An Estimation of the Economic Impact of Spousal Violence in Canada, 2009

1. Introduction

The purpose of this report is to estimate the economic impact of spousal violence that occurred in Canada in 2009. This section begins with a brief discussion of the nature and extent of spousal violence in Canada, followed by the issue of gender in spousal violence, and the purpose of estimating the costs of crime. The subsequent sections describe the methodology used and provide estimates for three categories of economic impact: the impact borne by the justice system (criminal and civil), the impact borne by primary victims, and the impact borne by third parties.

1.1 Spousal Violence in Canada

Spousal violence (SV) is a widespread social issue that has garnered significant attention over the last three decades in social science and legal arenas, as well as in the media. Violence between spouses is unique in that the parties share a complex relationship involving physical, emotional, and economic bonds, with many violent spousal relationships also complicated by the presence of children. A victim of spousal violence is susceptible to sustaining long-lasting physical, emotional, and financial consequences. These effects are not confined only to the victim; they also stretch far beyond to the victim's family, friends, and employers. All members of society are also affected, whether through the additional financial strain imposed on the health care system or the lost future productivity of children exposed to spousal violence. Governments and other organizations are active in combating spousal violence, and a considerable amountof Canadian resources are redirected from other potential uses to address this issue. Despite decades of public awareness and prevention campaigns in Canada, spousal violence remains a devastating reality for many Canadian families of all social, economic, and cultural groups. National statistics show that, with the exception of spousal homicides, the incidence of spousal violence in Canada has not decreased over time (see Perreault and Brennan 2010).

Spousal violence is not a specific offence in and of itself in the Criminal Code, but many acts that constitute spousal violence are crimes in Canada. Offences often associated with spousal violence include common assault, assault with a weapon, sexual assault, homicide, forcible confinement, uttering threats, criminal harassment, and failure to provide the necessities of life. While psychological abuse and financial abuse are also forms of spousal violence (Mechanic et al. 2008; Adams et al. 2008), they are not necessarily considered crimes in Canada (Johnson and Dawson 2011). However, there is provincial and territorial legislation (for example, Prince Edward Island's Victims of Family Violence Act R.S.P.E.I. 1988, c. V-3.2; Manitoba's The Domestic Violence and Stalking Act, C.C.S.M. c. D93) that does address these forms of abuse. This legislation provides for responses in the civil context and complements the provisions outlined in the Criminal Code.

One way of describing the magnitude of spousal violence is through police-reported data. In 2009, 46,918 incidents of spousal violence were reported to police, representing 11% of all police-reported violent crime in Canada.Footnote 1 Incidents involving female victims accounted for 81% ofall incidents, highlighting the gendered nature of spousal violence.Footnote 2 Among these police-reported incidents, the majority (71%) involved current partners, while 29% involved former partners. Most violent incidents between spouses involved offences that had low severity ratings as defined by Statistics Canada, with common assault representing the largest proportion of reported offences (63%). Assault with a weapon or causing bodily harm accounted for 13% of spousal violence while uttering threats and criminal harassment made up smaller proportions at 10% and 7% respectively. At its most severe, spousal violence can end fatally. In 2009, there were 65 spousal homicides in Canada, 11% of all homicides in that year. As in past years, women continued to be more likely than men to be victims of spousal homicide; in 2009, almost three times as many women as men were killed by a former or current spouse. In addition, another 8 deaths were caused by other violations such as criminal negligence causing death.

Self-reported data, another source for examining spousal violence, reveal that the majority of spousal violence incidents do not come to the attention of police. The 2009 General Social Survey (GSS) finds that 1,186,000 Canadians aged 15 and older in the provinces reported being physically or sexually victimized by a spouse in the preceding five years. In 2009 alone, 335,697 Canadians were victims of 942,000 spousal violence incidents. Less than one quarter (22%) of victims stated that the incident in the previous five years had come to the attention of police, a rate that has declined from 28% in 2004 (Brennan et al. 2011). The GSS data also show that of those victims who do alert the police, many do so only after experiencing multiple episodes of spousal violence. For instance, 28% of victims reported that they had been victimized more than ten times before they contacted the police. More than three quarters of spousal violence victims reported being emotionally affected in addition to sustaining physical injuries. The GSS findings show that spousal violence is more likely to occur between ex-spousesthan current spouses, as 17.8% of people with a previous relationship experienced violenceeither before or after the separation while 3.8% of people in a current relationship experiencedviolence. Findings from the GSS also indicate that females continued to report more serious forms of spousal violence than males, despite equal victimization rates when all incidents were included.

