Abstract
The empirical research on the relationship between mortgage foreclosures and crime continues to evolve, and, with few exceptions, the results generally show that there is a relationship between community measures of mortgage foreclosure and crime levels. Lacking in this literature are studies that considered how foreclosures may impact domestic violence. For many families, home mortgages have significant meaning as an indicator of financial stability and represent a long-term commitment to a community through home ownership and the accompanying social and financial investments. Families who are threatened with the loss of their home to foreclosure experience a decrease in their financial status, and the stress triggered by the crisis may place them at greater risk for family violence. In this study, we examine the relationship between mortgage foreclosures and family violence. Using longitudinal panel data for Massachusetts cities and towns, our study focuses specifically on the years 2005–2009, the period that includes the Great Recession. We find that after controlling for other community indicators of economic health, higher levels of monthly mortgage foreclosures lead to higher levels of domestic violence.
Introduction
The rise in home mortgage foreclosures during the mid-2000s was a harbinger of the “Great Recession” that hit the United States from 2007 to 2009. According to Bocian, Li, and Ernst (2010), during this time period, an estimated 2.5 million residential foreclosures were completed, many of which were owner-occupied housing. This trend alarmed the public and policy makers, because it signified that many households across the country were dealing with financial crises. Journalistic accounts of the foreclosure crisis began calling attention to the negative impact on local communities (Mummolo & Brubakerr, 2008; Johnson, 2008) and families (Armour, 2008).
Concerns about the effects of residential displacement on local communities left in the wake of property foreclosures stimulated social scientists’ interest in the relationship between foreclosures and criminal behavior. Over the past several years, a growing number of studies have examined the relationship between foreclosures and crime. According to a review by Wolff, Cochran, and Baumer (2014), these studies theorize that foreclosures affect crime because they accelerate indicators of social disorganization and disorder that lead to crime. Foreclosures affect community crime by reducing the capacity for informal social control both socially, due to rapid residential changes, and physically, by creating crime opportunities that accompany growing signs of disorder when properties are abandoned.
The empirical research on the relationship between mortgage foreclosures and crime continues to evolve, and, with few exceptions, the results generally show that there is a relationship between community measures of mortgage foreclosure and crime levels. Foreclosures have been found to affect crime at varying levels of geography, ranging from the census block face level (Ellen, Lacoe, & Sharygin, 2011) to census tract and city (Baumer, Wolff, & Arnio, 2012; Lersch, Sellers, & Cromwell, 2014), and county levels (Goodstein & Lee, 2010, but see Kirk & Hyra, 2012). However, there has been some inconsistency across studies in the types of crime affected by foreclosures. For example, Baumer, Wolff, and Arnio (2012) found foreclosures are related to burglary rates at the census tract level but not to robbery rates, while Arnio, Baumer, and Wolff (2012) found foreclosures affect burglary and robbery at the county level. Goodstein and Lee (2010) found a relationship between foreclosures and burglary, larceny, and aggravated assault. Other studies combine crimes into violent and/or property crimes, with mixed findings. Ellen, Lacoe, and Sharygin (2011) and Immergluck and Smith (2006a) found an effect of foreclosures on violent crime but not on property crime. Yet Baumer et al. (2012) found a significant positive effect on both violent and property crime. Still others report that any effects on property and violent crime are likely to be short term (Katz, Wallace, & Hedberg, 2013).
Although the impact of foreclosures on some specific types of violent crime has been investigated, only one published study to date has considered how they may affect levels of domestic violence. Analyzing Census Tract data in Tampa for 2008, Lersch, Sellers, and Cromwell (2014) found that foreclosure rates were positively associated with domestic dispute calls for police service, even after controlling for neighborhood demographic characteristics. While causality was unable to be determined due to the cross-sectional nature of the study, these results suggest that foreclosures have some connection with rates of domestic violence. Lersch and colleagues call on future researchers to “more carefully consider the temporal relationship between foreclosures and domestic disturbances” (p. 11). This study seeks to answer that call.
