FEDERAL, STATE, AND LOCAL GOVERNMENTS collectively spend over $260 billion annually for criminal justice, an expenditure that has increased by 210 percent since 1982. Much of the financial burden falls on local government. In 2010, city and county governments contributed over 50 percent of total direct criminal justice expenditures. The bulk of law enforcement spending originates at the local level (67 percent): nearly 40 percent of judicial/legal expenditures and one-third of corrections expenditures are local. State government contributes under one-third (30.5 percent) of total justice expenditures and 57 percent of corrections costs.
While state-level corrections costs constitute approximately 2.5 to 3 percent of total state expenditures, the relative burden at the local level is much higher. Jurisdictions differ in terms of how much of local criminal justice activities are funded by city, county, or state revenue. Overall, law enforcement (police) is paid out of municipal budgets. Sheriff’s agencies and the courts are typically funded by counties, as is local corrections (jail, pretrial, and probation). Local police generally consume a minimum of one-third of municipal budgets. Courts and other judicial services require between 35 and 50 percent of county budgets.
Regardless of the perspective—federal, state or local—the United States spends an extraordinary amount of public tax revenue on the administration of criminal justice. The $260 billion annual expenditure is higher than the gross domestic product of 80 percent of the nations in the world.
The question prudent policymakers, elected officials, and taxpayers should ask is: What is the return on investment. Has this been a cost-effective use of public resources? Can we develop a more cost-effective strategy going forward? But first, an example.
I have served several times as a consulting expert for inmates being considered for discretionary parole release. One case in particular illustrates many things about the justice system that are problematic, including lack of informed decision making, poor judgment, and wasteful expenditure of public resources. This case involves a sex offender (call him Mr. Davis) who in the 1970s abducted a teenage boy and sexually assaulted him. He was convicted and sentenced to prison. Davis served ten years and was released on parole supervision. Two years into his parole, he abducted another boy and molested him. Davis was convicted on the new charge and sentenced to twenty-five years in prison. I got involved in this case when Davis was completing year twenty-two of the twenty-five-year sentence. At this time, Davis was seventy-three years old, impudent, arthritic, and in poor health. He had been airlifted from the prison to a hospital on several occasions for life-threatening medical emergencies. His parole plan had him moving to another state to live on a farm with his nephew, who was a retired U.S. Marshall and an active deputy sheriff. His psych exam indicated no major concerns and he acknowledged responsibility and remorse for his crimes. While it is clear that Mr. Davis committed some bad crimes, it was clear to me that he had aged out of crime. My estimate of his risk was essentially zero. The risk assessment used by the parole board was a static assessment and, among other problems, only considered age in terms of whether the applicant is under or over age twenty-five. In essence, an offender in this system would have the same assessed risk at age ninety-six as at age twenty-six. The parole board showed no appreciation for how this assessment is inappropriate. Needless to say, Mr. Davis was denied parole release. This example illustrates tough on crime decision making gone haywire. Davis posed no threat to the public, but the parole board wanted to be tough on crime. In addition, they relied on a faulty assessment instrument and ignored dispositive contrary information. Unfortunately, this case is not the exception. The primary point here is that the individuals responsible for the decision to keep Davis in prison have no accountability for the financial cost of their decision.
I start by assessing the financial impact of crime. Table 8.1 presents average estimates of the criminal justice cost of selected offenses. These criminal justice cost estimates were developed by McCollister, French, and Hai (2010), based upon an extensive review and analysis of a variety of crime costing methods and unit costing studies that generate estimates for particular types of crimes. The criminal justice costs associated with each of the thirteen offenses included in the McCollister, French, and Hai research reflect a law enforcement component, a legal and adjudication component, and a corrections component. Keeping in mind that there are several assumptions that are employed in the estimation procedure, these are estimates, and as such there is error, I can cautiously extrapolate the overall criminal justice costs. Since the legal/adjudication and corrections costs are postoffense, I use 2010 arrest data from the Uniform Crime Reports (UCR) for 2010 to determine the number of arrests for each crime type. Multiplying the cost per offense by the number of arrests provides the aggregate criminal justice cost estimate. These thirteen offenses, which constituted only 21 percent of all 2010 arrests reported in the UCR, cost federal, state, and local governments over $18 billion in 2010. Eighteen billion dollars is quite a sum of money, but pales in comparison to the $260 billion in total annual criminal justice expenditures.
Table 8.2 provides estimates (from McCollister, French, and Hai 2010) associated with tangible crime costs and intangible crime costs for the selected thirteen offenses. Tangible cost estimates are based on the criminal justice costs, victim costs, and loss of productivity of perpetrators committing the crimes. The intangible costs include pain and suffering (McCollister, French, and Hai 2010). The total tangible costs for each offense are based on 2010 arrests. The total intangible costs (victim costs, among others) are based on National Crime Victimization Survey estimates of crimes committed in 2010. Taking this broader view of the cost of crime is rather sobering. These estimates indicate that the overall financial impact of these thirteen crimes is nearly $200 billion annually. Seeing the financial impact on victims in terms of direct costs as well as the intangibles of the emotional and physical harm raises the stakes in seeking more effective and cost-effective crime solutions. Again, if the criminal justice system more effectively reduced recidivism, many victimizations and much of the cost of victimization could be avoided.
