Correctional Service Canada
Symbol of the Government of Canada

Common menu bar links

Compendium 2000 on Effective Correctional Programming

Warning This Web page has been archived on the Web.


Cost-Effective Correctional Treatment


The evidence is overwhelming -- human service-based interventions reduce criminal recidivism; punishment does not. Recently synthesized findings based on over 500 studies spanning five decades of research clearly indicate that any kind of human-service based treatment reduces recidivism on average by 10% (Andrews et al., 1990; Lipsey, 1992; Lösel, 1995). Moreover, treatment approaches that follow empirically validated principles of effective intervention (Andrews & Bonta, 1998) yield substantially higher reductions ranging from 26% to 40% (Andrews, Dowden, & Gendreau, 2000; Lösel; 1996).

Clearly, we know what works with certain offenders. Equally important, however, is determining whether or not effective interventions are cost-effective from an economic perspective (Elliott, 1980; Morris & Braukmann, 1987; Rossi, Freeman & Lipsey, 1999; Weimer & Friedman, 1979). Efficiency evaluations, more commonly known as cost-benefit and cost-effectiveness analyses, strive to promote optimal resource allocation. In economic terms optimal resource allocation is achieved when no one is made better off without simultaneously making at least one person worse off. In practice however, net social welfare is rarely improved without negatively impacting at least one person. Consequently, economists have adopted a less stringent decision rule for evaluating economic efficiency. Specifically, a given resource allocation strategy is considered efficient if and when an overall net monetary benefit is realized. Thus, a correctional treatment program that costs $1,000,000 but generates $3,000,000 in terms of reduced criminal justice costs would be considered cost-effective.

Cost-benefit analyses (CBA) generate conclusions such as “every dollar spent on program X saves the taxpayer $10.00 in the long run”. In contrast, cost-effectiveness analyses (CEA) report the benefits of a given program in substantive rather than monetary terms. Thus, a cost-effectiveness analysis might conclude, “sex offender treatment costs $12,000.00 per potential victim saved” (Marshall, 1992). Thus, while cost-benefit approaches monetize program benefits, cost-effectiveness methods do not. Cost effectiveness analyses, simply determines for example how much it costs to save one human life, or to prevent one victim from experiencing the emotional pain and suffering associated with sexual assault.

Since the 1 960s, government regulatory boards have either recommended (e.g., Treasury Board of Canada, 1976; 1998) or required (e.g., United States Office of Management and Budget, 1989; 1996) that cost-benefit and cost-effectiveness techniques be used to assist policy decision-makers in the allocation of public resources. Surprisingly, efficiency evaluations have been noticeably absent from the psychological treatment literature. For example, a recent review revealed that less than five percent of psychologically-based treatment outcome evaluations conducted between 1967 and 1991 included CBA or CEA results (Yates, 1998). Similarly, Welsh and Farrington (2000) were only able to identify seven studies featuring control groups that applied cost-benefit analysis to correctional program evaluation. Nonetheless, efficiency evaluations are rapidly becoming popular, particularly within the criminal justice realm (Aos, Phipps, Barnoski, & Lieb, 1999; Cohen, 1999; Donato & Shanahan, 1999). Further, as Cohen aptly notes, “criminal justice researchers and policy makers will increasingly be confronted with cost-effectiveness and benefit-cost analyses -- whether they like it or not” (1999; p. 2).

This chapter describes some of the potential benefits to victims, society and offenders that can result from effective correctional programming. Conversely, it also reviews the costs associated with providing such treatment. The chapter proceeds to address some of the more common methodological issues surrounding efficiency evaluations. Lastly, it highlights the results of corrections-based efficiency evaluations that have been conducted over the last 10-15 years.


Crime impacts victims, their families, as well as their friends. Robbery victims lose money. They may also suffer physical or emotional trauma that will require short term and quite possibly, long term medical care. Similarly, family and friends may be forced to take time off work without pay to care for the needs of a crime victim.

Correctional programs that are successful in preventing offenders from committing new crimes may not only save potential victims money but they may also avert future pain and suffering. Thus, programs that reduce recidivism generate benefits in the form of future reductions in crime-related costs that would have otherwise accrued in the absence of treatment. At first glance the identification of program benefits appears seemingly straight-forward. However, as Rossi et al. (1999) note identifying such benefits is somewhat subjective, dependent upon one's perspective, be it the program participant, the program sponsor, the victim, or society at large. Further, the literature has described benefits that are direct and indirect as well as those that are tangible and intangible.

Direct program benefits are intentional. They are known before the program is implemented and are expected to accrue as a direct result of the program. In contrast, indirect benefits are unintentional and unplanned. They are best thought of as positive, albeit secondary by-products of the program. Thus, from a contemporary rehabilitative standpoint, reductions in recidivism would constitute a direct benefit while enhanced self-esteem would represent an indirect benefit.

Benefits can be further differentiated in terms of whether or not they are intangible or tangible (Kiessling, 1976; Laplante & Durham, 1983). While tangible benefits are quantifiable and can be expressed in monetary terms, intangible benefits can not (Cohen, 1999; Rossi et al., 1999). Although, recent techniques have provided a means of translating seemingly intangible benefits (e.g., human life) into monetary terms (Cohen, Miller, & Rossman, 1994; Miller, Cohen, & Wiersema, 1996). Further, goods and services traded in private or public markets will usually be considered tangible in nature.

Table 27.1 describes various domains where potential victims, families, and friends can benefit from effective correctional programming. More specifically, not only can appropriate treatment save lives but it can also prevent future victims and their families from having to experience the pain and emotional suffering that inevitably accompanies any type of crime. Moreover, treatment can prevent future victims from having to take unpaid leave from work or from having to pay for medical expenses that may follow victimization. Lastly, treatment can spare future victims and their families from having to engage in crime avoidance behaviour.

TABLE 27.1 Potential Correctional Program Benefits to the Victim

  • Out-of-pocket expenses (costs not reimbursed by health care or insurance companies: medical costs-physical & emotional trauma, forgone/lost productivity-lost wages, school days, housework days, and property losses-stolen, damaged, money)
  • Crime avoidance behaviour (e.g. purchasing alarms, locks, weapons)
  • Reduced quality of life (pain & suffering associated with non-fatal injuries, activities that can no longer be performed as a result of the injury)
  • Loss of life
  • Crime avoidance behaviour (e.g. restricted activity-no running at night)
  • Out-of-pocket expenses (family and friends attending to physical needs of victim resulting in forgone productivity, e.g. lost wages, lost school days, lost housework days)
  • Negative spill over to family & friends (crime avoidance behaviour -- purchasing locks, alarms, weapons)
  • Pain and suffering experienced by family and friends when a loved one is injured/murdered
  • Negative spill over to family & friends (crime avoidance-restricted activity-no running at night)

Out-of-pocket expenses

Out-of-pocket expenses refer to tangible, financial losses that are incurred by the victims of crime as well as their family and friends. They do not include expenses that are reimbursed to victims through private insurance companies or government health care systems. Thus, expenses associated with damaged or stolen property, stolen money, medical bills (for physical injury or emotional trauma), unpaid work days, or lost house-work days that are not absorbed by the system but are the responsibility of the victim are considered out-of-pocket expenses (Cohen, 1988, 1998; Cohen et al., 1994; Holahan, 1973; Rajkumar & French, 1997; Weimer & Friedman, 1979). Similar expenses incurred by the family and friends of a victim are also considered tangible, out of pocket expenses, albeit indirect. Examples include: a parent who takes unpaid leave to care for an injured child or a husband who takes unpaid leave to accompany his wife while she testifies in a court of law. While indirect effects to family and friends are recognized in theory, corresponding cost estimates have yet to be generated.

