The Relationship between Women's Economic Resources, Decision-Making Power, and IPV Using Structural Equation

Background: Can women’s access to resources increase their decision-making power and risk of intimate partner violence (IPV) at the same time? Theory and evidence seem to be divided. Economic bargaining theory and evidence indicate that women with more access to resources increase their decision-making power and decrease their IPV risk (Aizer, 2010; Doss, 2013; Farmer & Tiefenthaler, 1997). In contrast, theory and evidence supporting male backlash and gender inconsistency theories predict increases in IPV risk (e.g., Atkinson et al., 2005; Macmillan & Gartner, 1999). Furthermore, several studies considering both outcomes found that women’s access to resources increases both decision making power and IPV risk(Ahinkorah et al., 2018), partially supporting both theories. This inconsistency might be due to endogeneity issues between two outcome variables (autonomy and IPV), which are difficult to address with methods such as regressions. To examine this relationship, this study used structural equations and hypothesizes that women’s economic resources relative to their partners has a direct positive effect on autonomy and an indirect negative effect on IPV.

Methods: Using Colombian Demographic and Health Surveys, a nationally representative sample of girls and women ages 15 to 49, this study used two endogenous variables (decision-making and IPV) and one exogeneous variable (economic resources). Decision-making was captured by four observed indicator variables about who usually makes household decisions (e.g., use of family earnings). IPV was measured by eight observed indicator variables (e.g., hit, slapped, punched). Economics resources was captured with five observed variables measuring women’s resources relative to her partner’s (e.g., owning assets alone or jointly with partner) and earnings (whether she earned less, the same, or more than her partner), education, and household wealth quintile. Analysis was performed using Stata and Mplus.

Results: All path coefficients were statistically significant. To improve model fit, I followed modification indices using Mplus and allowed the correlation of some measurement errors under the endogenous latent variable violence. The final model indicated an excellent fit. CFI = 0.978. RMSEA = 0.022 with a 90% CI [0.020 ,0.024]. The results confirm the conceptual model hypothesizing that that women’s resources relative to their partner’s increases their autonomy and decreases their risk of IPV. Other things being equal, for each increase in the standard deviation of women’s economic resources, autonomy increased by .341 units (p < .001) and IPV decreases by .407 units (p < .001).

Conclusions: This study found that women’s relative economic resources to their partners have a direct effect on autonomy and is mediated by IPV. Findings indicate that women’s access to resources increases their household decision-making power and is also protective of IPV, suggesting that these two outcomes should be considered together. Since this was cross-sectional data, it is difficult to establish order of events regarding economic resources and IPV or autonomy. However, the reverse path was tested, and the model was under-identified. Future research should test the mechanism underlying the relation between women’s economic resources, decision-making, and IPV.