TABLE 8.1 The Direct Criminal Justice Cost of Crime
| OFFENSE |
CJS COST |
2010 ARRESTS |
TOTAL |
| Murder |
$392,350 |
11,201 |
$4,400,000,000 |
| Rape/sexual assault |
$26,480 |
20,088 |
$532,000,000 |
| Aggravated assault |
$8,640 |
408,488 |
$3,500,000,000 |
| Robbery |
$13,830 |
112,300 |
$1,553,000,000 |
| Motor vehicle theft |
$3,870 |
71,487 |
$277,000,000 |
| Arson |
$4,390 |
11,296 |
$49,590,000 |
| Burglary |
$4,130 |
289,769 |
$1,200,000,000 |
| Larceny/theft |
$2,880 |
1,271,410 |
$3,660,000,000 |
| Stolen property |
$6,842 |
94,802 |
$649,000,000 |
| Vandalism |
$4,160 |
252,753 |
$1,051,000,000 |
| Forgery/counterfeit |
$4,605 |
78,101 |
$360,000,000 |
| Embezzlement |
$4,820 |
16,616 |
$80,000,000 |
| Fraud |
$4,372 |
187,887 |
$821,400,000 |
| Annual total: |
|
|
$18,132,990,000 |
TABLE 8.2 The Tangible and Intangible Costs of Crime
| OFFENSE |
TOTAL TANGIBLE COST |
TOTAL INTANGIBLE COST |
TOTAL |
| Murder |
$14,400,000,000 |
N/A |
$14,400,000,000 |
| Rape/sexual assault |
$829,000,000 |
$48,700,000,000 |
$49,529,000,000 |
| Aggravated assault |
$7,954,000,000 |
$99,971,000,000 |
$107,925,000,000 |
| Robbery |
$2,400,000,000 |
$12,570,000,000 |
$14,970,000,000 |
| Motor vehicle theft |
$753,000,000 |
$165,000,000 |
$918,000,000 |
| Arson |
$186,000,000 |
N/A |
$186,000,000 |
| Burglary |
$1,800,000,000 |
$1,160,000,000 |
$2,960,000,000 |
| Larceny/theft |
$4,480,000,000 |
$120,825,000 |
$4,600,825,000 |
| Stolen property |
$758,400,000 |
N/A |
$758,400,000 |
| Vandalism |
$1,228,000,000 |
N/A |
$1,228,000,000 |
| Forgery/counterfeit |
$411,000,000 |
N/A |
$411,000,000 |
| Embezzlement |
$91,000,000 |
N/A |
$91,000,000 |
| Fraud |
$945,000,000 |
N/A |
$945,000,000 |
| Annual total: |
|
|
$198,922,225,000 |
Criminal Careers
While the cost of crime analysis is quite illuminating, it is even more compelling when we consider that the majority of offenders are repeat offenders. In effect, each time we investigate and arrest a suspect, place him or her in custody, pass the case to the prosecutor’s office, hold a preliminary hearing or magistration, engage pretrial services, detain some in jail pretrial, provide legal counsel (for roughly 80 percent of felony defendants), indict the case, dispose of the case in court at a plea hearing, hold a sentencing hearing, and then transfer the offender to corrections, costs are incurred. A felony arrest may cost $150 to $200, taking into consideration transportation of the suspect, police time booking the individual, and preparation of the incident/arrest report. Typical booking costs $175. Detention in jail costs $45 per day (booking and detention data are for Travis County [Austin] Texas).
What most of the cost analyses and cost-benefit studies fail to incorporate into their models is the impact of lifetime offending and desistence. There is good reason for this omission because the assumptions required produce a considerable amount of error in the forecasts. We do not have very precise data on aggregate criminal career trajectories. Clearly, the amount of repeat offending depends on a wide variety of factors, including age, gender, type of offenses (violent, property, drug), education, marital status, employment, among others. However, we can generate a reasonable, informed estimate of what this looks like in the aggregate. The following is based on a personal correspondence with Al Blumstein at Carnegie Mellon University. Best estimates indicate the following. Assume a cohort of 100 individuals who were arrested in 1990. Within an approximate two-year window, 60 percent of them have been rearrested. There is clearly an important role here for the efficiency of law enforcement; this procedure is silent regarding crimes committed that are either unknown to law enforcement or known but not leading to an arrest. The remaining 40 percent stay arrest free into the future. Of the 60 percent who were arrested a second time, approximately 70 percent of them (42 of the original 100) are arrested for a third time within twenty or twenty-one months. The next round involves roughly 80 percent of those with a third arrest (33 of the original 100) who are arrested again within nineteen to twenty months of the prior arrest. The pattern then persists with around 80 percent of each prior arrest group (26 of the original 100) being rearrested in a slowly diminishing time window. A minority of offenders commit the majority of crimes. Moreover, this smaller group of high-risk, high-criminogenic-need habitual offenders is responsible for most of the tangible and intangible costs of crime. The point is simple. Our failures to interrupt the cycle of reoffending that is all too typical of criminal offenders causes the needless expenditure of phenomenal amounts of public funds.
Costs and Benefits Over Time
The costs and benefits of policies and programs weigh heavily in the policy development process. Unfortunately, in the majority of states, policymakers and elected officials are not properly informed of the fiscal impacts of policy changes. Poorly performed cost-benefit or cost-impact analyses fail to provide the true financial effects of criminal justice reforms over the long term. Research conducted by the Center on Budget and Policy Priorities and the American Civil Liberties Union (Leachman, Chettiar, and Geare 2012) reveals a substantial lack of proper financial analysis of proposed criminal justice legislation. The ACLU study examined the fiscal notes for over 600 legislative bills on sentencing and corrections in forty-nine states that were enacted in 2009, 2010, and 2011. Fiscal notes are the financial assessments of pending legislation.
Leachman, Chettiar, and Geare (2012) report that for about 40 percent of the 600 bills, the states did not even prepare fiscal notes, thus they lacked an official certification regarding the fiscal impact of the legislation. Moreover, for those states that did prepare fiscal notes, the majority of the analyses of fiscal impact failed to assess that impact beyond two years. Without an assessment of the longer-term financial impacts, legislators are unable to appreciate the longer-term fiscal benefits of such reforms. Many criminal justice reforms that target recidivism will, by design, require time to realize their impact. Failure to provide a financial analysis, or focusing the financial analysis on just one to two years, is clearly inappropriate for capturing a realistic, longer-term economic impact. Leachman, Chettiar, and Geare (2012: 2) conclude:
the vast majority of states do not accurately perform these fiscal notes in a way that is useful to legislators. … By following these best practices, states can calculate both short- and long-term costs and cost savings for criminal justice bills and easily provide that information to legislators, who can then know the true fiscal implications of reform proposals before voting on them. … Improved fiscal notes will illuminate the short- and long-term benefits of ending our dependence on incarceration, bolster reforms that will reduce prison spending and save states millions that they can spend on other vital services.