Crime avoidance behaviour

Crime victims may experience a heightened sense of fear associated with the possibility of future victimization. In theory, this fear may manifest itself in various forms, collectively coined, crime avoidance behaviour. For example, “....during the weeks or months following the (rape), women frequently make costly changes in their lifestyles; this may involve moving to a “better” neighbourhood, buying expensive security systems, or avoiding work situations which they suddenly perceive as dangerous” (Burt & Katz, 1985; p. 333). Tangible crime avoidance behaviours include purchasing products (e.g., alarms, weapons, and locks) and services (e.g., self-defence courses, security guards, guard dogs) designed to reduce the probability of future victimization. In contrast, intangible crime avoidance behaviours focus on lifestyle restrictions such as refusal to jog or take public transportation at night. Further, it is not inconceivable that victim-related crime avoidance behaviours may spill over and promote similar behaviours among family and friends (Cohen et al., 1994). Once again, cost estimates for either direct or indirect forms of crime avoidance behaviour have yet to be generated.

Loss of life and reduced quality of life

Successful programming can reduce fatalities (i.e., loss of life) and prevent potential victims and their families from experiencing a reduced quality of life that invariably accompanies non-fatal injuries. Unlike out-of-pocket expenses, quality of life factors emphasize the intangible consequences of crime (Cohen, 1988, 1998; Cohen et al., 1994; Miller et al., 1996; Rajkumar & French, 1997). These factors attempt to capture the mental anguish as well as the actual physical pain and suffering associated with victimization and its consequences. Thus, efficiency evaluators recognize the importance of incorporating factors such as the emotional cost to a victim who can no longer tie her shoes as the result of a debilitating car accident caused by a drunk driver, or the cost to a father who can no longer play baseball with his son due a crime-related injury. Similarly, they also acknowledge for example, the robbery victim who is forced to endure lifelong chronic pain resulting from a gunshot injury or a rape victim who can no longer sleep through the night due to persistent nightmares (Cohen et al., 1994). Lastly, sparing family members and friends from having to watch a loved one cope with the after-math of a crime or deal with the death of a loved one are also recognized as additional benefits, albeit secondary or indirect. Given that current procedures for estimating victim pain and suffering invariably include familial pain and suffering, Cohen et al. recommends their exclusion.


Crime impacts almost every segment of society from the tax-payer, to the person who has never been directly victimized by crime but fears it nonetheless, and to the people and organizations tasked with the responsibility of apprehending, prosecuting, incarcerating and treating the perpetrators of crime. As Table 27.2 illustrates effective correctional programming can generate various benefits to society that are either direct or indirect in nature or tangible or intangible.

TABLE 27.2 Potential Correctional Program Benefits to Society

  • Criminal justice system (police, adjudication, corrections)
  • Offender's forgone taxable income
  • Victim forgone productivity
  • Medical expenses covered by health care system (physical and emotional)
  • Insurance and disability claims including administrative costs
  • Victim services (e.g. counselling, temporary shelter, financial aid)


  • Future crime prevention (broken cycle of violence)
  • Other non-criminal justice crime prevention programs (e.g. Neighbourhood Watch, Crime Stoppers)
  • Transfer payments to offender's dependants
  • Crime avoidance behaviour by potential victims (e.g. alarms, locks, weapons)
  • Crime avoidance behaviour by potential victims (e.g. restricted activity-no night running)
  • Generalized societal fear of crime

Virtually all efficiency frameworks include reductions in criminal justice expenses that would have otherwise accrued in the absence of treatment. For example, successful programming can reduce the number of offenders who will be sought out by the police, prosecuted in a court of law, and subsequently incarcerated in a correctional facility. This in turn reduces the cost of administering the criminal justice system, a burden that is inevitably shouldered by the taxpayer (Cohen, 1998; Cohen et al., 1994; Gray, 1979; Holahan, 1973; Miller et al., 1996).

Additionally, a large portion of effectively treated offenders will also obtain legitimate employment. As a result, society will benefit from an additional source of taxable revenue that would have otherwise been absent if the offender had not been treated and consequently, had resumed criminal activity (Miller et al., 1996; Rajkumar & French, 1997; Seashore, Haberfeld, Irwin, & Baker, 1976). Similarly, reduced recidivism rates translates into fewer victims in need of time away from work, school or from house-hold duties. Thus, society will suffer fewer productivity losses that would have otherwise accrued in the absence of effective programming.2 Fewer victims also means reduced health care costs, fewer insurance claims and a decreased demand for victim-related services offered outside of the traditional health care system (e.g., temporary shelter and volunteer counselling services). All of which may result in reduced taxes, possibly a reallocation of resources, and lower insurance premiums.

The literature has also identified a number of plausible indirect benefits to society.3 Laurence and Spalter-Roth (1996) note that domestic violence initiatives may inadvertently benefit society by breaking the cycle of violence. Thus, while a program's primary goal may be to treat the present-day perpetrator it might also generate secondary, positive effects such as preventing child victims or witnesses of domestic violence from manifesting similar behaviours in adulthood. Similarly, Cohen et al., (1994) have argued that if the impact of the program was sufficiently large the need for non-criminal justice organizations such as Neighbourhood Watch and Crime Stoppers would eventual decline, thus resulting in substantial cost savings or a redistribution of resources. Some studies (Knox & Stacey, 1978) have also included reductions in transfer payments to the dependants of offenders as valid, tangible benefits to society. However, others have argued for their exclusion (Mallar & Thornton, 1978). Additionally, successful programming could also result in global reductions in society's generalized fear of crime as well as the extent to which society engages in crime avoidance behaviour (Cohen et al., 1994; Gray, 1979; Holahan, 1973; Miller et al., 1996; Phillips & Votey, 1981). However, once again attempts to quantify such effects in a reliable manner have yet to emerge.


The perpetrators of crime do not escape unscathed. If caught, they may be sentenced to a prison term resulting in loss of freedom, forced separation from loved ones and missed employment opportunities. Similarly, their family and friends may also be adversely impacted. Nonetheless, correctional efficiency evaluators such as Cohen (1998) have dismissed offender-related benefits given that it would be philosophically inappropriate to include such factors within a cost-benefit framework. Moreover, such factors are more likely to dissuade rather than persuade policy decision-makers regarding the merit of offender treatment. Regardless, an overview of potential programming benefits that may accrue to the offender is provided for discussion purposes.

As Table 27.3 demonstrates offenders may receive direct, tangible benefits from programming. Employment-specify programs or those that target factors that help offenders maintain steady employment (e.g., substance abuse, interpersonal skills training) will increase the number of offenders who obtain and maintain jobs, thereby increasing legal income. Hence, success-fully treated offenders will receive legal income that would have otherwise been lost if the offender had maintained a criminal lifestyle.4 Additionally, it is arguable that successful programming can generate direct, albeit intangible benefits. Some rehabilitated offenders may no longer experience anxiety or the constant fear of detection that may or may not accompany certain criminal individuals. Similarly, they will no longer experience the pain and suffering associated with non-fatal crime-related injuries. As well, the odds of losing one's freedom or being fatally wounded during the commission of a crime will also decrease considerably.

