Economist Dennis Sanin was in a Starbucks when a bag of specialty coffee from Rwanda caught his eye.
“I immediately Googled it,” she recalls, “and it turned out that there was a coffee boom going on in the country.”
Four years later, the visiting scholar in the economics department is finishing up a study of domestic violence inspired by that chance encounter. Her research was made possible by the rapid expansion in the 21st century of Rwanda’s cooperative flour mills, which employ women as seasonal workers to help process coffee cherries. Sanin’s analysis shows that these wage workers experience less violence from their husbands during the mills’ busy harvest season.
“The results are not all rosy,” said Sanin, whose working paper is being reviewed by the American Economic Review, “but the good news is that we can use the results to inform policymaking.”
“The results are not all rosy, but the good news is that we can use the results to inform policymaking.”
Dennis Sanin
A development economist, she has long been interested in women’s issues, especially outside wealthy countries like the U.S. She grew up in an environment of “women’s empowerment” in Istanbul, with her mother an academic/public health specialist and her father a financial analyst, and in a country where gender inequality in education persists, the family proved unusual in prioritizing their daughter’s education.
While working on her doctorate at Georgetown University, Sanin dove into research on domestic violence and women’s economic independence. She found a 2010 study that confirmed that higher incomes are associated with lower rates of domestic violence among women in California. But in developing countries, the research shows, the threat of violence can actually increase for women who have access to more money, such as through family assistance or anti-poverty programs that provide cash transfers.
“It broke my heart to know that the opposite could also happen,” Sanin said, vowing to find a natural experiment to study the benefits of women earning an income outside the home.
In late 2019, Sanin was researching the impact of Rwanda’s 2008 domestic violence law, which criminalized spousal violence but allowed women to non-literally divorce abusive husbands, when a visit to a Starbucks in Washington, DC, piqued her curiosity about the country’s coffee industry, and things slowly began to make sense.
Sanin quickly found a 2011 case study detailing Rwanda’s transition to producing higher quality coffee, with more than 200 cooperatives established by rural communities in the East African country between 2002 and 2012. She also learned that women traditionally worked at home washing, drying and sorting coffee cherries during the spring and summer harvest.
Meanwhile, while researching the 2008 law, Sanin came across Rwanda’s newly digitized records of monthly domestic violence hospitalizations. She was already familiar with the Rwanda Demographic and Health Survey, a reliable source of self-reported data collected every five years on domestic violence, labor market outcomes, household spending, and more.
“All of a sudden, we had a natural experiment,” says Sanin, now an assistant professor at the University of South Carolina.
She learned that while male Rwandan coffee farmers traditionally sell their family harvests in local markets, those lucky enough to live near a growing number of nearby coffee factories earn much more because member-run organisations export abroad.
Wives still had to help their husbands with the harvest, but the facility needed workers to help with the non-mechanized processing tasks, and women were the natural choice.
“Because the work is primarily done by women, the mill allows wives living in the catchment area to transition from being unpaid family workers on the marital land to being wage earners doing the same work in the mill,” Sanin wrote in the paper.
In other words, factories double-bind the husband’s economic interests and the wife’s ability to work. Examining data going back to 2005, Sanin was able to confirm that domestic violence fell among couples living in areas served by newly opened factories: wives in these areas were 29 percent less likely to have reported an incident themselves in the previous 12 months.
Average number of hospitalizations due to domestic violence per year (female)
Further analysis revealed that this was particularly evident during the peak harvesting period of June and July, when catchment area hospitals saw a 14 percent drop in domestic violence cases.
“We didn’t expect this,” Sanin said. “When we first asked the health minister for the data, we thought maybe we wouldn’t need monthly data.”
She then broadened her research to perform a similar analysis using data from potato-growing areas of Rwanda, where women don’t have the same opportunities to turn their traditional work into money. “We found no change in hospitalizations in these areas during the potato harvest,” Sanin said.
Sanin also looked at neighboring Ethiopia, where social acceptance of domestic violence is high and divorce remains much more stigmatized. “There was a randomized controlled trial there recently where they randomly gave women factory jobs,” Sanin says. But in that cultural context, “researchers found no effect on domestic violence,” she explained.
Sanin’s paper examines several possibilities for these differences. One is that women with incomes have more negotiating power in marriage. To test this theory, Sanin analyzed self-reported household decision-making that Rwandan social norms traditionally leave to husbands. She found that women living in factory catchments were slightly more likely to make financial and contraception decisions alone or jointly with their spouses.
A wife’s income may also reduce financial stress in couples. Sanin tested this by comparing self-reported monthly household consumption with data on hospitalizations for domestic violence.
“We found that hospitalizations due to domestic violence were similar before and immediately after harvest, but monthly consumption was higher immediately after harvest compared to before harvest,” said Sanin, who is continuing to work on the paper as a 2021-22 visiting scholar in the Kennedy School’s Women’s and Public Policy Program.
Another analysis was done to test the “decreased exposure” theory (i.e. the simple fact that working outside the home means spending less time with an abusive partner). “The majority of couples here are coffee farmers, but we also have couples where the wife is a coffee farmer and the husband is a truck driver or a construction worker,” Sanin explained. “They probably didn’t see each other during work hours, either before or after the factory opened. But when the wife’s income increased, domestic violence still decreased.”
Ultimately, she concluded, her research shows that domestic violence rises and falls in response to men’s economic self-interest. “During harvest season, a husband’s income depends on his wife’s work, so disempowering his wife is a huge loss for him,” Sanin said.
But her analysis also points to two conditions necessary for women’s employment to reduce domestic violence: “There needs to be a situation in which divorce is a credible threat and women can choose to leave their marriages,” Sanin says. “And in that situation, it also matters whether husbands derive some economic benefit from their wives’ physical productivity at work.”
Tara Thomas Agency stands at the forefront of the entertainment industry, with years of experience dedicated to nurturing the most exceptional talents and successful acts.