THE COSTS AND BENEFITS OF INCARCERATION
Incarceration is the most expensive form of corrections. The per-inmate costs vary from $22,000 to $50,000 and more depending on jurisdiction, custody level, and other circumstances, such as physical health and age. On top of that, most states have funding obligations that figure into the cost of prison that are routinely not reported in their corrections budgets. On average, an additional 14 percent of prison costs excluded from reported corrections costs are due to obligations for employee benefits, pension contributions, retiree healthcare contributions, capital costs, legal judgments and claims, and hospital care for inmates, among others. Illinois leads the nation in prison costs excluded from the corrections budget (33 percent), followed by Iowa (26 percent) and Texas (24 percent).
It should be a fairly safe conclusion that given the relatively high cost of incarceration, high recidivism rates, and the marginal impact of incarceration on crime rates, incarceration alone as a correctional strategy is generally cost ineffective. Let’s look at the evidence.
A study conducted by the Pew Center on the States (2012) focuses on the cost- effectiveness of increased punishment severity over the past two decades, measured as increased time served. The research is based on release cohorts from Michigan, Florida, and Maryland. Since 1990, the growth in time served was relatively consistent by type of crime: drug crimes (36 percent increase), property crimes (up 24 percent), and violent crimes (up 37 percent). The question posed in the Pew study is: What was the cost and impact on public safety of these increases? Based on a risk analysis of individuals released from prisons in the three states included in the research, they found that 14 percent of releasees in the Florida cohort, 18 percent of the releasees in Maryland, and 24 percent in Michigan could have been released between three months and twenty-four months earlier without any negative impact on public safety. The extra sentence length cost $54 million in Florida, $30 million in Maryland, and $92 million in Michigan.
There has been a fair amount of attention in recent years on the marginal utility of incarceration. Diminishing returns of incarceration refers to the public safety benefit of continuing to imprison larger numbers of offenders. The logic is clear: locking up the most dangerous and highest-risk offenders can have a significant public safety impact; however, as larger and larger numbers of offenders are incarcerated, increasingly lower-level offenders who are lower risk enter the inmate pool, thus in turn, the public safety payoff diminishes. Washington State research shows the effect: in 1980, the number of crimes committed by the average inmate was sixty-two; it dropped to thirty-seven in 1990 and to eighteen in 2001. As noted in Pew Center on the States (2009: 18):
Back in 1980, state researchers found, each prison bed represented a positive benefit-to-cost ratio. But during the 1990s and the first part of this decade, prison expansion captured less and less harmful offenders, leading to a dilution of impact. Put simply, after 20 years, locking up more drug and property offenders in Washington began to cost more than it was worth.
Nearly all of the cost-benefit research conducted on corrections focuses on alternatives to incarceration. There is very little directly on incarceration. One exception is Aos and colleagues (2001), who computed cost-benefit ratios for incarceration in Washington State. The 2001 ratios range from 0.37 for drug offenders to 2.84 for property offenders and 2.74 for violent offenders. These cost-benefit ratios declined dramatically since 1980, a reflection of the earlier discussion on the diminishing public safety returns of incarceration. A similar analysis of incarceration in North Carolina shows comparable reductions in the cost-benefit ratios over time (Yearwood et al. 2007).
COST-BENEFIT ANALYSES OF IN-PRISON INTERVENTION PROGRAMS
Whatever the computed cost-benefit ratios for incarceration, the question is whether alternative correctional strategies pay greater dividends per dollar invested than simple punishment. The answer is that there are many, many alternatives that do precisely that. What follows is not an exhaustive inventory of cost-benefit analyses. Instead, the research discussed here is illustrative of the cost-effectiveness potential of prison-based alternatives that provide more than just loss of liberty.
The research shows that prison-based treatment and intervention can produce substantial returns. For example, estimates from Washington and Iowa (respectively) show the cost-benefit ratio for in-prison cognitive behavioral programs is $49.55 and $37.70; in-prison drug treatment is $8.25 and $3.69; vocational education is $12.43 and $4.12; and correctional education is $18.11 and $2.91. These are just examples from two states. Returns on investment are a function of many factors. However, these results show that there is substantial potential to engage much more cost-efficient strategies than simply punishment.
The state of Connecticut implemented a multitier in-prison substance abuse treatment program. Eighteen-month postrelease rearrest was used as a measure of the impact of the program compared to similar inmates who did not receive the treatment. The substance abuse program, especially the more intensive tiers, significantly reduced postrelease recidivism and was cost-effective. The cost-benefit ratios, which measured the cost impact of reincarceration avoided, ranged from 1.8 to 5.7, depending on the tier of treatment (Daley et al. 2004).
New Jersey also implemented prison-based substance abuse treatment for inmates nearing their release date. Like the programming in Connecticut, the New Jersey in-prison substance abuse treatment was determined to be effective in reducing postrelease recidivism one year after release compared to similar controls that did not receive treatment. The net economic benefits were estimated to be between $4,307 and $6,209 per participant (French, Fang, and Fretz 2010).
The Washington State legislature passed the Drug Offender Sentencing Alternative (DOSA) in 1999. DOSA provides for judicial discretion in sentencing drug-involved felony offenders providing they are not sex offenders or violent offenders and if their instant offense involved a limited amount of a controlled substance. DOSA gives the court the option to impose the alternative of a split prison and community release sentence, providing the offender agrees to participate and complete drug treatment while confined. There are two primary groups of offenders sentenced under DOSA: drug offenders and drug-involved property offenders. Cost-benefit research conducted by the Washington State Institute for Public Policy shows a cost-benefit ratio range of $7.25 to $9.94 for the drug-involved offenders, and $0.93 for drug-involved property offenders. The alternative sentence and treatment is very cost-effective for drug offenders and neutral for drug-involved property offenders.