As Table 27.3 indicates correctional programming may also inadvertently generate indirect benefits that are tangible as well as intangible. For example, what an offender learns during programming may transfer or spill over to his/her friends, family, or children (Levine, 1983; Nas, 1996; Rossi et al., 1999). Offenders who gain improved job search techniques such as resume writing and interviewing skills may share this knowledge with family and/or friends who in turn are able to secure better jobs and enhance their own earning potential. Additionally, correctional programs may unintentionally improve an offender's overall well being, family functioning, as well as non-familial relationships. These factors are viewed as indirect given that the primary goal of offender treatment should be to generate reductions in recidivism rather than increases in an offender's overall well being. Although all of the aforementioned benefits are theoretically possible they have yet to be examined in practice.

TABLE 27.3 Potential Correctional Program Benefits to the Offender

  • Forgone legal earnings after taxes
  • Quality of life (fear of detection, pain & suffering from non-fatal injuries)
  • Loss of life
  • Loss of freedom
  • Positive spill over of treatment effects to friends & family
  • Improved well-being
  • Improved familial relations
  • Improved non-familial relations



The literature has not fully explored potential victim-related costs associated with program delivery (see Table 27.4). Certain programs such as restorative justice initiatives may impose direct monetary expenses on victims such as foregone wages and transportation costs associated with program participation. Similarly, the notion of rehabilitating rather than punishing perpetrators of crime may evoke feelings of injustice from individual crime victims as well as victim advocacy groups. However, to date, the literature has not developed methods for quantifying such factors. Further, it is likely that the development of such techniques will be challenging if not entirely impossible.

TABLE 27.4 Potential Correctional Program Costs to the Victim

  • Out-of-pocket expenses associated with program participation (e.g. travel costs, lost wages, school days, housework days associated with some restorative justice initiatives)



  • Feelings of injustice associated with treating offenders from immediate victims of crime and victim advocacy groups



From a societal perspective, the literature has primarily focused on direct program-related costs that are tangible in nature (see Table 27.5). Examples include: the cost of hiring staff to deliver a program, the cost of purchasing the necessary equipment and materials to run a program, and the cost of operating the actual facility from which a program is run.

Some researchers have identified indirect costs that may be associated with correctional programs (see Table 27.5). A correctional industry program (e.g., CORCAN) may cause displacement effects whereby newly trained ex-offenders who are willing to work at a lower rate, replace higher paid, prosocial workers (Laplante & Durham, 1983; Levine, 1983; Rossi et al., 1999). Intensive community-based treatment programs may inadvertently interfere with an offender's probability of securing and maintaining steady employment. This results in a reduction to the offender's overall contribution to the workforce that may in turn reduce available taxable income. As well, treatment may inadvertently prolong incarceration which in turn increases incarceration costs (T3 Associates, 1999). Lastly, society may reject community-based programs (e.g., halfway houses) or any form of human-service based intervention for that matter, on the grounds that treatment is incongruent with get tough on crime policies and that it heightens society's generalized fear of crime. It should also be noted that it is unclear in terms of whether or not societal opposition to treatment should be considered separately or conjointly with victim advocacy opposition. To date, no studies have attempted to quantify any of these indirect costs.

TABLE 27.5 Potential Correctional Program Costs to Society

  • Program costs (personnel, facilities, equipment, materials)


  • Displaced workforce
  • Forgone productivity (i.e. offender's taxable income)
  • Prolonged incarceration costs (e.g. not released until program completed)
  • Inconsistent with ‘get tough on crime policies'
  • Heighten fear of crime (e.g. halfway house located in one's neighbourhood)



The efficiency literature has generally neglected direct program costs for offenders. However, as Table 27.6 demonstrates offenders may in theory, experience indirect costs. A community-based program may unintentionally reduce an offender's income during the duration of program delivery. Similarly, the practice of favouring institutional over community-based treatment coupled with a parole board's reluctance to grant early release to individuals who have not experienced a full bevy of treatment programs prior to release may inadvertently result in prolonged incarceration. This in turn may produce intangible emotional discomfort for the offender.

TABLE 27.6 Potential Correctional Program Costs to the Offender





  • Forgone legal earnings after taxes
  • Prolonged incarceration (emotional distress associated with loss of freedom).

In sum, the correctional efficiency literature has identified numerous potential benefits and costs associated with appropriate programming. The offender, the victim and society stand to benefit substantially from effective correctional programming. However, it is unlikely that reductions in global phenomena such as crime avoidance behaviour, societal fear of crime, the cycle of violence, or the reliance on non-criminal justice prevention programs could be causally linked to any one specific program (Cohen, 1998). Further, while a distinction has been made between activity-related crime avoidance behaviour and society's generalized fear of crime, in practice separating the two may prove impossible. Lastly, several of the previously reviewed costs and benefits exist merely as hypothetical constructs rather than quantifiable entities. Nonetheless, the literature has developed defencible methods for quantifying many of the existing crime-related consequences and program costs.


Efficiency evaluators assign dollar values to the costs and benefits of correctional programming using a variety of methods. While most tangible victim and criminal justice costs can be estimated from survey data and financial records, the majority of crime-related costs are not directly accessible. Consequently, efficiency evaluators rely on a number of indirect estimation procedures including shadow prices, the jury compensation method and the willingness to pay approach. Unlike direct estimation, indirect estimation can be controversial and complicated (Rossi et al., 1999).

Financial records

For the most part, programming (e.g., staff, materials) and criminal justice costs (e.g., policing, incarceration) can be estimated directly from financial records (Cohen et al., 1994; Cohen, 1999; Miller et al., 1996). However, at times it may be difficult to separate marginal from fixed costs. While fixed costs remain constant regardless of whether or not a program is implemented, marginal costs vary as a function of the program. They represent the added or incremental cost associated with treatment delivery, independent of the cost of running a correctional institution (e.g., electricity, food, clothing). A basic principle underlying all efficiency evaluations is that cost be restricted to marginal expenses (Cohen, 1999; Levine, 1983; Nas, 1996). Thus, standard institutional operating costs (e.g., electricity) that exist regardless of whether or not a program is implemented should be excluded. Similarly, other criminal justice expenditures such as the annualized cost of maintaining a police station should be excluded given that they typically do not change as a function of a unit decrease in crime. Unfortunately, marginal costs are not always readily available. Consequently, evaluators sometimes use average or aggregate costs, a procedure that typically overestimates expenditures (Laurence & Spalter-Roth, 1996; Tonry 1990).

Survey data

Victim survey data is commonly used to assess monetary victim expenses including medical costs, lost property, and forgone earnings (Cohen, 1988, 1998; Cohen et al., 1994; Miller et al.,1996). The criminal justice literature has primarily relied on the American National Crime Victimization Survey (NCVS; Bureau of Justice Statistics, 1998) as a means of estimating tangible victim costs. The NCVS polls households regarding information pertaining to recent criminal victimizations. The survey however, has been criticized for underestimating victim-related costs. For example, the survey only elicits information about medical costs incurred during the past six months. Further, it does not inquire about mental health costs, forgone housework or school-related productivity. Additionally, such surveys implicitly assume that respondents can provide accurate cost estimations for expenses commonly paid for by insurance companies (Cohen et al., 1994; Rajkumar & French, 1997). Miller et al. (1996) circumvented some of these problems by combining the survey data with primary file information obtained from workers compensation and hospitalization records. Interestingly, this procedure resulted in substantially higher cost estimates than those derived from the NCVS. For example, Miller et al. report cost estimates that are 10 to 20 times higher for certain violent crimes than those reported in the NCVS (Cohen, 1999). However, cost estimates derived from Canadian victimization surveys and the Canadian health system are still required if accurate Canadian-based cost-benefit analyses are to be conducted.

