Association for Public Policy Analysis & Management

Economic downturns are associated with an increase in child abuse and neglect (Lindo et al., 2013; Wood et al., 2015). Economic downturns are also linked to increases in intimate partner violence (IPV) (Schneider et al., 2016), which can further increase child abuse risk (Zolotor et al., 2007). However, not all economic downturns impact family violence in the same way in all communities and for all families (Lindo et al., 2013). Moreover, the effects of economic downturns are not limited to only those families that experienced job loss or income loss, as is often assumed (Gassman-Pines et al., 2015). Yet, the primary policy responses to economic downturns in the U.S., Unemployment Insurance (UI) and Trade Adjustment Assistance (TAA), provide benefits only to workers who have lost jobs and may leave many families vulnerable. Effective family violence prevention strategies must fill the gaps left by these traditional policy responses to economic downturns. The purpose of this paper is to develop a more complete understanding of what types of economic downturns place which families at risk of family violence.

 

Data

This paper uses child maltreatment data from the National Child Abuse and Neglect Data System (NCANDS) and IPV data from the National Incident-Based Reporting System (NIBRS). NCANDS and NIBRS data are available for twenty-nine states for the years 2004 to 2014. Both NCANDS and NIBRS provide incident-level information, including the date and nature of the incident and the victims’ and perpetrators’ demographic information. The NCANDS and NIBRS data are linked with state-level monthly data on community-wide job losses from the U.S. Department of Labor (DOL) and state-level quarterly data on men’s and women’s job losses from the U.S. Census Bureau to measure community-level economic conditions. Finally, these data are also linked with annual state UI and TAA expenditures data from the U.S. DOL.

Analytical Strategy

The analyses will identify which families are most at risk of family violence following job losses by examining racial differences and differences by children’s ages. Second, this paper will assess the differential effects of men’s and women’s job losses, as well as differential effects by the gender of the perpetrator. Finally, this paper will test to what degree the two existing primary policy responses, UI and TAA, buffer these effects. By identifying whether traditional policy interventions buffer this effect, this project also identifies remaining policy gaps.

To assess the association between job losses and family violence, this paper will use fixed effects analyses to control for any observed and unobserved factors that might impact both the economy and family violence. Specifically, the model will include year and month fixed effects to isolate the effect of job losses that were aberrations relative to the overall economy in the country in a given month and year. It will also include state fixed effects to control for stable differences in the economy and family violence between states. The study sample size will be 3,480 state-months.