COST-BENEFIT ANALYSES OF ALTERNATIVE CORRECTIONAL INTERVENTION PROGRAMS
There is a growing body of cost-benefit research that indicates that correctional treatment in general is cost-effective. Even absent cost-benefit analyses, if such alternatives reduce recidivism and are cheaper than incarceration, they are likely to be a cost-efficient strategy compared to incarceration alone.
Some of the early work, summarized by Welsh and Farrington (2000), shows favorable cost-benefit ratios (ranging from $1.13 to $7.14, meaning that for every dollar invested, the benefits range from $1.13 to $7.14 depending on the intervention). The programs evaluated are largely community based and include pretrial diversion, employment training, and substance abuse treatment. The Washington State Institute for Public Policy (WSIPP) reviewed and meta-analyzed over 570 evaluations of adult correctional programs, juvenile correctional programs, and community-based crime-prevention programs. The WSIPP (2011) provides an extensive menu of the costs and benefits of intervention and treatment programs that have a low risk of failure. These include juvenile justice interventions (for example, Multisystemic Therapy, Aggression Replacement Training, Functional Family Therapy, Multidimensional Treatment Foster Care) and adult justice interventions (Dangerously Mentally Ill Offenders, Electronic Monitoring, Correctional Education, CBT, Work Release, and Vocational Education among others). They also include a wide variety of community-based crime prevention programs for children and families, community mental health programs, public health, and substance abuse treatment, and others. For example, Aos and colleagues (2001) report that on average, intensive supervision treatment-oriented programs can produce a nearly 17 percent reduction in recidivism and an $11,500 net economic benefit per offender. Community-based drug treatment can produce an average 9 percent reduction in recidivism and a $10,000 net economic benefit per offender. In-prison vocational education can reduce recidivism by 9 percent and provide a nearly $14,000 net benefit. On the other hand, the evidence confirms some of the conventional wisdom about ineffective programs. For example, adult boot camps have a zero recidivism return and a zero economic benefit, as do most domestic violence education programs and life skills education.
In a similar vein, the Iowa Department of Corrections conducted a comparable assessment of correctional interventions in that state. They found substantial cost-benefit ratios for both institutional programs and community-based programs for parolees and probationers.
Yearwood and colleagues (2007) estimated the cost-benefit ratios associated with community supervision in North Carolina. This analysis is simply the supervision costs, with no provision for added services targeting behavioral change. The results, comparing the cost benefit for community supervision to that for incarceration is as follows: violent ($22.58 versus $4.04 for incarceration), property ($47.89 versus $8.56), and drug ($6.29 versus $1.13). Again, there is little provision in these estimates for enhanced programming to mitigate public safety risk. Nevertheless, the analysis does decisively show the relative costs and benefits of community corrections compared to incarceration.
Diversion and Community-Based Alternatives
Community supervision is much less expensive than incarceration. Prison (in 2008) costs on average $79 dollars per day (nearly $29,000 per year); average probation cost is $3.42 per day ($1,250 per year), and parole is $7.47 per day ($2,750 per year). With careful screening for risk and provision of known recidivism-reducing strategies and interventions, public safety can be enhanced with considerable cost savings.
Diversion from incarceration, whether probation, deferred adjudication, jail diversion, and drug court, among others, has a quite favorable cost-benefit track record. The WSIPP and Iowa Department of Corrections research indicates that several community-based correctional programs analyzed are cost-effective. Examples, with their cost-benefit ratios, include: drug treatment ($5.11 to $7.35), cognitive-behavioral therapy in the community ($19.46 to $35.70), intensive supervision probation with treatment ($2.28 to $2.78), and community employment training and job assistance ($2.88 to $35.13). Community-based drug treatment utilizing a therapeutic community model (often in the context of a work release facility) yields an average cost-benefit ratio of $8.87 (Aos et al. 2001).
Jail diversion for individuals with co-occurring mental illness and substance use disorders is recognized as an effective way to divert offenders to necessary treatment services. Cowell, Broner, and Dupont (2004) assessed the effectiveness and cost- effectiveness of diversion of mentally ill, substance abusing individuals with justice system involvement. The diversion is either pre- or postbooking. Eligible individuals presented with bipolar disorder, depression, or schizophrenia and co-occurring substance abuse or dependence disorders. This assessment occurred at four different sites: Memphis, Tennessee; Lane County, Oregon; New York City; and Tucson, Arizona. There was variation in outcomes across the four sites, due to factors such as prebooking diversion versus diversion sometime subsequent to booking, as well as the type and intensity of treatment. The primary cost saving was due to diversion from jail. At the same time, when diversion leads to linking individuals to treatment, those costs can be quite high, depending on what treatments are involved (for example, inpatient mental health treatment). At two of the sites (Memphis and Tucson), the diversion resulted in psychiatric improvement. At the others, there was no significant increase in alcohol and drug use and rearrest, a modestly positive outcome in light of the characteristics of the population served. Due to a variety of research limitations, Memphis was the only site for which a cost-effectiveness estimate was derived. The finding is that for each one-point improvement in the Colorado Symptom Index (a measure of mental health functioning), the jail diversion program spent $1,236. It is important to keep a few things in mind. First, this is a challenging population of individuals who have a wide variety of significant criminogenic needs. Second, individuals with co-occurring disorders tend to be more service resistant than others, making intervention more difficult. Moreover, the primary cost saving is averted jail days and the primary expenses are for mental health, substance abuse, and physical health treatments. To the extent that jail diversion programs like these can even partially address the psychiatric problems of this population and interrupt the cycle of justice system involvement, future cost savings can be significant as mental health costs stabilize and diminish and jail days continue to be averted. This can be substantial in the long term as this population has historically cycled in and out of the justice system over a considerably long period of time. The point is that these cost-effectiveness studies are essentially static in time, without statistical and fiscal consideration for longer-term effects (that is, ten, twenty, or thirty years out).