Shadow pricing

Economists commonly quantify costs based on market prices. The market price for most tangible commodities is readily avail-able and hence traded openly by buyers and sellers in the market place. However, more often than not, market prices are distorted for various reasons including government subsidies and monopolies. When this occurs, researchers rely on shadow prices (Rossi et al., 1999; Weimer & Friedman, 1979).

Shadow prices refer to the true market price that would exist under perfectly competitive market situations or when the demand for a given commodity equals its supply. In essence, a shadow price hides behind or shadows the observed market price (Laplante & Durham, 1983; Levine, 1983; Nas, 1996). Assume for example that a researcher is evaluating a new pilot program for violent offenders. Initially, post-graduate clinical students deliver the program. Further, the students receive academic rather than monetary credit for their participation. In this situation it would be more accurate for the evaluator to determine the cost of the program based on estimated salaries of paid treatment deliverers rather than clinical students (Nas, 1996; Rossi et al., 1999).

Willingness to pay

In some circumstances a market does not exist for a given crime-related cost such as human life or emotional pain and suffering. While a cost-effectiveness analysis is a viable option (Levine, 1983), economists and criminal justice researchers have developed alternative methods for quantifying intangible victim costs. Two current methods include the willingness to pay approach and the jury compensation method.

The willingness to pay approach estimates how much society is willing to pay to reduce the risk of crime-related death by extrapolating from non-crime studies that have examined for example, how much society is willing to pay to ensure safe work-places or safe vehicles. The most common study estimates how much society values human life by examining worker's willingness to accept riskier jobs in exchange for a premium wage. Two independent reviews of this literature (Viscusi, 1993; Miller, 1990) involving 70 different studies revealed that the statistical value of a human life has been estimated between $500,000 and 7 million (average: 2.7-5 million).

Critics argue that the willingness to pay approach is not only morally wrong but that it is also technically flawed in that it assumes people correctly perceive the risk associated with a given behaviour (Zerbe, 1998). Further, it is also possible that people would be willing to pay substantially more for reducing the risk of death from a violent crime rather than a workplace accident. Nonetheless, the assignment of monetary value to human life has become standard operating practice for prominent government regulatory boards, albeit the Treasury Board of Canada (1998) has adopted a somewhat more conservative approach. Further, most efficiency experts would agree that currently, this method remains unrivalled given that earlier cost studies grossly under-estimated the value of human life by relying solely on forgone productivity. This method not only failed to consider the intrinsic value of life (e.g., pain and suffering, enjoyment of life) but it also implicitly placed less value on the elderly and the young (Cohen et al., 1994; Laurence & Spalter-Roth, 1996).

Jury Compensation Method

Cohen and his colleagues (1994; Miller et al., 1996) pioneered the jury compensation method as a means of quantifying the fear, pain, suffering, and reduced quality of life experienced by non-fatally injured victims of crime. Originally, damages awarded to accident victims in civil cases were used to approximate the monetary cost of similar injuries incurred by crime victims. However, the method assumes that injuries resulting from crime should be afforded the same weight as similar injuries resulting from incidents of a non-criminal nature. Intuitively it seems reasonable to assume that crime-victims are more likely to experience increased pain and suffering that is uniquely related to crime-specific psychological trauma (Cohen, 1999). Interestingly, recent litigation trends in the United States have obviated this problem. In 1996, Miller et al. analyzed 2,112 jury awards and settlements to assault, rape and burn victims. The lawsuits generally involved third party negligent suits such as a rape victim who sued the owner of an underground parking garage for poor lighting. Further, the analysis focused exclusively on the portion of the award designed to compensate the victim for pain, suffering, and lost quality of life.

Based on their analysis, the authors provide monetary quality of life estimates for serious felony offences (e.g., child abuse, sexual assault, assault, robbery, drunk driving, arson, property offences) by combining the jury information with tangible victim cost estimates obtained from survey data (e.g., NCVS). The authors estimate that the pain and suffering associated with rape for example, is valued at approximately $81,400.00 while the pain and suffering of being robbed is valued at $5,700.00. They do not provide estimates for white-collar crime, environmental crimes, Food and Drug Act violations, anti-trust breaches or crimes of treason. While tangible estimates are available for aggregate fraud losses (Titus, Heinzelmann, & Boyle, 1995), intangible estimates have yet to become available. Even Cohen (1999) concedes that it may be impossible to quantify all types of crimes, particularly those involving anti-trust violations or treason.

Proponents of this method argue that not only are jury awards stable and predictable but given that society has chosen the civil court system as an acceptable means of redressing victims, jury awards represent a reasonable proxy for assessing crime-related pain and suffering (Cohen, 1988, 1999). Further, the jury compensation method has been used outside of the criminal justice realm (e.g., Consumer Product Safety Commission, Zamula, 1987 cited in Cohen, 1998). As well, this method is perceived as being less subjective compared to it's predecessor which estimated the cost of victim pain and suffering based on public opinion surveys (Phillips & Votey, 1981; Sellin & Wolfgang, 1964). Lastly, criminal justice researchers have embraced rather than rejected the quantification of human life, pain and suffering both in theory (Gray, 1994; Kiessling, 1976) and in practice (Cohen, 1998; Rajkumar & French, 1997).

Discounting costs and benefits to present value

Before costs and benefits can be compared they must be adjusted to ensure an ‘apples to apples' comparison. More specifically, costs and benefits expected to accrue in the future must be converted into present dollar value before making a meaningful comparison (Cohen, 1998; Laplante & Durham, 1983; Levine, 1983; Nas, 1996). Suppose for example that you wish to purchase a new set of golf clubs. Further assume that it is the middle of winter and that your local sports store is having a golf equipment sale. Now ask yourself, how much must the original price be discounted or reduced by in order to make the purchase worth your while? Remember that you won't be able to use your new clubs until the spring. Here in lies the essence of discounting; a dollar spent today is not the same as a dollar received tomorrow. Future benefits must be discounted or depreciated to present value when compared with costs incurred today (Cohen, 1998). This principle perhaps explains why so many of us find it difficult to save money for our retirement when we would rather spend it on present-day activities such as holidays and entertainment. While discounting is a standard economic technique, the selection of an appropriate discount rate can be a rather dubious task.

There are no authoritative guidelines for choosing an appropriate discount rate. The criminal justice literature has used rates as low as 2% and as high as 15%. While prominent experts (Cohen, 1998, 1999) recommend 2% and 3% rates, government regulatory boards (e.g., U.S. Office of Management and Budget, 1996; Treasury Board of Canada, 1998) recommend considerably higher rates ranging between 7% and 10%. It is important to note that the lower the discount rate, the greater the value placed on benefits that accrue in the future. Conversely, higher discount rates deflate the value of future benefits. Further, discount rates can have a dramatic influence on one's results. As the Treasury Board of Canada (1998) observed, “most projects look good at a 5% discount rate and poor at a 15% discount rate.” Nonetheless, discounting future benefits or costs to present value is a mandatory requirement for any cost-benefit analysis that can not be avoided.