This report presents a great many numbers: national statistics, numbers from other studies, calculations and financial estimates. It is important to remember that behind these numbers are actual people – women and men – who have experienced spousal violence. The following Text Box displays a summary of the demographic profile of the victims, so that readers may have a better understanding of who they are. All data are from the 2009 GSS and include victims of both current and ex-spousal violence.

Text Box 1.1: Who were the victims?

Just over half of the victims of spousal violence were female (54%). The majority of these females were aged 25 to 44 (67%). Almost three quarters (70%) had post-secondary education (24% a university degree, 31% a college or technical diploma, and 15% some university of college). Just over one-tenth (12%) had elementary schooling or did not complete high school and the remainder (18%) had a high school diploma.

In terms of annual personal income (i.e. not their household income), 5% of females had no income and 5% had an income over $100,000. More than half (62%) had an income less than $39,999. Household income was under $19,999 for 12% of women.

The majority of females (80%) lived in urban areas and most were Canadian-born (86%).Footnote 3 Almost three-quarters (70%) of females spoke English as their main language, with a quarter (25%) speaking French and 5% speaking neither English nor French as their main language. Less than one-tenth (7%) cared for children and/or performed housework as their primary occupation while more than half (57%) were employed or seeking employment. More than a quarter (28%) volunteered or cared for children other than their own.

Less than half (46%) of the victims were male. The majority of these males were aged 25 to 44 (58%), while a further 23% were aged 45 to 54. Almost four-fifths (79%) had post-secondary education, 16% had a high school diploma, and 5% had elementary level education or had not completed high school. In terms of annual personal income, one-fifth (20%) earned more than $100,000, more than half (61%) earned from $20,000-$69,000, 7% earned less than or equal to $19,999, and no males reported no income.

Eighty-seven percent of males lived in urban areas and 85% were Canadian born. In terms of main language spoken, 71% of males spoke English, 24% spoke French, 1% spoke both and 4% spoke neither English nor French as their main language. In terms of their main occupation, 71% of males were employed or seeking employment, 3% were in school, 2% reported caring for children or doing housework, and 18% volunteered or cared for other children.

The GSS collects quantitative data and with these data paints a picture of sorts of the prevalence and nature of spousal violence in Canada. In reading the numbers in the pages that follow, do remember the victims.

1.2 Measuring Spousal Violence

Johnson and Dawson (2011, p. 65) note that “the frequency, severity, consequences, and context of intimate partner violence are gender-specific with distinct victimization experiences for men and women.” However, ascertaining the true differences in victimization rates and experiences between genders is difficult, and there is no consensus among academic researchers on the true composition of spousal violence. Some research has shown that both frequency and severity of spousal violence victimization are greater for female victims than for male victims (AuCoin 2005; Tjaden and Thoennes 2000). Other research indicates that the frequency of female victimization is greater and that females sustain more injuries but that severity of injuries is greater for male victims than female victims (Felson and Cares 2005). Research also suggests that the frequency is equal, but that women experience more severe victimization (Archer 2000).

Kimmel (2002), Felson and Cares (2005), Johnson (2008), and Allen (2011) offer some reasons for the contradicting results found in different studies. The major differences can be explained by the different types of surveys used in these empirical studies; the rates and severities of violence revealed crucially depend on the survey questions (definitions), assumptions, and samples. There are two main types of spousal violence surveys: police-reported crime surveys and self-reported victimization surveys based on the Conflict Tactics Scale (CTS).

Police-reported crime surveys only include incidents that are reported to police (the UCR2 in Canada) and are perceived as a crime, implying that only incidents that conform to Criminal Code offences are included. Studies using police-reported crime survey invariably find higher rates of violence against women (Kimmel 2002). Self-reported victimization surveys based on the CTS (the GSS in Canada) take samples from the general population and often report equivalent rates of violent victimization for men and women. This latter type of survey is more general and includes minor violent acts together with more severe incidents. A third type of survey is shelter-based (the Transition Home Survey (THS ) in Canada). Surveys of this type show prevalence and severity results similar to police-reported crime surveys with women being the overwhelming victims. It is important to note that many shelters are only open to women and as such, results are not representative of the population (Statistics Canada 2009).