Mental health courts are a relatively new diversion approach to addressing the needs of mentally ill individuals with justice system involvement. They are limited in number in part because of the high cost associated with treating individuals with mental health disorders and often co-occurring substance abuse disorders. We also know little about their effectiveness and cost-effectiveness, especially compared to diversion programs like drug courts. Ridgley et al. (2007) assessed the cost-effectiveness of the Allegheny County, Pennsylvania, Mental Health Court. The analysis included one- and two-year follow-up comparisons for similar individuals who were not diverted to community-based treatment. The evaluation indicates that the Allegheny County court experience is one of essentially offsetting jail and treatment costs for the first year of follow-up. The evidence also indicates that over a longer period (two years) the program may actually save public resources. In short, it was successful at diverting nonviolent offenders with serious mental illnesses out of traditional criminal adjudication and punishment and into community-based mental health treatment and other supportive services. It appears that this program does so without incurring substantial incremental costs.
California’s Proposition 36, passed in 2000, provides for selected California drug offenders to be diverted from incarceration to probation with drug treatment. The Substance Abuse and Crime Prevention Act (SACPA) was a substantial change to criminal justice policy. SACPA not only provides for diversion for nonviolent drug offenders with no serious or violent prior felonies or misdemeanors, it also provides that probationers and parolees who committed nonviolent drug offenses and who violate drug-related conditions of supervision can elect drug treatment rather than revocation. The law generally allows three opportunities to reenter drug treatment for violations. An assessment of the impacts of SACPA (Anglin et al. 2013) shows an economic benefit of $2,317 per offender. While a true cost-benefit ratio was beyond the scope of the research, the authors project a $97 million net savings from the first year of the program. Anglin and colleagues (2013: 1100) conclude:
Thus, moving offenders from the correctional system and into the treatment system, where they are exposed to rehabilitative efforts, should provide long-term savings through reduction in the overall incarcerated population, or at least an attenuation of its historical growth rate.
In 2009, New York State passed reforms to the Rockefeller drug laws, infamously some of the most punitive drug laws in the nation. The reform eliminated mandatory prison sentences for most felony drug offenders and provided for judicial diversion for many drug and drug-involved property offenders to drug treatment (Center for Court Innovation 2013). The cost difference between diversion to treatment and prereform comparison cases (including case disposition and the sentence) is $5,564. Taking recidivism into consideration, the net cost savings per diverted offender is $5,144 or a cost-benefit ratio of $2. Including victimization costs changes the ratio to $3.56. The aggregate cost savings to the state amounts to over $7 million per year.
A third example of diversion comes from the District of Columbia, which estimates the cost benefit of electronic monitoring while on probation. The comparison is regular probation. Roman et al. (2012) estimate an 80 percent probability that the electronic monitoring program will be cost-effective and determine the cost benefit due to recidivism reduction to be approximately $4,600. Another cost-benefit analysis by Yeh (2010) shows a cost-benefit ratio of $12.70 for the combination of home detention and electronic monitoring for probationers and parolees. The benefit includes justice system savings as well as broader social benefits.
The Juneau County Diversion Program diverts nonviolent misdemeanor offenders from traditional criminal adjudication and punishment in an effort to reduce the costs of processing these low-level offenders. The program matches offenders with mentors, participants attend life skills workshops, complete a minimum of twenty hours of community service, and pay full restitution. Upon successful completion, the charges are dropped. Analysis by Bong and colleagues (2012) shows the average annual net present value of the program is $25,734, based on four components of savings: the criminal justice system, the jail, community service, and enhanced employability and earnings for program completers.
The Program for Offenders, Inc. (TPFO), located in Pittsburgh, is a nonprofit community-based corrections facility that provides residential jail services to offenders diverted from prison. Participants are assessed for criminogenic needs and are provided a variety of services, including case management, inpatient substance abuse treatment, counseling, employment assistance, life skills training, education, and referrals for intensive outpatient substance abuse treatment, mental health treatment, housing and other services not directly provided by TPFO. The estimated cost-benefit ratio is approximately $4.09 (Yamatani 2012).
The Pima County Arizona Drug Treatment Alternative to Prison (DTAP) program is probation-based drug treatment that provides enhanced recovery support services for selected nonviolent drug offenders with at least two prior narcotics offenses who are diverted from prison to residential, community-based treatment. A cost-benefit analysis conducted by Herman and Poindexter (2012) reports that the Pima County diversion program, compared to a matched control group, saves the state of Arizona an average of $8,807 per offender compared to incarceration.
A similar drug offender diversion program was implemented in King’s County (Brooklyn), New York, in the 1990s. It was designed to divert nonviolent felony drug offenders to community-based drug treatment. A six-year follow-up showed that, compared to incarceration, the recidivism results of the diversion program with treatment had a cost-benefit ratio of $2.17.
Zarkin et al. (2012) conducted a simulation analysis of the lifetime costs and benefits of diversion of drug offenders from incarceration to community-based drug treatment. The authors develop two different scenarios and compare long-term outcomes (recidivism) with the no-diversion baseline data. The scenarios differ with regard to a 10 percent probability of diversion and a 40 percent probability of diversion, so scenario two is more aggressively scaled up. The results indicate in the 10 percent scenario, in which 44,200 offenders are diverted, the lifetime net benefit to the criminal justice system is $4.8 billion. The 40 percent scenario, in which 118,300 are diverted, the criminal justice cost saving is $12.9 billion. Estimation of the broader societal economic impact of diversion (which includes victimization costs, criminal justice costs of arrest, courts, incarceration, and health care costs) ramps up the net economic benefits to $8.5 billion for the 10 percent scenario and $22.5 billion for the 40 percent scenario.
