from the South African government too – it is also true that there are still many unanswered questions about both the causes and consequences of domestic violence – and the policies that are most likely to result in positive change. One reason for that is data. The numbers I report above are of women who reported rapes and assaults; the true numbers are likely much higher.
Another reason is the lack of rigorous models to help us think through the ways in which abusive men exert control over their partners, both economically and emotionally. Without clear models, it is difficult to understand the full range of tactics used by abusers, from direct violence to more subtle forms of coercive control, such as limiting access to finances or restricting employment opportunities. This gap hinders the development of effective policies, leading to reactive, fragmented solutions that often fail to address both the immediate harm and long-term impacts of abuse.
That was the motivation for a new study published in the Quarterly Journal of Economics. The study has two strengths: Access to incredibly rich Finnish data, which the authors then use to build an economic model with clear policy conclusions. At first glance, the authors investigate how cohabiting with an abusive partner affects women’s economic outcomes, particularly in terms of employment and earnings. But drill down, and the results allow the authors to build a model that shows how abusive partners use economic suppression to limit women’s outside options, making it harder for them to leave the relationship, a finding with clear policy implications.
First, though, the data. The study links police reports in Finland to administrative records on employment, cohabitation, and demographic data. Unique identifiers for both perpetrators and victims allow the authors to determine if they were living together at the time of the offence.
The authors first estimate the impact of cohabiting with an abusive partner on women’s labour market outcomes. In the five years leading up to the relationship, women’s employment and earnings increased by 20% and 30%, respectively. However, once the abusive relationship began, their economic conditions worsened, with employment dropping by 4% and earnings by 6% in the year following cohabitation.
Of course, there might be many things that explain such a correlation. To get at a causal explanation, the authors compare the outcomes of victims with those of similar women who begin cohabiting with similar partners. They find that women in abusive relationships experience employment rates that are 6.7 percentage points lower than their matched controls in the year after cohabitation, and this decline lasts for at least five years. Victims also see their earnings drop by an average of €1600 per year over five years, a 12% loss compared to their earnings before cohabitation.
But perhaps there is reverse causality: women that lose their job might make bad relationship choices. To test this, the authors show that the fall in earnings is not due to changes in the firm, in co-worker outcomes, or local labour market shocks. This suggests that the decline in women’s economic outcomes is due to the abusive relationship, not external economic shocks leading to poor relationship choices. Recognising that cohabiting with an abusive man involves other partner characteristics that may influence female labour supply, they also rule out factors such as fertility changes or differing gender norms as causes.
Another concern might be that abusive men are somehow different to non-abusive men, and that it is the relationship, rather than the abuse, that cause these negative economic outcomes. To test this, the authors compare victims’ outcomes in non-abusive relationships as a counterfactual. They find that victims’ employment is 4 percentage points lower in abusive relationships than in non-abusive ones, indicating that the economic decline is specific to abusive relationships.
They also examine whether men reported for physical abuse negatively affect the economic outcomes of all their partners. Women who cohabit with known abusers, even without reporting violence, show significant declines in employment. Women who live with men reported for violence against another woman, but not themselves, also experience reduced employment, while no such effect is found for men who committed violence against other men. These findings suggest that men prone to abusing women consistently harm their partners’ economic outcomes.
What to do with all these findings? The authors use it to construct a new economic model that aims to explain how abusive relationships affect women's job prospects and earnings. In short, they assume two types of men: abusive and non-abusive. Women cannot tell which type their partner is at the start of cohabitation. Abusive men harm their partners’ economic independence by limiting their ability to work or sabotaging their job opportunities. This makes it harder for women to leave the relationship.
The model suggests that abusive men use two main strategies: physical violence and coercive control. Coercive control involves actions like limiting access to money or preventing women from working. By doing this, the abusive partner reduces the woman’s options to leave, as she becomes financially dependent. This economic suppression can make it difficult for women to escape the abusive situation, even when the abuse becomes more serious.
One key idea is that women with intermediate economic independence before cohabitation are most at risk. Those with very low or very high earnings are less likely to experience severe economic damage. That is because women with very low earnings may already be financially dependent and thus have little to lose, so the impact of economic suppression is less noticeable, while women with very high earnings tend to have more resources and support networks, making it harder for abusive partners to sabotage their financial independence.
The model also shows that women who gain more economic independence through outside support like shelters are more likely to leave abusive relationships. And here the policy implications become obvious: First, expanding access to shelters that provide financial support is crucial to help women gain the security needed to escape abusive relationships. Second, policies should focus on improving women’s access to employment and skills training, enabling them to build or regain their financial independence. Finally, legal frameworks must address not only physical violence but also coercive control, recognising economic abuse as a serious issue and offering stronger protection against financial manipulation.
Let’s hope that the government’s ‘Emergency Response Action Plan on Gender-Based Violence and Femicide’ can implement the lessons from studies like these; without real action, it’s just another promise that won’t protect the women who need it most.
An edited version of this article was published on News24. Support more such writing by signing up for a paid subscription. The images were created with Midjourney v6.