What’s in thoseCost-of-crime Estimates?

Rudy HaapanenCenter for Public Policy Research, UC Davis

Association for Criminal Justice Research (CA)Spring Conference

March 25, 2010

PROGRAM COST/BENEFIT

BENEFITS AND COSTS OF PREVENTION PROGRAMS FOR YOUTH Does prevention pay? The Washington State Institute for Public Policy has provided credible evidence that for each dollar spent on “research-based” prevention programs for youth, more than a dollar’s worth of benefits will be generated (i.e., benefits exceed the costs). Following are the benefits accrued per dollar of cost for the Blueprints programs that were analyzed.

From the Institute of Behavioral Science (IBS) at the University of Colorado at Boulder,

Blueprints for Violence Prevention

PROGRAM COST/BENEFIT

BENEFITS AND COSTS OF PREVENTION PROGRAMS FOR YOUTH Blueprints Model Programs Benefit per Dollar Outcome of Cost • Big Brothers Big Sisters $ 1.01 Drugs, Crime, Educational • Functional Family Therapy $13.25 Crime reduction • Life Skills Training $25.61 Drug reduction • Midwestern Prevention (Project STAR) $ 5.29 Drug reduction • Multidimensional Treatment Foster Care $10.88 Crime reduction • Multisystemic Therapy $ 2.64 Crime reduction • Project Towards No Drug Abuse $55.84 Drug reduction • Nurse-Family Partnership $ 2.88 Crime reduction

From the Institute of Behavioral Science (IBS) at the University of Colorado at Boulder,

Blueprints for Violence Prevention

PROGRAM COST/BENEFIT

Source: Evidence-based juvenile offender programs: Program description, quality assurance, and cost, WSIPP, June 2007

Early Studies

Prison is cost-effective• Zedlewski, E. (1987) Making Confinement Decisions.

National Institute of Justice Research in Brief – National estimates of crime avoidance, control, and

justice system response expenditures ($99.8 billion) – Divided by the number of crimes committed in 1981

(42.5 million, as estimated from victimization surveys) – Calculated a per crime cost of $2,300

Early Studies

Prison is cost-effective• Zedlewski, E. (1987) Making Confinement Decisions.

National Institute of Justice Research in Brief – Applied this figure to average self-reported crimes prior

to prison (from Rand Surveys) of 187 to 278 per year• inflated and highly skewed (median = 15)

– $430,000+ in avoided crime cost per year of prison use– This (huge) sum was argued to justify an expansion of

prison use as an incapacitation strategy

Early Studies

Early Release is Cost-effective• James Austin (1986) Using early release to relieve prison

crowding: A dilemma in public policy. Crime & Delinquency– Costs of crimes committed by those released early

compared to prison costs avoided by the early release– NIJ estimates for per-crime costs, based on short-term,

out-of-pocket expenses to victims (victimization surveys)– Crime costs overshadowed by avoided prison costs

(early release is sound policy)

Early Studies

Early Release is Cost-effective• James Austin (1986). Using early release to relieve prison

crowding: A dilemma in public policy. Crime & Delinquency– In contrast to Zedlewski’s $2,300 per average crime,

Austin’s figures were much lower – A rape was valued at only $350, the average out-of-

pocket expense reported by rape victims at the time – Out-of-pocket expenses may not adequately capture

the long-term impact on victims, especially for violence

Current Estimates

Two Types of Cost:• Criminal Justice System Costs

– Incremental costs (Law enforcement, courts)– Average costs (incarceration, probation)

• Victim and Social Costs– Out-of-pocket expenses– Intangible expenses and “social harm”

CJS Costs

• Analyze how operating budgets vary as a function of the number of crimes in a jurisdiction– Budgets apportioned by number and types of crimes for

law enforcement, DA offices, courts (WSIPP)– Budgets apportioned by amount of time spent on tasks

associated with specific types of crime (Home Office)– Average Costs per offender for jail, prison, juvenile

facilities times the probability of sentence for each crime type times the number of crimes of that type

CJS Costs

Types of Crimes and Resource Costs Analyzed in the WSIPP Cost-Benefit Model Types of Crime Types of Resource Costs Incurred