The most commonly evaluated justice alternative is probably drug court. It is also the justice program for which the most cost-benefit research has been conducted. Drug courts can produce substantial reductions in recidivism and relapse (as discussed in chapter 5). At eighteen months after completion, drug court participants are less likely to need employment, educational, and financial services. These results are generally evident across a variety of offender types. Cost-benefit analyses have been conducted in drug courts across the country including, but not limited to, Washington, Oregon, Ohio, Vermont, California, Maryland, Michigan, Florida, Georgia, Illinois, Pennsylvania, South Carolina, Texas, and New York (Carey, Finigan, and Pukstas 2008; Carey and Waller 2011; Finigan, Carey, and Cox 2007; NPC Research 2009; Rossman et al. 2011a; Shaffer, Bechtel, and Latessa 2005; Washington State Institute for Public Policy 2003). Cost impacts and cost-benefit ratios all indicate favorable cost-effectiveness across the board. Cost-benefit ratios range from $1.74 to $9.43. Per participant net (of comparison group costs) cost savings due to recidivism reductions include $1,362 (St. Louis City Adult Drug Court), $1,982 (Maryland Drug Treatment Court Program), $3,451 (Baltimore City Drug Court), $4,133 (St. Joseph County Indiana Drug Court), $4,251 (Vigo County Indiana Drug Court), $4,600 (Howard County, Maryland), $4,896 (Montgomery County, Maryland), $5,809 (Vermont Drug Courts), $6,744 (Multnomah County, Portland, Oregon), $7,040 (Monroe County Indiana Drug Treatment Court), and $11,366 (Prince George’s County, Maryland). In those cases, for example Ohio, in which the cost-benefit ratio is relatively low or the cost-effectiveness is marginal, the analysts indicate that they were unable to control for offender risk. They suggest that had the risk principle been faithfully implemented so that higher-risk offenders populated drug courts, the cost results would be more favorable.
John Roman, an expert on drug courts, estimates that if drug courts were taken to scale to accommodate the roughly 1.5 million drug-involved offenders in need of substance abuse treatment, it would cost more than $13 billion annually. However, that investment would return more than $40 billion annually in benefits. Roman’s testimony before a House of Representatives subcommittee included the prediction that drug court effectiveness could be enhanced and cost controlled by utilizing a HOPE Court accountability model in conjunction with drug courts. This would allow the use of accountability/sanction courts for those who do not need drug treatment in order to desist from offending and therefore reserving the more expensive drug court model for those who cannot desist without substance abuse treatment.
Reentry
Despite the recent national focus on reentry spearheaded by the Urban Institute, there has been relatively little cost-effectiveness or cost-benefit analysis of reentry programs. Here is what is known. Welsh (2004) poses the question of whether there is a compelling economic argument for correctional treatment for offenders leaving prison. Welsh reports the cost-benefit results of a review of fourteen fairly early studies (most from the 1970s, 1980s, and 1990s) that focus on community-based correctional treatment. Most exhibited a favorable cost-benefit ratio, ranging from $1.13 to $2.82 to $7.14 to $53.73 to $270.00. Some of the studies monetized the recidivism effects and others monetized other outcomes such as substance use, education, and employment. Welsh (2004: 5) concludes
Offender reentry programs are crucial in an effort to reduce recidivism rates. But it may be that what comes before this end stage—in the form of correctional treatment programs—is equally, if not more, important. From the cost-benefit studies reviewed here, it would seem that a case can be made for increasing treatment resources for offenders, and this may improve offenders’ chances for a successful return to the community.
The Iowa Department of Corrections conducted a cost-benefit analysis of some of their reentry programs. Their analysis was modeled after the analytic approach developed by a partnership between the Pew Center on the States and the Washington State Institute for Public Policy. The Iowa Department of Corrections analysis determined the following reentry programs provided significant cost benefit: cognitive-behavioral programs ($34.30), drug treatment ($8.98), intensive supervision based on the risk-need-responsivity principle ($7.18), electronic monitoring ($6.43), employment training and job assistance ($5.02), and intensive supervision with treatment ($5.01).
McCollister and colleagues (2004) assessed the longer-term costs and benefits of in-prison substance abuse treatment in California (the Amity in-prison therapeutic community) and subsequent community-based aftercare postrelease from prison (the Vista aftercare programs). While substance abuse treatment required an average investment of $7,041 per participant over a five-year follow-up period (including primary treatment and aftercare), the averted incarceration days attributed to the program produced a cost-benefit ratio of $65. The additional investment of $5,311 in treatment yielded eighty-one fewer incarceration days (13 percent) among Amity participants relative to controls—a cost-effectiveness ratio of $65. When considering the average daily cost of incarceration in California ($72), these results suggest that offering treatment in prison and then directing offenders into community-based aftercare treatment is a cost-effective policy.
The State of Oregon’s Criminal Justice Commission (Officer, Bajpai, and Wilson 2011) conducted a cost-benefit analysis of an offender reentry program for releasees who received substance abuse treatment while incarcerated. The reentry program is designed to continue substance abuse programming once released, as well as provide assessment of the need for community-based services including mental health, employment counseling, career development, housing, and GED completion. The goal is to increase self-sufficiency and reduce recidivism. The evaluation research shows that the treatment group had significantly fewer rearrests (both felony and misdemeanor) than the control group. The cost-benefit analysis reveals a positive economic impact ($6.73 return for every dollar invested in the reentry programming).
The Maryland Reentry Partnership Initiative (MRPI) was established in 1999. It is a partnership of service providers that coordinate services for releasees returning to particular Baltimore neighborhoods. The goal is to provide coordinated and comprehensive reentry services including housing, substance abuse treatment, mental health treatment and counseling, education, and employment and vocational training, among others. The program is designed to identify inmates prior to release, provide prerelease planning and preparation, and then assist the transition by assessing service needs and providing community-based services. The evaluation indicated lower recidivism for the MRPI group compared to the controls. The total net economic benefit per participant is estimated to be $21,500. The cost benefit is estimated to be $3.00 (Roman et al. 2007). The greatest benefit from the program, in terms of cost and crime, is the reduction in serious crime. This finding of reduction in more serious crimes has been reported by other evaluators of reentry programs.