Even without confirming causality, the findings of Lersch and colleagues (2014) are not entirely surprising. For many families, home mortgages have significant meaning as an indicator of financial stability and represent a long-term commitment to a community through home ownership and the accompanying social and financial investments. Families who are threatened with the loss of their home to foreclosure experience a decrease in their financial status, and the stress triggered by the crisis may place them at greater risk for family violence. Financial stress has been reported to be the cause of an increase in calls to domestic violence hotlines. For example, in September 2008, The National Domestic Violence Hotline (NDVH) reported a 21% increase in calls compared to 2007. Additionally, among 7,868 callers who agreed to participate in the NDVH’s study in late 2008, 54% noted a change in their household’s financial situation over the last year and 64% reported abusive behavior had also increased (Fielding, 2009).
In this study, we examine the causal relationship between mortgage foreclosures and family violence. Using longitudinal panel data for Massachusetts cities and towns, our study focuses specifically on monthly foreclosures for the years 2005–2009, the period that includes the Great Recession (Russo, Mischow, & Schinski, 2015). In the sections that follow, we first review the empirical literature that illustrates the unique role of community economic context on domestic violence. Second, we explain the relevance of using foreclosures as a unique indicator of economic stress. We then describe the data and analytic methods used for our study followed by the results and discussion of findings.
The Community Context of Economic Stress and Domestic Violence
Researchers have been interested in the relationship between economic stress and family violence for decades. Although it is well documented that domestic violence occurs across class lines (Fagan & Browne, 1994), those facing greater economic hardship are especially vulnerable (Renzetti, 2009). Economic strain may create heightened risk of interpersonal violence for individuals, for couples, and even other family members. According to MacMillan and Gartner (1999), financial hardship may generate feelings of stress and frustration that become expressed in acts of physical aggression and controlling behavior. This may be particularly acute during economic downturns, even for those who are otherwise able to keep negative behavior under control (Schneider, Harknett, & McLanahan, 2014). Others describe the emergence of violent and controlling behavior in relationships as a response to loss of control in one’s economic domain (Stets, 1995). Among couples, violence was reported to increase if a woman was partnered with an unemployed male (MacMillan & Gartner, 1999).
Unemployment and income are key indicators of economic strain examined in the literature. At the individual level, employment has been found to deter or prevent violence, functioning as a protective factor, while unemployment is considered to be a risk factor (Bowlus & Sietz, 2006; Mistry, Low, Benner, & Chien, 2008). Using data from national surveys, Fagan (1993) found a relationship between income and family violence that varied by race and geography. Spousal assaults were more likely to be reported in low-income groups living in central cities and were highest among the poorest African American families. In 1998, Miles-Doan empirically investigated the relationship between community economic structure and family violence reported to the police. Her study reported an association between a composite measure of community resource deprivation and rates of intimate partner violence. These findings indicate that community structural and economic dimensions are pertinent to the study of family violence.
The contextual effects of community economic disadvantage were examined by a more recent body of literature. Drawing on social disorganization theory to explain the influence that neighborhood economic structure may have on intimate partner violence, these studies often employ composite measures of neighborhood disadvantage created from indicators of economic stress that include Census measures of poverty, family structure, unemployment, public assistance, and race. These measures are then appended to area-identified survey data. These studies show that after controlling for household and individual economic circumstances, community-level measures of economic disadvantage were associated with an increased likelihood of self-reported couple violence (Benson & Fox, 2004; Benson, Fox, DeMaris, & VanWyk, 2003; Fox & Benson, 2006; but see Lauritsen & Schaum, 2004).
Interaction effects between neighborhood disadvantage and individual and household factors have also been reportedly found to explain couple violence. VanWyk, Benson, Fox, and DeMaris (2003) report an interaction effect between neighborhood disadvantage and low levels of social support for women which increases the likelihood of intimate partner violence. This body of work emphasizes the importance of the community context in explaining intimate partner violence. The studies are limited, however, by their reliance on static measures of community characteristics and are thus unable to consider how changes in economic conditions over time may be related to domestic violence.