Comparing costs and benefits

The two most common methods for comparing costs and benefits are benefit-cost ratios and net present value. A benefit-cost ratio compares the present value of benefits to the present value of costs. Benefit cost ratios generate statements such as every dollar spent on program X generates $30.00 in benefits or economic returns. Benefit-cost ratios in excess of 1.00 are considered economically efficient. In contrast, the net present value is simply the present value of all benefits, discounted at the appropriate discount rate, minus the present value of all costs discounted at the same rate (Nas, 1996; Treasury Board of Canada, 1998). If the net present value is positive, the program is judged economically efficient. Benefit-cost ratios are beneficial in that they facilitate comparison across studies. However, they made be misleading given that different studies incorporate different benefits and hence will generate benefit-cost ratios that may vary more as a function of one's accounting perspective rather than the actual cost-effectiveness of the program in question (Weimer & Friedman, 1979).

Dealing with uncertainty: Sensitivity analysis

Efficiency evaluations rely extensively on a number of assumptions. Consequently, economists recommend that an efficiency analysis be accompanied by a sensitivity analysis (Nas, 1996; Rossi et al., 1999). Briefly, a sensitivity analysis requires that the researcher re-analyze the results across a number of different assumptions. If the results remain positive across a wide range of possibilities (e.g., varied discount rate, varied statistical value of life estimate, varied program effectiveness) one can be reasonably assured that the results are accurate.


Correctional program efficiency evaluations have proliferated substantially since they first emerged in the 1 960s. However, methodological shortcomings precluded earlier studies from reaching meaningful conclusions. More often than not, correctional researchers failed to describe cost estimation procedures in sufficient detail. Further, sensitivity analysis and discounting procedures were rarely used. As well, researchers often relied on inferior study designs. Earlier studies also excluded intangible program-related benefits given that quantification techniques had not yet been developed, hence grossly underestimating potential economic returns. Lastly, the lack of consistency in terms of which program-related benefits correctional researchers chose to examine precluded meaningful comparisons across studies.

Fortunately, the last decade has witnessed substantial progress in the quality of corrections-based efficiency evaluations. Consequently, this section highlights key efficiency findings from recent years that have focused specifically on correctional program outcomes. It should be noted that the review emphasizes methodologically superior studies.

Treatment versus punishment

Zedlewski (1985), a staff economist for the U.S. National Institute of Justice concluded that prisons are a highly cost-effective means of reducing crime. More specifically, he estimates that every dollar allocated towards imprisonment could generate $17.00 in tangible criminal justice savings. However, as others have aptly noted (Greenberg, 1990; Zimring & Hawkins, 1988) Zedlewski's conclusions are predicated on a number of faulty assumptions. For example, he incorrectly assumes that incarceration has a general deterrence effect. Critics (Greenberg, 1990) have clearly illustrated that the research in this area is less than reliable. Further, Zedlewski's cost saving estimates are based largely on the assumption that the crime rate will drop substantially if society incarcerates offenders who would have otherwise received probation or monetary fines. In essence, he argues that net widening5 will save society money. He arrives at this conclusion based on self-report crime data obtained from 2,190 incarcerated offenders who on average, reported committing 187 property offences annually. He erroneously assumes that individuals who have been sanctioned either by probation or monetary fines offend at the same frequency and intensity as individuals serving carceral sentences who undoubtedly are higher risk cases. Thus, he incorrectly concludes that the incarceration of one individual who would have otherwise been in the community under a probation order will prevent 187 property crimes from occurring.

Meta-analytic reviews provide the most persuasive evidence against the argument that punishment is economically efficient. The most recent review by Andrews et al. (2000) demonstrates that punishment is actually related to slight increases rather than decreases in recidivism. Thus, if punishment can not reduce recidivism it surely can not reduce future criminal justice costs. Additional evidence in support of treatment is available from recent research sponsored by the Washington State Institute for Public Policy.

Aos et al., (1999) analyzed 108 correctional treatment outcome studies using a cost-benefit approach. Traditionally, meta-analytic techniques have been used to aggregate the findings of a large number of treatment studies to ascertain the average impact that treatment has on reducing recidivism. This review represents the first attempt at estimating average cost-savings across various programs using meta-analytic techniques.

The authors focused on studies conducted during the last 20 years that compared recidivism rates between an experimental treatment group and a comparison group that did not receive treatment. Further, only methodologically superior studies were included. Cost savings were reported for several different treatment categories from two perspectives: the taxpayer and the victim. The taxpayer's perspective focused exclusively on criminal justice savings (e.g., police, adjudication, and corrections) while the victim's perspective incorporated criminal justice savings as well as monetary victim losses (e.g., medical and mental health care expenses, property damage and losses, reduced future earnings). Intangible victim costs such as pain and suffering and loss of life were excluded.

In sum, the review demonstrated that on average, every dollar spent on human-service orientated interventions (N = 88) saves the taxpayer approximately $5.00, and the victim, $7.00.6 Conversely, punishment-orientated interventions such as boot camps and intensive supervision programs that rely on expensive strategies such as random curfew checks, electronic monitoring, and urinalysis testing (N = 20) yielded substantially lower economic returns ranging from 50¢ to 75¢ for every dollar allocated to the program. Thus, human-service interventions satisfy standard economic efficiency threshold criteria while punitive interventions do not.

The inability of intermediate sanctions such as boot camps, intensive supervision, and electronic monitoring to demonstrate cost-effectiveness is not surprising. A recent meta-analysis that reviewed 20 studies that evaluated intermediate sanctions concluded that these programs have virtually no impact on recidivism (Gendreau, Goggin, Cullen, & Andrews, 2000). Thus, it is not surprising that they do not reduce criminal justice costs. Further, opponents have long argued that intermediate sanctions are actually more expensive than traditional forms of punishment due to net widening and close monitoring (Gendreau, Paparozzi, Little, & Goddard, 1993; Clear & Hardyman, 1990; MacKenzie & Parent, 1992; Tonry, 1990). For example, given the option, judges are more likely to sentence convicted offenders to boot camp or electronic monitoring who would have otherwise been sentenced to less costly, albeit more traditional forms of punishment such as probation. Further, the close monitoring that accompanies intensive supervision programs (ISP's) actually increases the probability of detecting and processing technical violations that would have otherwise gone unnoticed under regular supervision. Thus, additional costs associated with revoking and eventually incarcerating the offender are incurred. Lastly, critics (Tonry, 1990) have argued that diversion programs such as ISP's do not generate the substantial cost-savings that supporters purport due to an over reliance on fixed rather than marginal costs. As Tonry (1990) aptly observes the savings incurred by diverting one offender from prison in reality are quite small, amounting to no more than a bit of food and record keeping. Large cost savings will only materialize if and when diversion programs either cause existing prisons to close or prevent new prisons from being built.

Juvenile offender treatment

Promising findings have resulted from three relatively recent juvenile efficiency evaluations. Greenwood, Model, Rydell, and Chiesa (1996) conducted a cost-effectiveness comparison of California's three-strikes law versus early intervention programs. Two of the most promising intervention programs included graduation incentives and parent skills training. Graduation-incentive programs financially compensate disadvantaged high school students to encourage graduation. Alternatively, parenting skills programs teach parents how to deal effectively with aggressive children. The study estimates that California's three-strikes law will reduce crime by 21% at an annual increased incarceration cost of 5.5 billion dollars. However, graduation incentive programs coupled with parenting skills training could approximately double the crime reduction rate for 1/5 of the cost.