The Ready, Willing and Able program, based in New York City and Philadelphia, assists individuals recently released from prison by providing transitional housing, employment, and other support services. The goal is for releasees to independently maintain housing, employment, and sobriety. Cost-benefit analysis indicates the program benefits exceed the costs by approximately 20 percent.
The Nebraska Department of Correctional Services implemented a serious and violent offender reentry program in 2003. The intent was to target particularly dangerous and high-risk offenders and facilitate reentry. The first phase of the program was designed to provide substance abuse and mental health treatment, education, and life skills training while still incarcerated. Phase one also provided for the creation of transition teams responsible for developing individualized reentry plans for each releasee. Phase two was the transition to the community while receiving services from the state and private treatment providers. They also were on electronic monitoring until they demonstrated the ability to function without it. Phase three continued treatment services, but essentially no correctional control. The evaluation indicated a lower recidivism for program participants, compared to controls. The cost-benefit analysis resulted in an estimated net annual cost savings to the justice system of $5,849 per participant. When victimization costs are included, the total annual savings amount to $10,637 per program participant (Sample and Spohn 2008).
Dangerous mentally ill inmates who are released from prison present a particular risk to public safety. Odds are that they receive little effective mental health treatment while incarcerated and likely experience the criminogenic impact of incarceration. Postrelease supervision and control are important, however absent effectively addressing the mental health problems, that public safety risk persists. The Dangerous Mentally Ill Offender Program (DMIO) is designed to assess and intervene with more effective treatment for dangerous, mentally ill prison releasees. This is a postrelease program in Washington State for mentally ill releasees who are assessed to be a danger to themselves or others. The DMIO program begins while offenders are still incarcerated. Once identified as a dangerous mentally ill offender, they will be assigned to a treatment provider and will receive pre-engagement services during the three to four months prior to release, as well as treatment and transitioning planning. Upon release and up to five years out, DMIO participants receive community-based services including mental health and substance abuse treatment, housing assistance, medical care, and other support services. Mayfield (2009) reports that the four-year follow-up evaluation shows significant reductions in felony recidivism for program participants. The cost savings to the justice system as well as savings to potential crime victims average about $21,600 per participant. After four years, the cost-benefit ratio is $1.64 returned for every $1 spent on this program. A three-year follow-up reports a cost-benefit ratio of $1.24 of benefits to the justice system and crime victims (Mayfield and Lovell 2008).
Zhang, Roberts, and Callanan (2006) conducted a cost-benefit analysis of a statewide parole program in California. The California Preventing Parolee Failure Program (PPFP) was developed in response to record high recidivism rates of prison releasees. A California Department of Corrections analysis identified several factors that were related to parole failure: substance abuse, unemployment, illiteracy, and homelessness. The PPFP provided education, training, treatment, and assistance designed to mitigate these criminogenic circumstances. Initial positive findings of recidivism reduction and cost savings led to expanded funding and a new name, the Preventing Parole Crime Program (PPCP). The budget tripled in 1998. The 2006 evaluation results indicated that the PPCP reduced returns to prison among paroles who received services under the program, compared to controls. The reduction in reincarcerations led to aggregate cost savings of $21 million during the study period. The cost-benefit analysis resulted in an estimated net return on investment of 47 percent, or a 47 percent net return for every $1 invested in the PPCP.
Roman and Chafin (2006) assessed the economic impact of reentry initiatives for jail inmates. The initiatives included services such as provision of life skills and substance abuse counseling, employment and education services, and healthcare, among other services designed to assist with re-entry. The economic analysis compared the reentry program participants’ outcomes to business as usual processing in the justice system without reentry services. The analysis indicated that under different conditions of program cost, it only requires modest reductions in offender recidivism to render jail-based reentry programs cost-effective. They conclude that the case for implementing jail-based reentry programs is strong based on cost-effectiveness analyses.
The Washington State Institute for Public Policy in conjunction with the Pew Center on the States has developed a cost benefit tool, as has the Vera Institute (Cost Benefit Knowledge Bank). These tools are easily available for practitioners to use to assess cost-effectiveness and estimate cost-benefit ratios.
The end game here is to reduce recidivism and crime. I have discussed a variety of evidence-based strategies that have demonstrable track records for being effective and cost-efficient. The financial impacts on the criminal justice system of engaging proven practices and implementing them in effective, appropriate ways are substantial. These impacts include reductions in incarceration costs, the cost of law enforcement and the courts, as well as victimization costs. Shapiro and Hassett (2012) have taken these economic impacts and extrapolated the effects to particular U.S. cities. For example, they estimate that a 10 percent reduction in violent crime in Boston will produce a $5 million savings to local government and $73 million in intangible losses to victims. A 25 percent reduction in violent crime would result in a $12 million annual savings, reduce the direct costs to victims by $18 million, and the intangible victim costs by $180 million. They estimate that a 10 percent reduction in homicide in the Boston area would boost the value of residential real estate by $4.4 billion. A 10 percent reduction in violent crime in Chicago translates into a $24 million savings to local government, a $43 million savings in direct costs to victims, and $420 million in victim intangibles. Residential real estate value is expected to increase by $2.2 billion with a 10 percent reduction in murder in Chicago. The scenario for Houston is a $17 million annual savings in direct costs to local government, $27 million for victim direct costs, and $265 million in victim intangible costs. The impacts are equally impressive for Dallas, Jacksonville, Milwaukee, Philadelphia, and Seattle.
The takeaway from the Shapiro and Hassett analysis and forecasting is that the economic impact of crime reduction affects the expense side as well as the revenue side of government. The expense side is in terms of the criminal justice cost savings; the revenue side is in terms of things such as increases in property values, which in turn translate into higher property tax revenue. Some of the savings to victims may translate into increased consumer spending, resulting in increases in sales tax revenues. Such revenue increases could in part be used to fund or increase funding for initiatives that have been demonstrated to be effective crime prevention programs; for example, programs identified by the Washington State Institute for Public Policy as having quite favorable cost-benefit ratios, including a variety of early childhood education programs such as Nurse-Family Partnerships, parent training, children’s mental health interventions, and youth mentoring programs.
