Employment as a Pathway to Reducing Domestic Violence in India: Insights from NFHS-5

Domestic violence (DV) is a major public health and gender equality concern. While women’s employment is often assumed to reduce vulnerability, evidence remains mixed due to issues of endogeneity and selection bias. This study examines the impact of women’s employment on DV in India, emphasizing the mediating role of empowerment. Using data from the National Family Health Survey (NFHS-5, 2019–2020), we applied econometric and causal inference methods. Mean differences in DV by employment were using t-tests. To address selection bias, propensity score matching (PSM) was conducted. Endogeneity was tested and corrected using two-stage least squares (2SLS) for continuous DV and two-stage ordered probit (2SOP) for severity, with heterogeneity across subgroups. Mediation was assessed by two-stage residual inclusion (2SRI), with empowerment (PCA based) as mediator. DV prevalence was higher among employed (36.8%) than unemployed women (27.2%); PSM confirmed this with an 8.1% absolute standard ratio. After correcting for endogeneity and covariates, employment reduced DV scores (β = -0.171, p < 0.01) and severity (β = -0.081, p < 0.05). Stronger protective effects were observed for younger (β = -0.719, p < 0.001), rural (β = -0.148, p < 0.05), and less-educated women (β = -0.269, p < 0.05). Mediation analysis showed direct (β = -0.378, p < 0.001) and indirect (β = -0.028, p < 0.001) protective effects through empowerment. Women’s employment lowers domestic violence, strengthened by empowerment. Results endorse causal inference perspectives, dispute male backlash theory, and stress policies promoting employment, empowerment, and gender equality.