1. Murder/Manslaughter 1. Police and Sheriffs’ Offices2. Rape/Sex 2. Superior Courts and County Prosecutors3. Robbery 3. Juvenile Detention, with Local Sentence4. Aggravated Assault 4. Juvenile Detention, with JRA Sentence5. Felony Property Crimes 5. Juvenile Local Probation6. Drug Offenses 6. Juvenile Rehabilitation, Institutions7. Misdemeanor Crimes 7. Juvenile Rehabilitation, Parole

8. Adult Jail, with Local Sentence9. Adult Jail, with Prison Sentence10. State Community Supervision, Local Sentence11. Department of Corrections, Institutions12. Department of Corrections, Post-Prison Supervision13. Crime Victim Monetary Costs14. Crime Victim Quality of Life Costs

Source: Table D.2a Benefits and Costs of Prevention and Early Intervention Programs for Youth: Technical Appendix. Washington State Institute for Public Policy (Aos et al., 2001).

CJS Costs

Estimating by modeling budgets in terms of crimes• WSIPP estimates of costs for law enforcement, prosecution, and courts were

developed by comparing jurisdictions across Washington State

• Operating budgets were regressed on arrests:

Police and Sheriff's Offices

ln(Oper. Exp.)=9.55+.212ln(FVA)+.181ln(nFVA)+.266ln(nFA)+.203ln(TR)R2Adj=.84N=341

• Felony violent arrests (FVA),• Felony non-violent arrests (nFVA), • Non-felony arrests (nFA), • Traffic infraction filings (TR).

CJS Costs

Estimating by modeling budgets in terms of crimes

• Implied modelCrime Volume Operating Budget

• Is there any theoretical justification or empirical support for this model?

CJS Costs

Estimating by modeling budgets in terms of crimes

• More plausible alternative modelCrime Volume

PopulationOperating Budget

• Even here, however, there is little theoretical or empirical support for a direct relationship between crime and CJS budgets

• Especially for small, incremental changes in crime

CJS Costs

Estimating CJS costs by apportioning staff time• The Home Office in the UK and some researchers in the US

(Fowles et al, 2005) sought to allocate time spent by LE staff on various crimes

• Disaggregate CJS budgets by type of activity• Even if budgets remained the same, reductions in crime

could result in diverting resources to other needed activities (traffic control, parking enforcement, minor crimes, prevention)

• “Opportunity cost” model

CJS Costs

• Source: Exhibit 3 in Aos et al., 2008.

Victim Costs

• Cohen, M. (1988). Pain, suffering, and jury awards: A study of the cost of crime to victims. Law and Society Review.– Re-analysis of Austin’s data, incorporating figures from

jury awards for pain and suffering as part of the costs of crime

– Using these data, even a small number of serious crimes can be quite costly

– May justify even expensive options such as prison – But we need to consider ALL alternatives

Victim Costs

Miller, Cohen & Wiersema (1996) produced the first set of crime-specific cost estimates for NIJ

• these estimates went beyond out-of-pocket expenses to include lifetime estimates of:

– lost wages/productivity;– medical care;– mental health care;– police and fire services; – victim services;– property losses; and– pain, suffering, and lost quality of life.

Victim Costs

Miller, Cohen & Wiersema (1996)• Additional analyses incorporated estimates for risk of death

for non-homicide crimes and • included an estimated value of a “statistical life”

proportional to the risk of death for each crime • The value of a statistical life was based primarily on wage

differentials for riskier jobs

Victim Costs

Source: Miller, T., Cohen, M., & Wiersema, B. (1996). Victim costs and consequences: A new look, NIJ Research Report, page 9.

Victim Costs

Victim Costs

• These concepts and the logic underlying these estimates have not been disputed for the most part

• The Miller, Cohen & Wiersema (1996) estimates have been included in a number of subsequent estimates

• Are included in the WSIPP analysis and the Blueprints estimates

Victim Costs

• Critics point out: Costs have little basis in economic reality

– The quality of life loss estimates were based on jury awards for pain and suffering in civil cases (by definition, unusual situations and atypical victims)

– These awards are not tied to any true cost in actual dollars but, rather, serve as a monetary symbol of the seriousness of the consequences to the victim

– Hence, the dollar figures themselves have little basis in economic reality (What does it mean, for example, to value a rape at 62.1 burglaries--the ratio of the “costs”)?