WHAT DOES THE COST-BENEFIT RESEARCH TELL US ABOUT THE RECOMMENDED PATH FORWARD?
This section provides a brief summary of what is known about the cost benefit of the recommendations made in prior chapters. At the outset, the evidence is clear that since incarceration alone does not effectively reduce recidivism and is extraordinarily expensive, it is reasonable to conclude that alternatives that are demonstrated to be effective at reducing recidivism and cost less are, by definition, more cost-efficient than incarceration. However, the incapacitation function of incarceration going forward is necessary. It is important, however, that the justice system transition the use of incarceration for serious violent offenders, truly habitual offenders, and those for whom there is clear, compelling evidence that rehabilitation is not possible. At the same time, incarceration should not be void of programming. The vast majority of the offenders that are incarcerated in the future will be released and will have significant risk when released. Some of that risk can be mitigated by implementing programing while incarcerated. The evidence indicates that in-prison substance abuse treatment as well as cognitive-behavioral interventions targeting other criminogenic deficits are cost-effective.
Many of the recommendations I have discussed are not programs per se and thus are not as amenable to cost-benefit analysis. For example, problem-solving prosecution and collaborative sentencing are statutory, procedural, and cultural changes that facilitate the goals of recidivism reduction, targeted incarceration, and behavioral change. Others are too new to have cost-benefit analyses. For example, targeting neurocognitive deficits and impairments and swift and certain sanctioning are relatively new initiatives with evaluation research speaking to effectiveness, however, the specific economic analyses are yet to be conducted.
The primary focus of what this book is recommending is a concerted and dramatically expanded effort at behavioral change. On balance, the evaluation research and the cost-benefit analyses indicate that is a prudent and profitable path.
First, the research indicates that diversion is a wise investment. Whether probation, deferred adjudication, jail diversion, diversion courts, and others, the research indicates that when properly implemented, these treatment-based alternatives to incarceration have substantial economic benefits, including cost savings to the criminal justice system, local government, communities, and victims. Moreover, the cost-benefit research supports efforts to take many of these initiatives to scale. The evidence regarding the effectiveness of drug diversion is so compelling and the cost-benefit impacts are so favorable that the economics clearly support expansion of capacity. The cost-benefit results are also compelling for programs that address a variety of other criminogenic deficits and impairments. Programs that are designed to assist with educational deficits, employment, vocational training, mental health, and housing, among others, when implemented in community-based settings, are not only effective in helping to reduce recidivism, they are also cost-effective.
The evidence regarding probation is especially important. Probation is a critical opportunity to advance the goals of behavioral change, reduce recidivism, reduce victimizations, and enhance cost-effectiveness. There is tremendous potential there and the research indicates that additional investment in probation is clearly warranted and fiscally prudent. Again, community-based substance abuse treatment, cognitive-behavioral therapy, employment and job training, and intensive supervision with treatment are a few of the probation-based interventions that provide positive ROIs.
Reentry is another clear opportunity for improvement and the research indicates that reentry programming can be cost-effective. Substance abuse treatment and aftercare, cognitive-behavioral interventions, employment, housing, job assistance, electronic monitoring, mental health treatment, and education programming, as well as prerelease planning and preparation, can be cost beneficial.
The results of the cost-benefit research can be helpful in transitioning our thinking away from the tough rhetoric of crime control and seeming disregard for scientific evidence to policymaking that is informed by an emphasis on what is effective and financially prudent. There is clear guidance provided by the cost-benefit research regarding where to invest wisely in the future.
SUMMARY
The goal of this chapter has been to provide policymakers with information regarding the potential for alternatives to incarceration to be not only effective, but cost-efficient as well. Much of our effort over the past forty years and much of the money spent on that effort has not provided a reasonable return on investment. Let’s be fair. Policy development and implementation is a result of many factors, some of which concern whether certain initiatives are effective and whether they are cost-efficient.
In the course of the discussions in chapters 3 through 7, I have provided alternatives and options that are grounded in scientific evidence and for which the scientific community has demonstrated effectiveness. The inventory of effective practices and programs is quite impressive and provide policymakers with a wide variety of viable options. Moreover, I have also shown in this chapter that engaging in actual behavioral change, which is the goal of most of these alternatives to incarceration and control, can be cost-effective. The evidence indicates that for the programs for which cost-benefit analyses have been conducted, most return more financial benefit to the justice system and society at large (potential victims, property values) than the upfront investment.
The brevity of this chapter serves to illustrate that the assessment and evaluation of the financial impacts of correctional alternatives has not kept pace with the evaluation of program effectiveness. We are just recently beginning to learn that evidence-based practices are reasonable choices, not only because they effectively change behavior and reduce recidivism and crime, but also because many do so by returning more fiscal benefit relative to the cost incurred. Moreover, the dividends from these programs are likely significantly understated as they ignore the longer-term financial impacts of desistence from crime of career offenders—breaking the cycle of repeat offending, reductions in crime, reductions in justice system contacts, and reductions in victimizations.
Much of what is recommended in the preceding pages is based on prioritizing and addressing in a meaningful sequence or cluster the primary criminogenic needs of offenders. What is missing in the cost research are estimates of the fiscal benefits of targeting multiple criminogenic needs, that is, the cumulative, marginal effect.
While cost-benefit research has not yet provided the full picture of return on investment of the variety of alternatives discussed in the preceding pages, at the end of the day, we know that incarceration is quite expensive and has limited effectiveness. If the prison population declines, the marginal utility of incarceration may, all else equal, improve. However, the fact that alternatives are effective and in most instances cost less is assurance that the path outlined in this book is fiscally prudent.