A causal model would be more supportable if changes over time in economic stressors predicted changes in levels of domestic violence. This approach was taken in two recent studies. Lauristen, Rezey, and Heimer (2013) examined national trends in intimate partner violence reported to the National Crime Victimization Survey (NCVS) from 1980 through 2011 and found no relationship between unemployment and intimate partner violence rates. Petersen (2011) also used the NCVS to examine the relationship between employment and intimate partner violence trends from 1993 to 2005. He found that during the 1990s, a time of economic growth, there was a significant positive relationship between unemployment and intimate partner violence, but no relationship during the recession period from 2001 to 2003. Both studies reported relatively stable survey trends in intimate partner violence during the 2000s, but the lack of consistent findings between national unemployment levels and intimate partner violence reported in these victimization studies may be a function of the varied time spans considered, the level of aggregation employed, or both. Petersen’s (2011) macro-level finding is important because it suggests that specified periods of economic growth and recession should be examined with regard to intimate partner violence.
The onset of the Great Recession, the time period in which we are specifically interested, started in the mid-2000s. The timing of this recent recession suggests that a more focused choice of years be examined that may capture the economic shock with more precision. Moreover, the use of national unemployment estimates in the Lauritsen, Rezey, and Heimer (2013) and Peterson (2011) studies may mask important differences in trends for smaller geographic units such as cities and towns. For instance, a study by Schneider, Harknett, and McLanahan (2014) used survey data to examine the relationship between local area-level unemployment rates and intimate partner violence for the years from 1999 through 2010. They found that worsening labor market conditions are associated with increases in the prevalence of violent/controlling behavior in marriage.
In sum, there is evidence to suggest that community-based measures of economic disadvantage may influence levels of domestic violence independent of individual and household factors. While it is understood that middle- and upper-class families experience domestic violence, families living in disadvantaged communities face a heightened risk of intimate partner violence. This has particular relevance for research on mortgage foreclosures, as there are studies reporting that the foreclosure crisis hit economically disadvantaged communities harder than others (Rugh & Massey, 2010). Two important issues we consider in this study are whether foreclosures predict levels of domestic violence during pronounced recession and whether the effect should be considered separately from other local indicators of economic stress, such as income or unemployment.
Mortgage Foreclosures as a Neighborhood Economic Stress Indicator
We conceptualize the role of foreclosures as an additional indicator of economic stress operating at the community level. This conceptualization is informed in part by the empirical literature on mortgage lending practices, which reports evidence that community dynamics are predictive of subprime lending practices, which in turn were linked to the surge in foreclosures (Rugh & Massey 2010). The subprime market provided opportunities for traditionally underprivileged households, often located in disadvantaged communities, to become homeowners. The predatory, high-cost lending practices that the subprime market supported resulted in many vulnerable home ownerships, because it allowed borrowers with poor credit histories and unstable income sources to take on risky mortgages, leaving them at greater risk of foreclosure.
Some researchers have documented that minority populations are at heightened risk of exposure to predatory lending and foreclosures, regardless of neighborhood factors such as age of housing stock and ownership levels (Jayasundera, Silver, Anacker, & Mantcheya, 2010). In Massachusetts, the subprime share of both Black and Hispanic household purchases increased from 6% in 1998 to almost 44% in 2005. The share of subprime purchases increased more modestly for White households, from over 2% in 1998 to 12.5% in 2006 (Gerardi & Willen, 2008). Been, Ellen, and Madar (2009) found that the likelihood of receiving a high-cost loan for borrowers increased as racial segregation of the area worsened. Research by Teasdale, Clark, and Hinckle (2012) found that subprime lending foreclosures had a substantial impact on property and public order crime.