Some researchers claim that the violence captured in CTS surveys is taken out of context and hence, gender symmetry in intimate partner violence (IPV) is misleading. Allen (2011) explains that the main survey types paint different pictures of IPV because they actually capture different types of IPV. When IPV is subdivided into intimate terrorism (IT) and common couple violence (CCV), the ostensibly contradictory results from the different survey types can be reconciled. IT is characterized by one spouse using severe violence to gain power, control, and domination over the other spouse, while CCV is characterized by mostly minor violence triggered by conflict issues common to many relationships. Graham-Kevan and Archer (2003) find empirical evidence that distinguishing between IT and CCV does indeed reconcile the findings based on different types of surveys. Using a CTS survey, they find that 87% of IT is perpetrated by men and that 55% of CCV is perpetrated by women.

To investigate the gender differences of spousal violence victimization, the present report disaggregates the costs by gender of victim. Data from the Canadian police-reported crime survey and the self-reported victimization survey are used, each where appropriate.

1.3 Costs of Crime

Examining the economic impact of social phenomena is an established approach in social science research, and cost estimation methods have been extensively developed in the field over the past decades (Cohen 2005). The topic of crime, in particular, has garnered the attention of policy-makers and of researchers who have subsequently conducted studies on crime's economic impact. While a minority would disagree with the notion that enumerating the costs of crime is an effective and worthwhile exercise (Zimring and Hawkins 1995), proponents of crime costing contend that an understanding of the economic impact of crime can be important to policymakers and may assist in the proper allocation of resources both within the criminal justice system and between different social issues.

Economic exercises are often criticized for attempting to place a monetary value on intangibles. An individual's life and the pain and suffering of a victim are concepts that do not have a price determined by natural economic forces, as in a marketplace of goods where supply and demand determine prices. It is therefore difficult to attach a monetary figure, and some people may even claim that it is insensitive to do so. However, it can be argued that the benefits of performing costing exercises outweigh the potential negatives, and that estimating the costs of crime provides practically useful information.

As Cohen (2005) states, “the cost of crime” is equivalent to the “benefit of reducing crime”, which in turn is tantamount to “the monetary amount that society would be willing to spend to prevent a criminal incident from occurring”. It is important to remember that, regardless of the relative wealth of a society, economic resources are scarce. Estimates of the economic impacts of different social phenomena are therefore crucial in determining (through marginal condition) the efficient allocation of resources to programs competing for public funds (e.g., a larger police force, increased health care capacity, new transportation infrastructure, or more public parks).

Policy development implies the allocation of scarce public resources. Guiding questions for policy-making governments include:

  1. How do we distribute public funds efficiently?
  2. How do we distribute public funds fairly?

The efficient distribution of money entails getting the maximum impact given the limited funds with no regard to equality of spending. If spending all of the public money on one issue will generate greater overall benefits than allocating even a small portion of funds to another issue, efficiency would dictate the former option. However, doing so may be considered unfair and a decision to provide at least some funds to address each of the issues might be desirable. These guiding questions are often in tension and a balance must be found between efficiency and fairness.

Cost estimates assist in and facilitate the answering of these questions. Money can be viewed as a universal, objective language, and projecting objective monetary units onto subjective intangibles, such as mental health effects or pain and suffering, is a way of standardizing the competing social issues and allowing for a direct comparison. Costing exercises are fraught with challenges and those pertaining to this report specifically are described in the following Methodology section.

Several victimization costing exercises have been conducted in Canada, but few have focused specifically on spousal violence. The present research will attempt to fill this knowledge gap and will serve as a resource for all stakeholders working together to end spousal violence and to assist the victims, their children, and the many others affected.

Table 1.1 lists previous work on costing of domestic violence, intimate partner violence, and violence against women in Canada specifically. Walby (2004) and Varcoe et al. (2011) provide thorough reviews of the international literature of the costs of domestic violence up to 2010.

Table 1.1: Costing Studies on Domestic Violence and Intimate Partner Violence in Canada
Authors Description Results Results
(2009 $)Table note
Day (1995) Estimates health-related economic costs of violence against women in Canada (includes medical costs, work loss, transition homes, etc.). $1.54 billion (1993) $2.05 billion
Greaves et al. (1995) Estimates economic costs of sexual assault, intimate partner abuse, and child sexual assault against women in Canada in health, criminal justice, social services/ education, and labour/ employment. $4.23 billion (1994) $5.55 billion
Kerr and McLean (1996) Estimates economic costs of violence against women in BC in areas such as policing, corrections, income assistance, lost work time, transition homes, etc. $385 million (1994/1995) $502 million
Varcoe et al. (2011) Estimates the costs associated with intimate partner violence for women leaving abusive partners. $6.9 billion (2011) $6.62 billion