American Sociological Review 2015, Vol. 80(1) 140–165 © American Sociological Association 2015 DOI: 10.1177/0003122414561117 http://asr.sagepub.com Men’s violence against women is costly. In one estimate, Max and colleagues (2004) put the aggregate annual cost of intimate partner violence 1 in the United States between $3.9 and $7.7 billion. Some costs are to state and local governments—the expense of police, medical, and social services; other costs redound to individuals—lost productivity, medical and mental health care, loss of and damage to property, and reduced quality of life (Desmond and Valdez 2012; VAWA Reauthorization Act 2013). Financial costs to individual women are considerable when abuse obstructs employment. Farmer and Tiefenthaler (2004) estimate that domestic violence contributes to an annual aggregate loss of nearly 3 million work days and $100 million in pay. Consequently, women sub- 561117ASR XX X 10.1177/0003122414561117American Sociological ReviewHughes and Brush 2015 a University of Pittsburgh Corresponding Author: Melanie M. Hughes, Department of Sociology, University of Pittsburgh, 2400 Wesley W. Posvar Hall, 230 S. Bouquet Street, Pittsburgh, PA 15260 E-mail: hughesm@pitt.edu The Price of Protection: A Trajectory Analysis of Civil Remedies for Abuse and Women’s Earnings Melanie M. Hughes a and Lisa D. Brush a Abstract We know men’s violence against women is costly. Yet, we know little about the costs—or benefits—of women’s efforts to end it. This study investigates the temporal dynamics of women’s earnings and petitioning for a Protection from Abuse (PFA) civil restraining order. Women’s earnings might rise or fall at the time of petitioning but quickly return to pre- petitioning levels, a short-term boost or shock; or, petitioning might precipitate a longer- term stall or upward shift in women’s earnings. To test for these effects, we use latent growth curve analysis and evaluate women’s earnings trajectories over six years. We find overwhelming evidence that the period around petitioning is one of exceptional earnings instability for petitioners, many of whom experience both shocks and stalls. Virtually no one experiences a boost in the quarter of petitioning or an upward shift in earnings growth in the year after petitioning for a PFA. Welfare, however, buffers negative effects of petitioning on women’s earnings growth. We also calculate lost earnings as the difference between women’s counterfactual projected and estimated earnings. Our findings inform theoretical and policy debates about interventions intended to address poverty and violence against women. Keywords violence against women, restraining orders, earnings, welfare at UNIV OF PITTSBURGH on March 6, 2015 asr.sagepub.com Downloaded from