As suggested by Lersch and colleagues’ (2014) cross-sectional study of Tampa Census Tracts, area-based measures of foreclosures may directly impact levels of domestic violence in the community. Based on the economic stress model, we would expect that those communities most affected by foreclosures would have higher levels of domestic violence. A description of the foreclosure process helps to elucidate the causal mechanism we suggest. Home mortgages entering the foreclosure stage begin with petition to file. The owner typically has several months to sell, refinance, or come to some other agreement with the bank. This is the time when economic stress triggered by a foreclosure notice may be highest. From a community-level perspective, mortgages in the earliest stage of foreclosure are therefore perhaps the best measure for predicting domestic violence, since a completed foreclosure will usually result in the owners leaving the premises, and perhaps the community. Foreclosures may influence levels of domestic violence in a second, more indirect manner, whereby foreclosures drive down property values of nearby properties, thereby jeopardizing the financial stability of neighbors and injecting hardship (Immergluck & Smith, 2006b). Such a process may also lead to homeowner economic stress and increased levels of domestic violence in the community.
That some communities experience more foreclosures than others presents an opportunity to explore how community foreclosures influence levels of family violence. This work may extend our understanding of the community context of domestic violence by considering foreclosures as a unique measure of economic stress, moving beyond the static, census-based neighborhood indicators of economic disadvantage typically considered in contextual studies of family violence. Moreover, since the foreclosure crisis occurred over a period of several years, it allows for consideration of whether changes in levels of foreclosure, precipitated by the Great Recession, are related to changes in levels of family violence.
Data and Method
The hypothesis for this study is that higher levels of mortgage foreclosures in a community will lead to higher incidence of domestic violence. Our time span for the study covers the years 2005–2009. Housing prices peaked in 2005, signaling the high point in the housing bubble that preceded the foreclosure surge (Byun, 2010). The dependent variable for the analysis is the monthly count of incidents of family violence reported to the police, compiled in the National Incident-Based Reporting Program (NIBRS) for Massachusetts cities and towns. NIBRS data have been used in a number of studies that examine the context and consequences of intimate partner violence (Hirschel, Buzawa, Pattavina, & Faggiani, 2007; Pattavina, Buzawa, Hirschel, & Faggiani, 2007).
Massachusetts has a total of 351 cities and towns, and during the study period 280 cities and towns reported to the NIBRS program. Of the 280, 51 cities and towns did not report to NIBRS for the entire period or had 12 total months or more of missing data and were thus excluded due to the importance of covering the recession for this study. Monthly estimates were imputed for cities that had 5 months or less of missing data. This resulted in a total of 229 cities and towns in the final sample. 1
The NIBRS data include incident-level details of crimes reported to law enforcement agencies. Information is recorded on the victim, offender, and offense characteristics. We included all incidents classified as aggravated assaults, simple assaults, and intimidation for incidents where the victim and offender were current or former intimate partners or other domestic relations (e.g., siblings, parent–child, etc.). Our reasoning for extending the measure of domestic violence to include current or former intimate partners or other domestic relations is that entire families come under strain when faced with having to leave a home. For instance, foreclosures may exacerbate parents’ negative behaviors like child abuse. A study by Wood and colleagues (2012) found that, between 2000 and 2009, rates of child physical abuse and traumatic brain injuries were associated with area-level foreclosures.
Monthly residential foreclosure data were obtained from the Warren Group, a commercial real estate data provider. We included only residential foreclosures of single family homes, multifamily homes, condos, apartment buildings, 2 and mixed use properties if living space was included. We treat month of petition filing as the starting point for the foreclosure process. After a mortgage has gone unpaid for a minimum of 3 consecutive months, the lender can file a notice (petition) of the intention to sue the property owner and reclaim the property if the loan is not paid off. 3 After the foreclosure petition is filed by the lender, the borrower may attempt to prevent the property from being foreclosed by restructuring the loan with the existing lender, refinancing the property with a different lender, or selling the property to a third party and satisfying the loan. The borrower may also turn over the deed to the property to the lender in lieu of paying off the loan (Shuetz, Been, & Gould, 2008). The independent variable of interest was measured as the rate of foreclosure petitions per 1,000 owner-occupied homes with a mortgage, using the number of owner-occupied homes as reported in the 2000 Census.