Aos et al., (1999) reviewed 21 human-service orientated juvenile treatment programs including parent skills training, diversion programs, and aggression replacement training. The results indicated that each juvenile treatment dollar will generate between $7.62 to $31.40 in future economic returns. Interestingly, juvenile offender treatment generated the highest benefit-cost ratios outperforming both adult offender treatment and primary prevention strategies.7

Cohen (1998) estimates that one chronic juvenile offender will cost victims and society between 1.3 and 1.5 million dollars in the long run. The majority of which will be associated with intangible victim costs (50%) followed by tangible victim costs (25%), criminal justice expenses (20%), and foregone offender productivity (5%). This implies that relatively small treatment effects could generate substantial cost savings. For example, a program that costs $500,000 to treat 100 chronic juvenile offenders would still be deemed cost-effective with a success rate as low as 1%. However, in reality success rates are substantially higher, particularly for innovative juvenile treatment programs such as multisystemic therapy (MST). Aos et al. (1999) reported that MST generates $13.45 dollars in returns for every invested program dollar. Interestingly, however, if one adopts Cohen's (1998) 1.3 to 1.5 million dollar estimate, MST could potentially generate $60.00 in economic returns for every program dollar.8 The discrepancy ($13.45 vs. $60.00) is most likely attributable to the fact that Aos et al. excluded intangible victim costs whereas intangible victim costs accounted for 50% of Cohen's 1.3 to 1.5 million-dollar estimate.

Sex offender treatment

The literature has produced conflicting results regarding the cost-effectiveness of sex offender treatment. For example, Aos et al., (1999) recently concluded that sex offender treatment is not cost-effective. Based on a review of six treatment outcome studies the authors report that every dollar allocated towards sex offender treatment yields no more than 25¢ in economic returns. However, more promising conclusions have been reached independently by Canadian, American and Australian researchers.

Marshall (1992) concludes that by treating 100 sex offenders, Canadian society will not only save 50 potential victims but that it will also save 4.4. million dollars in averted criminal justice costs. Marshall's estimates were based on the assumption that sex offender treatment has a 25% success rate and that the typical sex offender recidivist will commit at least two new sexual offences. Prentky and Burgess (1990) conducted a cost-benefit analysis of a program that had been treating child molesters in Massachusetts since 1959. Like Marshall, they concluded that the treatment of sexual offenders can result in substantial cost savings, specifically, they estimate that for every 100 treated sex offenders society saves 2.7 million dollars in averted criminal justice expenses. However, it should be noted that Prentky and Burgess did not have access to a random or matched control group. Instead, they used information obtained from Marshall who estimated the recidivism rate for untreated sex offenders based on his own Canadian-based research.

While Prentky and Burgess conclude that sex offender treatment can be cost-effective the absence of a reliable control group precludes confidence in the results. Similarly, critics (Quinsey, Harris, Rice, & Lalumière, 1993) would likely argue that Marshall's 25% success rate is grossly inflated. Both studies did not conduct detailed sensitivity analyses nor did they use discounting procedures or include victim-related benefits. However, recent findings from Donato and Shanahan (1999), and Hanson (personal communication, 2000) have addressed some of this issues.

Donato and Shanahan (1999) conducted a cost-benefit analysis of an intensive, sex offender treatment program delivered in Australian prisons to child molesters. The authors demonstrate that treating child molesters prior to release can be cost-effective provided that reductions in recidivism exceed 6% and that both tangible and intangible victim-related benefits are included. The authors further illustrate that a 14% reduction in sexual recidivism can generate almost 4 million dollars in economic returns for every 100 child molesters that are treated. In contrast, sex offender treatment will not generate positive economic returns if reductions in recidivism are less than 2%, regardless of whether or not tangible or intangible benefits are included. Donato and Shanahan's research represents one of the first attempts to incorporate tangible as well as intangible victim benefits into a sex offender efficiency evaluation.

A recent meta-analytic review of 42 sex offender treatment outcome studies involving 9,316 sex offenders demonstrated that on average, untreated sex offenders reoffend at rate almost twice as high (17.7%) as their treated counterparts (9.9%) (Hanson, personal communication, 2000). Thus, extrapolating from Donato and Shanahan's work, one can infer that the typical, contemporary sex offender program will generate positive economic returns provided that intangible victim-related benefits are considered. More specifically, the treatment of 100 sex offenders will save society between $41,6000 and 1.85 million dollars. However, primary efficiency evaluations involving well-controlled sex offender outcome studies remain noticeably absent. Thus, most individual jurisdictions can not state with certainty in terms of whether or not their own programs are cost-effective.

Education and employment

To date, 21 employment and/or education programs have been evaluated within a cost-benefit framework (see Aos et al., 1999; Knox & Stacey, 1978; Mallar & Thornton, 1978; Seashore et al., 1976). However, the majority were conducted by a secondary group of researchers (e.g., Aos et al., 1999). In sum, most employment and/or education programs generate positive economic returns. More specifically, every dollar allocated towards vocational and basic education programs yields cost savings ranging from $1.71 to $3.23. Similarly, job search and/or counselling programs generate positive returns ranging from $2.84 to $6.56. Mallar and Thornton conclude that one program dollar could generate returns as high as $53.73. Conversely, short-term financial assistance and subsidized job placements programs generate break even returns (e.g., 1 dollar spent = 1 dollar gained). Additionally, while Friedman (1977) concludes that work-release programs generate positive economic returns (e.g., $1 .64), Turner and Petersilia (1996) concluded otherwise.

Substance abuse treatment

The efficiency literature has paid considerable attention to sub-stance abuse treatment (Apsler, 1991; Britt, Gottfredson & Goldkamp, 1992; Goldschmidt, 1976; Hertzman & Montague, 1977; Holder, 1987; Leukefeld, Logan, Martin, Purvis, & Farbaee, 1998; Swint & Nelson, 1977). While drug diversion programs have generated modest returns (e.g., $1.69 to $2.18 for every program dollar), interventions classified as case management substance abuse programs have generated negative returns, such that every program dollar actually costs the taxpayer 15¢, and the victim 21¢ (Aos et al., 1999).

However, more encouraging findings are also available. For example, a recent Canadian study demonstrated that one of the Correctional Service of Canada's core substance abuse treatment programs generated approximately $2,000 in annual savings per offender (T3 Associates, 1999). Similarly, research conducted on substance abusers, rather than criminal offenders suggests that for every 100 substance abusers that are treated, society accrues between 1.4 and 2.2 million dollars in cost savings. Cost savings associated with the provision of substance abuse treatment have included criminal justice expenses, tangible and intangible victim losses, as well as offender foregone productivity (Rajkumar & French, 1997).


In sum, the available empirical literature presents a convincing argument that evidence-based correctional treatment is cost-effective. However, the review also illustrates the absence of well-controlled primary studies. Further, standard operating practices that outline common ingredients required of all corrections-based efficiency evaluations should be established (Welsh & Farrington, 2000). Thus, regardless of which program is being evaluated, one can argue that all evaluations should include pro-gram costs, criminal justice expenses, tangible and intangible victim costs, and forgone offender productivity. The development of such standards is necessary to promote consistency and hence, comparability across studies. The importance of this objective is made acutely clear by Eddy's (1992) statement, “much of cost-effectiveness analysis parallels clinical judgement” (p. 3344). This observation is particularly distressing given the repeatedly poor performance of human judges tasked with the responsibility of making predictions about future behaviour (Grove, Zald, Lebow, Snitz, & Nelson, 2000).

