The unintended effects of social protections on intra-household instability: insights from meta-analysis and the UN Joint Programme in Ethiopia

Social protections, particularly social assistance programmes such as cash transfers (CTs) are considered the most important tools to reduce poverty and empower women. They are prominently linked in the Sustainable Development Goals (SDGs), particularly in SDG 1 and SDG 5, which aim to end poverty, and achieve gender equality and empower all women and girls. The role of those programmes in reducing poverty and enhancing women's empowerment has been recognized by many empirical examinations. However, their unintended effect on family instability, intimate partner violence (IPV) in particular, remains overlooked. Even the existing few empirical examinations on the impact of those programmes on IPV provide mixed and inconclusive results. By employing a meta-regression analysis (MRA) method, we reconcile the conflicting findings of the existing empirical studies. Using 394 reported estimates from 15 studies published until August 2024 on the impact of Unconditional cash transfer (UCT) on IPV, we estimated the genuine underlying effect of UCT on IPV and examined if there is publication bias. The multivariate MRA also shows that UCT helps to reduce IPV. Specifically, it shows that after controlling for publication bias and possible sources of heterogeneity, UCT was found to reduce IPV on average by 85.6%. However, we didn’t find a significant publication bias in the issue. In addition, most theories and empirical studies argue that the main mechanism through which CTs impact IPV is women's empowerment (WE). However, none of them examined the joint impact of CT and WE on IPV. Hence, besides reconciling the existing conflicting findings and examining the presence of publication bias, we find it necessary to examine the joint effect of CT and WE on IPV. To do so, we used the World Bank data on the Rural Women’s Economic Empowerment (UNJP-RWEE) programme implemented to empower women in Ethiopia. In harmony with the meta-result, we found that UCT helps to reduce attitudes towards IPV, particularly physical domestic violence (PDV). However, women empowerment was found to increase attitudes towards PDV. Furthermore, UCT and women empowerment jointly increase attitudes towards hitting or beating a wife for neglecting children and for arguing with the husband by 38.9% and 45.7%, respectively. Overall, the result suggests that the increasing effect of IPV is primarily driven by women's empowerment. The results of this study provide valuable insights for policymakers and development agents working to enhance household welfare by enhancing women’s empowerment should reconsider the possible trade-off between women's empowerment and intra-household violence. Hence, they should not only focus on empowering women but also enhancing the awareness of male spouses and the community related to the existing masculinity hegemony and gender roles.