Central to our study is determining whether foreclosures have an effect on family violence independent from the more traditional community economic indicators such as unemployment and income. Kirk and Hyra (2012) argue that foreclosures are just another dimension to economic stress, and so any effects are thus confounded with other neighborhood indicators of disadvantage. To explore this possibility, we include local monthly unemployment estimates provided by the Massachusetts Executive Office of Labor and Workforce Development, and median household income reported in the 2000 Census. We also include a measure of residential stability available from the 2000 Census, measured as the percentage of the population age 5+ living in the same house for the last 5 years. According to a review by Beyer, Wallis, and Hamberger (2015), this variable has been considered in some research on intimate partner violence with mixed results. We believe this variable is relevant to our analysis because it reflects a prior level of resident stability existing in the community that could explain some of the relationship between foreclosures and intimate partner violence. We also included from the 2000 Census the percentage of the city that is minority (i.e., non-White). Finally, we controlled for city size by including a dichotomous variable indicating a value of 1 for cities with a population over 50,000. Table 1 reports descriptive statistics for these variables.
Descriptive Statistics.
As the dependent variable is measured as monthly counts of domestic violence incidents, our choice of statistical techniques was informed by both the distribution of counts and the time-dependent nature of the data for each city in the analysis. Poisson models are appropriate for the analysis of count data, but only when the data are not overdispersed. We conducted a z-test of this assumption as outlined in Hilbe (2007), and the results indicated that overdispersion was present in the data. We therefore used a negative binomial modeling design that accommodates overdispersed data.
The coefficients are derived from a generalized estimating equation (GEE) that produces population averaged estimates. This type of panel method is desirable for several reasons. First, this method allows us to adjust for temporal dependence often associated with repeated measures of crime over time. According to Hilbe (2007), this is an especially desirable estimation method for panel data with many units (cities) and small time intervals (monthly). Given that the best predictor of what the crime level may be in a given month is what it was the month before, we apply an autoregressive (AR1) adjustment. Second, this option allows us to include time-invariant independent variables as predictors. We are especially interested in separating the unique effect of foreclosures from other indicators of disadvantage with which the effect of foreclosures may be confounded, including the percentage of the community that is minority, median household income, and the percentage of the community living in the same house for the last 5 years.
The foreclosure rate was lagged by 2 months since in Massachusetts that is the typical time period after a petition is filed during which homeowners make arrangements regarding their intentions related to the mortgage (Gerardi, Shapiro, & Willen, 2007). This would potentially be a very stressful time in the household. 4 Recall that homeowners may opt to try and sell the home, refinance or make other financial arrangements to keep the home, or let the bank take possession of the home. Dummy variable indicators for each year are included (with 2005 as the comparison) as well as calendar quarters (with the first quarter as the comparison) to adjust for seasonality. No significant issues with multicollinearity were detected. 5
Table 2 presents the results of the GEE population averaged model estimated with robust standard errors. We present four separate models. The first two models include estimates without foreclosure rates (Model 1) and with foreclosure rates (Model 2). According to Model 1, the only community-level factor significantly associated with family violence is median household income. Higher median incomes are associated with lower levels of family violence. The results from Model 2 show that the foreclosure rate is a significant positive predictor of family violence in the community. For every additional foreclosure per 1,000 owner-occupied homes with a mortgage, family violence incidences increase by an average of 2%, controlling for other variables. Median family income is also significant in this model, which suggests that foreclosure rates have an effect on family violence that is independent of other economic stress measures. 6 Monthly unemployment, percentage minority residents, residential stability, and city size had no significant effect on levels of family violence.
Generalized Estimating Equation Regression Predicting Violence.
Note. Models have an exposure of the logged number of household. Models estimated using robust standard errors.
aSee Ballinger (2004) and Hardin and Hilbe (2013).
*p < .05. **p < .01.
In order to determine whether foreclosures also predict violence outside the family, we used the same models to predict measures of violent offenses committed by strangers and by acquaintances. These results are presented in Models 3 and 4 of Table 2. With the exception of lagged foreclosure rates, the results were similar. Foreclosure rates were not significantly related to violence committed by strangers and acquaintances. This null finding bolsters the idea that it is the foreclosure rates influencing family violence rates, rather than it being a spurious effect associated with overall violence in the community.