Critics argue that efficiency evaluations are subject to a “garbage in, garbage out phenomena”. Regardless of how technically accurate an efficiency evaluation may be, the findings will be rendered unreliable if the primary impact evaluation was sufficiently flawed. Similarly, cost-benefit analyses rely extensively on uncertain assumption and at times, less then reliable cost estimates. Further, attributing monetary value to human pain and suffering as well as human life remains controversial. Lastly, the level of technical expertise required to conduct such evaluations may simply be inaccessible to some jurisdictions.

Clearly, it is difficult to justify maintaining programs that have no impact. However, even programs with a demonstrated impact may be difficult to maintain given political climate, public opinion or if limited resources intensifies competition among various programs. Further, efficiency analyses can enhance informed policy decision-making, reduce the costs of obtaining key objectives, and promote efficient resource allocation. While economic arguments may be insufficient to affect policy change given that the average person is more readily influenced by programs that are marketed as “state of the art” as well as programs that affect real people rather than anonymous statistical cases (Eddy, 1992) they can help bridge the gap between research and practice.

1 Correctional Service of Canada

2 When a victim is unable to work due to a crime-related injury two distinct losses may occur: victim losses (lost wages for unpaid workdays) and societal losses (foregone productivity) (Miller et al., 1996).

3 Given that offender treatment programs promote change at the individual rather than societal level, global changes in societal behaviour that can be directly linked to any one specific program are viewed as secondary or indirect effects, albeit positive.

4 In the present model an offender's income after taxes is considered a direct benefit to the offender, while the taxable component is considered a direct benefit to society.

5 Net widening refers to the practice of meting out hasher penalties (e.g. boot camps) to individuals who would have otherwise received less costly, traditional forms of punishment (e.g. probation) if the alternative (e.g. boot camp) had not been available to the sentencing judge.

6 Adult cognitive-behavioural treatment programs generate economic returns ranging from $2.54 to $11.48 for every invested program dollar.

7 Primary prevention programs target children who have had no formal involvement with the criminal justice system but are considered high risk for future contact (e.g., Big Brothers).

8 This latter estimate is based on the assumption that MST has a 20% success rate (Henggeler, Melton, & Smith, 1992)


Andrews, D. A., & Bonta, J. (Eds.), (1998). The psychology of criminal conduct. Cincinnati, OH: Anderson Publishing Co.

Andrews, D. A., Dowden, C., & Gendreau, P. (2000). Clinically relevant and psychologically informed approaches to reduced reoffending: A meta-analytic study of human service, risk, need, responsivity and other concerns in justice contexts. Manuscript submitted for publication.

Andrews, D. A., Zinger, I., Hoge, R. D., Bonta, J., Gendreau, P., & Cullen, F. T. (1990). Does correctional treatment work? A psychologically informed meta-analysis. Criminology, 28, 369-404.

Aos, S., Phipps, P., Barnoski, R., & Lieb, R. (1999). The comparative costs and benefits of programs to reduce crime: A review of national research findings with implications for Washington State. Washington, DC: Washington State Institute for Public Policy.

Apsler, R. (1991). Evaluating the cost-effectiveness of drug abuse treatment services. In Economic Costs, Cost-effectiveness, Financing, and Community-Based Drug Treatment, National Institute of Drug Abuse Research Monograph Series 113 (pp. 57-66), Rockville, MD: U.S. Department of Health and Human Services.

Britt, C. L., Gottfredson, M. R., & Goldkamp, J. S. (1992). Drug testing and pre-trial misconduct: An experiment on the specific deterrent effects of drug monitoring defendants on pre-trial release. Journal of Research in Crime and Delinquency, 29(1), 62-78.

Bureau of Justice Statistics (1998). Criminal Victimization 1997 (Report NCJ-173385) Washington, DC: U.S. Department of Justice.

Burt, M. R., & Katz, B. L. (1985). Rape, robbery, and burglary: Responses to actual and feared victimization, with special focus on women and the elderly. Victimology: An Internal Journal, 10, 325-358.

Clear, T. R., & Hardyman, P. L. (1990). The new intensive supervision movement. Crime & Delinquency, 36, 42-60.

Cohen, M. A. (2000). Measuring the costs and benefits of crime and justice. In D. Duffee (Ed.), Measurement and analysis of crime and justice, criminal justice 2000, 4, (pp. 263-315). Rockville, MD: National Institute of Justice/NCJRS.

Cohen, M. A. (1998). The monetary value of saving a high-risk youth. Journal of Quantitative Criminology, 14(1), 5-33.

Cohen, M. A. (1988). Pain, suffering, and jury awards: A study of the cost of crime to victims. Law and Society Review, 22(3), 537-555.

Cohen, M. A., Miller, T. R., & Rossman, S. B. (1994). The costs and consequences of violent behavior in the United States. In A. J. Jr. Reiss & J. A. Rother (Eds.), Consequences and control of understanding and preventing violence (pp. 67-166). Washington, DC: National Research Council, National Academy Press.

Donato, R., & Shanahan, M. (1999). The economics of implementing intensive in-prison sex-offender treatment programs. In Trends and issues in crime and justice, 134, (pp. 1-6). Canberra, Australia: Australian Institute of Criminology.

Eddy, D. M. (1992). Cost-effectiveness analysis: Is it up to the task? Journal of American Medical Association, 267(24), 3342-3348.

Elliott, D. S. (1980). Recurring issues in the evaluation of delinquency prevention and treatment programs. In D. Schichor & D. H. Kelly (Eds.), Critical issues in juvenile delinquency (pp. 237-262). Lexington, KY: D. C. Health and Company.

Friedman, L. S. (1977). An interim evaluation of the supported work experiment. Policy Analysis, 3(2), 147-170.

Gendreau, P., Goggin, C., Cullen, F. T., & Andrews, D. A. (2000). The effects of community sanctions and incarceration on recidivism. Forum on Corrections Research, 12(2), 11-13.

Gendreau, P., Paparozzi, M., Little, T., & Goddard, M. (1993). Does “Punishing Smarter” Work? An assessment of the new generation of alternative sanctions in probation. Forum on Corrections Research, 5(3), 31-34.

Goldschmidt, P. G. (1976). A cost-effectiveness model of evaluating health care programs: Application to drug abuse treatment. Inquiry, XIII, 29-47.

Gray, C. M. (Ed) (1979). The costs of crime. Beverly Hills, CA: Sage Publications.

Gray, T. (1994). Research Note: Using cost-benefit analysis to measure rehabilitation and special deterrence. Journal of Criminal Justice, 22(6), 569-575.

Greenberg, D. F. (1990). The cost-benefit of imprisonment. Social Justice, 17(4), 49-75.

Greenwood, P. W., Model, K. E., Rydell, C. P., & Chiesa, J. (1996). Diverting children from a life of crime: Measuring costs and benefits. Santa Monica, CA: RAND Corporation.

Grove, W. M., Zald, D. H., Lebow, B. S., Snitz, B. E., Nelson, C. (2000). Clinical versus mechanical prediction: A meta-analysis. Psychological Assessment, 12, 19-30.

Henggeler, S. W., Melton, G. B., & Smith, L. A. (1992). Family preservation using multisystemic therapy: An effective alternative to incarcerating serious juvenile offenders. Journal of Consulting and Clinical Psychology, 60(6), 953-961.

Hertzman, M., & Montague, B. (1977). Cost-benefit analysis and alcoholism. Journal of Studies of Alcohol, 38(7), 1371-1384.

Holahan, J. (1973). Measuring benefits from prison reform. In Benefit-cost and policy analysis. An Aldine annual on forecasting, Decision-making and evaluation (pp. 491-516). Chicago, IL: Aldine.

