Economic independence to beat gender based violence

Gender-based violence is a common problem around the world. The majority of the victims are women, who often suffer in the hands of their closest relatives or romantic partners, for daring to defy the restrictions imposed on them. A woman's desire to seek education, formal employment, or to escape an abusive partner may trigger violence or even her death. The most extreme form of gender-based violence is the killing of a woman or a girl on account of her gender, known as femicide. To get a sense of the extent of the problem, consider that, according to the Interior Ministry's statistics, in Italy a femicide is committed every three days. Moreover, this number has increased significantly since the beginning of the global pandemic. One academic study estimated that, in economic terms, the costs of intimate-partner violence (a subcategory of gender-based violence) alone reach 5% of global GDP. In short, gender-based violence is a large problem that calls for academic attention and policy solutions.

In my research, I studied this problem in the context of Turkey. Unfortunately, Turkey is another country where gender-based violence is common. It is a patriarchal society where more than 40% of women report suffering physical or sexual violence at least once in their lifetime. Since 2018, every year around 400 women are murdered, often by their relatives or intimate partners. Although the total number of femicides is high in Turkey, it is not uniform across the country. If we look at the rate of femicides (i.e. the number of femicides per citizen), it is twice as high in the top 25% of provinces relative to the bottom 25%. I investigated which factors may explain this variation.

I used statistical methods to look at the correlation between the rate of femicides and a number of socio-economic factors: ethnic composition, religiosity, and civil war experience, divorce rate, gender equality in education, and the level of economic development.
The first three factors (religiosity, ethnic composition, and civil war experience) capture various facets of patriarchal culture. All else equal, we may expect more women to be killed in places that are more religious, more conservative, and with a stronger legacy of political violence.

The next two variables (divorce rate and gender equality in education) aim to capture women's willingness to challenge patriarchal norms and men's commands. All else equal, more femicides may take place when more women are getting divorced and more women are getting educated.

Finally, the level of economic development, i.e. GDP per capita, may matter because it determines whether women have opportunities to win their economic independence. In less developed places women are more likely to be unemployed or work in low-paying jobs. Without economic independence, women who suffer violence at home may be unable to escape their abusers and may become femicide victims when violence escalates to murder.
My analysis shows support for most of these hypotheses. First of all, culture, (more specifically, ethnic composition) matters. Places in Turkey with a more patriarchal culture experience more femicides. However, neither religiosity or a legacy of political violence seem to have an effect once we control for ethnic composition. Divorce rate and educational gender equality also matter. As these factors increase, so does the femicide rate. Finally, economic development matters. Fewer femicides are committed in places that are more developed.

One may ask, do these factors interact? For instance, is the deadly effect of divorce mitigated if a province is more developed and therefore women have better opportunities to escape their abusers? The answer is yes. The deadly effects of divorce rate and educational equality are mitigated in more developed provinces. To get a sense of the importance of these factors, consider a (hypothetical) average province. Using our model, we can calculate how the femicide rate will change as we vary different parameters. For instance, if its divorce rate were higher, how much will the number of femicides change? It turns out that divorce is always associated with more killings, but in a relatively rich province, the femicide rate doubles, whereas in a poor province it quadruples.
These findings broadly support the theoretical argument that gender-based violence results from a combination of women challenging their status quo and men using violence to suppress such challenges. However, economic development seems to be an important factor that facilitates defiant women's safety. Unfortunately, in the Turkish context, my analysis suggests that more women are murdered as more women challenge patriarchal norms. However, environments that allow women to gain their economic independence also give them more bodily safety.