Discussion
The purpose of this study was to examine the influence of mortgage foreclosures on family violence during the Great Recession. Our results show that increases in mortgage foreclosures do lead to increases in levels of family violence in the community but not nonfamilial violence among strangers and acquaintances. These findings expand on the current literature on mortgage foreclosures and crime by conceptualizing foreclosures as a unique economic stress placed on families that may lead to violence in the home. This is a departure from prior studies that have grouped violent incidents together and have relied more heavily on theoretical explanations that emphasize how the physical changes in a community, left in the wake of foreclosures, explain levels of crime.
While there is ample evidence in the literature demonstrating the effects of individual, household, and community indicators of economic stress on family violence, the community-level indicators typically used in the research capture extreme disadvantage at the community level, such as poverty and public assistance, and are measured at a single point in time. We included as controls economic indicators intended to capture more variation in economic resources, such as median household income, as well as temporal variation of other stressors related to the recession, such as monthly unemployment. These factors may be more closely related to both the ability to avoid foreclosures and family violence. While both mortgage foreclosures and median family income had significant independent effects on family violence, unemployment did not, suggesting that there are important distinctions in the types of economic strain that can lead to domestic violence across communities.
By examining rates of citywide mortgage foreclosures by month over multiple years, our study adds another dimension to the study of family violence. Mortgage foreclosures offer a unique way to examine the role of economic stress on family violence at the community level. Our measures were close in time to the outcome and captured a small window of opportunity (months) where the stress is likely to be the greatest as families and homeowners contemplate resolution. Measuring foreclosures at the community level also provides the opportunity for future research to consider the possibility of spillover effects. For example, high levels of foreclosures may negatively affect the property values of nearby properties, indirectly causing stress on others living in the community. Foreclosures on rental properties may also create strain if tenants have to quickly relocate or otherwise are faced with an uncertain future regarding housing. More research is needed to unpack these specific mechanisms.
There are several limitations to our study. First, it is possible that some families may have left the community entirely during the several month lag period between the petition filing and resolution. If these are families with a high risk of violence, then leaving the community would remove any violent incidents from that community’s data, and perhaps displace these incidents into another community in our sample. Whether a move would intensify or reduce a families’ risk of violence after a foreclosure is an empirical question and should be considered in future research. Second, we did not have the capability of conducting a multilevel study where it would be possible to study the impact of foreclosures on family violence at both the household level and community level simultaneously. Recall that we considered only a short period of risk after a foreclosure notice, and it was not possible to study outcomes for families in the midst of experiencing this crisis. We also include cities and towns from a single state. The foreclosure crisis hit some states and communities harder than others (Pew Charitable Trusts, 2013) and our results may therefore be limited in generalizability.
Last, we used incidents of family violence reported to police for the study. It is a well-known fact that not all crime is reported to police. Fagan (1993) warns that studies examining economic and demographic variables in relation to family violence measured from official sources will likely suffer from bias due to selection processes. Those lower on the economic scales rely on public safety and public resources out of necessity, while those better off may seek help from private sources. During the period that we studied, it could be that the effect of foreclosures is confounded with other recession-driven forces relevant for understanding community-level reports to police. It is possible that reports to police may increase if other public resources or social supports typically available to victims become overwhelmed or discontinued during economic hard times. This is an issue for future research.
Policy Implications
While the foreclosure crisis seems to have abated in many communities in recent years, some in the financial community are skeptical about the long-term capacity of government programs to avert another foreclosure crisis. They argue that targeted relief is only temporary and has not reached the many Americans who need long-term support to stay in their homes, and the reality of foreclosure still looms large for many homeowners (Dayen, 2014). Furthermore, the reluctance of government oversight efforts to identify and prosecute those responsible for predatory lending practices sends the message to financial institutions that holding them accountable for foreclosure problems is not a priority (Schoen, 2013). Under these circumstances, any reprieve from surges in mortgage foreclosures may only be temporary.