Holder, H. D. (1987). Alcoholism treatment and potential health care cost saving. Medical care, 25, 52-71.

Kiessling, J. J. (1976). A cost-effectiveness analysis of the Ottawa Volunteer Program. Ottawa, ON: Ontario Ministry of Correctional Services.

Knox, G. W., & Stacey, W. A. (1978). Determinants of employment success among ex-offenders. Offender Rehabilitation, 2(3), 205-214.

Laplante, J. M., & Durham, T. R. (1983). An introduction to benefit-cost analysis for evaluating public expenditure alternatives. (Available from the Policy Studies Associates, P.O. Box 337, Croton-on-Hudson, NY 10520).

Laurence, L., & Spalter-Roth, R. (1996). Measuring the costs of dynamic violence against women and the cost-effectiveness of intervention: An initial assessment and proposals for further research. (Available from the Institute for Women's Policy Research, 1400 20th Street, NW, Suite 104, Washington, DC 20036).

Leukefeld, C. G., Logan, T. K., Martin, S. S., Purvis, R. T., & Farabee, D. (1998). A health services use framework for drug-abusing offenders. American Behavioral Scientist, 41(8), 1123-1135.

Levine, H. M. (1983). Cost-effectiveness: A primer. New perspectives in evaluation, 4. Beverly Hills, CA: Sage Publications.

Lipsey, M. W. (1992). Juvenile delinquency treatment: A meta-analytic inquiry into the variability of effects. In R. S. Cook, H. Cooper, D. S. Cordray, H. Hartmann, L. V. Hedges, R. J., Light, T. A. Louis, & F. Mosteller (Eds.), Meta-analysis for explanation (pp. 83-125). New York, NY: Russell Sage.

Lösel, F. (1995). The efficacy of correctional treatment: A review and synthesis of meta-evaluations. In J. McGuire (Ed.), What works: Reducing reoffending: Guidelines from research and practice (pp. 79-111). Chichester, UK: Wiley.

Lösel, F. (1996). Effective correctional programming: What empirical research tells us and what it doesn't. Forum on Corrections Research, 8(3), 33-37.

MacKenzie, D. L., & Parent, D. G. (1992). Boot camp prisons for young offenders. In J. M. Byrne, A. J. Lurigio, & J. Petersilia (Eds.), Smart sentencing: The emergence of intermediate sanctions (pp. 103-119). Newbury Park, CA: Sage Publications.

Mallar, C. D., & Thornton, C. V. (1978). Transitional aid for release prisoners: Evidence from the life experiment. The Journal of Human Resources, X111(2), 208-236.

Marshall, W. L. (1992). The social value of treatment for sexual offenders. The Canadian Journal of Human Sexuality, 1(3), 109-114.

Miller, T. R. (1990). The plausible range for the value of life: Red herrings among the mackerel. Journal of forensic economics, 3(3), 75-89.

Miller, T. R., Cohen, M. A., & Wiersema, B. (1996). Victim costs and consequences: A new look (NCJ-1 55282). Washington, DC: National Institute of Justice.

Morris, E. K., & Braukmann, C. J. (Ed.) (1987). Behavioral approaches to crime and delinquency: A handbook of application, research, and concepts. New York, NY: Plenum Press

Nas, T. F. (1996). Cost-benefit analysis: Theory and application. Thousand Oaks, CA: Sage Publications.

Phillips, L., & Votey, H. L. (1981). The economics of crime control. Beverly Hills, CA: Sage Publications.

Prentky, R., & Burgess, A. W. (1990). Rehabilitation of child molesters: A cost-benefit analysis. American Journal of Orthopsychiatry, 60, 108-117.

Quinsey, V. L., Harris, G. T., Rice, M. E., & Lalumière, M. L. (1993). Assessing treatment efficacy in outcome studies of sex offenders. Journal of Interpersonal Violence, 8, 512-523.

Rajkumar, A. S., & French, M. T. (1997). Drug abuse, crime costs, and the economic benefits of treatment. Journal of Quantitative Criminology, 13(3), 291-323.

Rossi, P. H., Freeman, H. E., & Lipsey, M. W. (1999). Evaluation: A systematic approach, 6th edition, Thousand Oaks, CA: Sage Publications.

Seashore, M. J., Haberfeld, S., Irwin, J., & Baker, K. (1976). Prisoner education: Project Newgate and other college programs. New York, NY: Praeger Publishers.

Sellin, T., & Wolfgang, M. E. (1964). The measurement of delinquency. New York, NY: Wiley.

Swint, J. M., & Nelson, W. B. (1977). Prospective evaluation of alcoholism rehabilitation efforts: The role of cost-benefit and cost-effectiveness analyses. Journal of Studies on Alcohol, 18(7), 1386-1404.

T3 Associates (1999). An outcome evaluation of CSC substance abuse programs: OSAPP, ALTO and Choices-Executive Summary (Available from Reintegration Programs, Correctional Service of Canada, 340 Laurier Avenue West, Ottawa, ON, K1A 0P9).

Titus, R., Heinzelmann, M. F., & Boyle, J. M. (1995). Victimization of persons by fraud. Crime and Delinquency, 41, 54-72.

Tonry, M. (1990). Stated and latent functions of ISP. Crime & Delinquency, 36, 174-191.

Treasury Board of Canada (1976). Benefit-cost analysis guide (Catalogue No. BT35-2-1976). Ottawa, ON: Minister of Supply and Services Canada.

Treasury Board of Canada (1998). DRAFT Benefit-cost analysis guide. Ottawa, ON: Author.

Turner, S., & Petersilia, J. (1996). Work release in Washington: Effects on recidivism and correctional costs. The Prison Journal, 76(2), 138-164.

United States Office of Management and Budget (1989). Regulatory Program of the United States. Washington, DC: U.S. Government Printing Office.

United States Office of Management and Budget (1996). Economic analysis of federal regulations under Executive Order 12866. Washington, DC: Author.

Weimer, D. L., & Friedman, L. S. (1979). Efficiency considerations in criminal rehabilitation research: Costs and consequences. In L. Sechrest, S. O. White, & E. D. Brown (Eds.), The rehabilitation of criminal offenders: Problems and prospects. Washington, DC: National Academy of Sciences.

Welsh, B. C., & Farrington, D. P. (2000). Correctional intervention programs and cost-benefit analysis. Criminal Justice and Behavior, 27(1), 115-133.

Viscusi, W. K. (1993). The value of risk to life and health. Journal of Economic Literature, 31, 1912-1946.

Yates, B. T. (1998). Toward the incorporation of costs, cost-effectiveness analysis, and cost-benefit analysis into clinical research. Journal of Consulting and Clinical Psychology, 62, 729-736.

Zamula, W. (1987). Social costs of drownings and near-drownings from submersion accidents occurring to children under five in residential swimming pools. (Available from Directorate for Economic Analysis, U.S. Consumer Product Safety Commission).

Zedlewski, E. W. (1985). When have we punished enough? Public Administration Review, 45, Special Issue, 771-779.

Zerbe, R. O. (1998). Is cost-benefit analysis legal? Three rules. Journal of Policy Analysis and Management, 17(3). 419-456.

Zimring, F. E., & Hawkins, G. (1988). Incapacitation: Penal confinement and the restraint of crime. New York, NY: Oxford University Press.


Previous PageTop Of Page Table Of ContentsNext Page