Crime declines where marriage is more prevalent and when individuals enter marriage. To explain this link between crime and household structure, I examine how forming a union transforms individual conduct by bringing previously unconstrained behaviors into the domain of household bargaining, thereby creating social constraints on behavior. I develop a simple household bargaining model that formalizes this process: spousal disapproval creates incentives for behavioral change, and the ability to enforce such change depends on bargaining power within the relationship. The model predicts that behavioral change depends on the interaction between individuals' baseline risk levels and their partners' bargaining power, producing the largest reductions in offending among high-risk individuals with powerful partners. Using panel data from the National Longitudinal Survey of Youth 1997 and proxying partner power with relative earning potential, I find evidence consistent with this prediction, particularly for men's offending. Importantly, the effects are as strong—and, in some specifications, stronger—for cohabitation as for marriage, supporting a bargaining mechanism over selection or identity-based explanations.
An improvement in women’s labor market opportunities can increase their bargaining power within the household. But such female empowerment can produce puzzling results. Some women improve their situation through household bargaining, while others experience increased household conflict (backlash). To reconcile these competing explanations, I develop a model inspired by Hirshman’s discussion of exit, voice and loyalty. In the model, empowerment enables previously suppressed dissatisfaction to surface in household bargaining as newfound voice. When a woman exercises her voice, two outcomes are possible. If her partner makes concessions, she can negotiate better terms. However, if her exit threat lacks sufficient credibility, her voice may trigger escalation. The relative strength of voice vs. exit credibility shapes household conflict dynamics. I show that empowerment produces a non-monotonic relationship: empowerment first increases conflict, then reduces it. This framework provides a cautionary interpretation of existing evidence: apparent household harmony may mask suppressed dissatisfaction, while increased household conflict can signal women's ability to challenge the status quo in their favor.