Government assistance programs to help people avoid foreclosures may keep home ownership intact for some, but domestic violence victims embroiled in a foreclosure proceeding or other economic stressor related to housing (i.e., loss of equity due to reduced property values in communities with high foreclosure rates) face additional challenges stemming from the victimization; such challenges require immediate action to keep them safe while resolving an impending foreclosure. While we cannot be certain that our findings reflect an actual increase in the domestic violence rate in response to foreclosures, as much domestic violence goes unreported, what we can conclude is that there is an increase in public help-seeking behavior from victims. Further, our findings are consistent with reports of increased calls to domestic service providers during the recession (Jensen, 2008, Dethy, 2009). Policy options that deal with these dual crises must link any discussion of housing alternatives addressing immediate victim safety concerns with options that help survivors who are economically dependent on their abusers avert further financial difficulties.
Some victims may feel pressure to stay in abusive relationships because they lack the economic means to leave without risking homelessness. Brown (1998) makes a strong connection between homelessness and domestic violence. She reported that 63% of homeless women had experienced domestic violence (Browne, 1998). Women who leave their abusive partners often stay with family and friends temporarily, and some combine this type of informal support with more formal support services, such as homeless or domestic violence shelters (Baker, Cook, & Norris, 2003). In a worsening economy, family and friends may be less capable of offering assistance. Moreover, the already tight budgets of service providers to victims may be further stretched or cut during a recession, which may limit the housing options for victims who wish to leave an abusive relationship (Renzetti, 2009). This type of aid, while necessary, is usually temporary and does not address the underlying problem of affordable housing. The result might be that victims end up depending on the police for help as a last resort. Our finding that incidents of domestic violence increase with rising mortgage foreclosures suggests that more housing options for victims are critically needed in response to factors that create housing instability.
Mortgage foreclosures pose additional barriers for women who want to leave abusers or evict them from the home. The legal pressure for a couple to jointly litigate and settle a case can exacerbate the violence or impede a victim’s willingness to address the abuse. Victims must make timely and crucial decisions that may have long-term impacts on both their safety and economic circumstances. For example, the victim must decide whether to stay in the house and if can she afford to make payments. If she gets housing support in a protection order, what is her plan for making payments on the house when the order expires or the batterer does not pay according to the terms of the order? Is bankruptcy a viable option if she can’t afford to make payments? State laws vary in terms of housing relief available to victims of domestic violence, and it is imperative that family and consumer lawyers, foreclosure defense attorneys, and welfare and housing advocates be able to screen for domestic violence, keep informed of the ways in which domestic violence limits their client’s options (National Consumer Law Center, 2009), and develop creative solutions to survivor’s housing needs.
Any housing-related choice presented to a survivor must also explicitly consider the potential for economic abuse, especially for victims who are financially dependent on their abusers or victims whose abusers have access to or control of their assets. In order to maintain control over the victim, some batterers may resort to identify theft to maintain control over bank accounts and credit cards. They may take equity from the home without informing the victim. Batterers may engage in myriad acts of economic control and sabotage to keep their partners financially dependent and expose them to increased risks of violence (Plunkett & Sussman, 2011). Advocates and attorneys must work with police to prevent and prosecute these acts. Given the complex intersectionality between domestic violence and mortgage foreclosure, Renzetti (2009) argues that it is essential that service providers collaborate with and value each other as part of a comprehensive social safety net available to victims. Moreover, concentrated efforts to keep these services intact during economic downturns should be a government priority.
Conclusion
Our study shows that external economic shocks to communities in the form of mortgage foreclosures increase family violence after statistically controlling for other factors, likely by contributing to the economic stress of those living in that community. Victims and their families may be affected in many ways by foreclosures. The policy implications of our study provide added incentive to both continue and expand the recent policies and practices adopted to stem the tide of foreclosures generally as well as to connect these efforts more specifically to the work of family law, welfare, housing, and other social service advocates working to secure long-term housing stability for domestic violence survivors throughout the United States. Indeed, many states have enacted policies and procedures to stem the number of foreclosures, and there is some evidence that foreclosures are on the decline. However, we should not lose sight of the fact that without housing options in place for victims, another surge in foreclosures could result in higher levels of family violence in the community.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.