Victim Costs

• Critics point out: Many crimes are excluded

– The estimates are for costs to victims, and only include crimes in which victimization of “unwilling” victims occurs.

– Exclude drug use or possession, consensual sex crimes (prostitution or soliciting), con games, or gambling.

– Similarly, exclude business or white-collar crimes such as embezzlement, fraud, or violations of health and safety codes

Victim Costs

• Critics point out: Many crime-related costs are excluded

– The estimates are limited to costs that can be reasonably attributed to individual victimizations

– Other social costs related to crime are difficult or impossible to attribute to specific incidents. • installation of home security systems and car alarms, • security services for buildings, • driving rather than walking at night, or • even the higher value of homes in “safer neighborhoods.”

Summary: CJS Costs

• CJS marginal cost estimation should carefully consider whether components of the criminal justice system can and will respond to small downward changes in crime with commensurate cost reductions

• Even large reductions in overall crime over the past 15 years have not been followed by reductions in criminal justice system expenditures

• Care must be taken to avoid suggesting that costs will decrease unless such a causal link can be demonstrated

Summary: Victim Costs• Victim cost estimates that incorporate intangible costs or

“social harm” are remote from any true cost in actual dollars• Although they may serve as monetary symbols of the

relative extent of suffering and reduced quality of life, the dollar figures themselves have little basis in economic reality

• To array these figures on the “benefit” side of a cost-benefit equation against real dollar expenditures on the “cost” side is to return to the problem that prompted the move toward cost-benefit analysis in the first place

What is the answer?

• Either way, the cost/benefit ratios currently in vogue appear to be based on methods better adapted to describing the extent to which individual and collective resources are devoted to dealing with “crime” in one way or another (that is, capturing all the costs)

• There is no current evidence that these costs will vary as a direct function of small changes in the number of crimes—what we would expect from the programs they are used to justify

• These approaches (and these estimates) have gained wide popularity partly because they seem to support treatment and rehabilitation programs

• Cost estimates that are high suggest that any rehabilitative program that reduces any amount of crime is cost-beneficial

References

Aos, S., Lieb, R., Mayfield, J., Miller, M., & Pennucci, A. (2004a). Benefits and costs of

prevention and early intervention programs for youth. Olympia, WA: Washington

State Institute for Public Policy.

Aos, S., Lieb, R., Mayfield, J., Miller, M., & Pennucci, A. (2004b). Benefits and costs of

prevention and early intervention programs for youth: Technical appendix. Olympia,

WA: Washington State Institute for Public Policy.

Aos, S., Miller, M., & Drake, E. (2006). Evidence-based policy options to reduce future

prison construction, criminal justice costs, and crime rates. Olympia, WA:

Washington State Institute for Public Policy.

Aos, S., Phipps, P., Barnoski, R., & Lieb, R. (2001). Comparative costs and benefits of

programs to reduce crime. Olympia, WA: Washington State Institute for Public Policy.

Austin, J. (1986). Using early release to relieve prison crowding: A dilemma in public

policy. Crime & Delinquency, 32(4), 404-502.

References

Cohen, M. (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. (2005). The Costs of Crime and Justice. New York: Routledge.

Fowles, R., Byrnes, E., & Hickert, A. (2005). The cost of crime: A cost/benefit tool for

analyzing Utah criminal justice program effectiveness. Commission on Criminal and

Juvenile Justice & Criminal and Juvenile Justice Consortium, University of Utah.

Miller, T., Cohen, M., & Wiersema, B. (1996). Victim costs and consequences: A new

look (NCJ-155282). Retrieved from the National Institute of Justice Research

website: http://www.ncjrs.org/pdffiles/victcost.pdf

Zedlewski, E. (1987) Making Confinement Decisions. National Institute of Justice

Research in Brief. USDOJ.

CPPR

The Center for Public Policy Research (CPPR) CPPR at the University of California, Davis serves state government and academic institutions.

We conduct research to bring science to policy.

We draw on the expertise of faculty and researchers at UC Davis and throughout the UC/CSU system.

Mission: To inform public policy through rigorous research and conscientious, impartial analysis of issues.

CPPR

Core CPPR Research Staff • Gail S. Goodman, Ph.D.—Professor & Director of CPPR• Michael J. Lawler, MSW, Ph.D.—Co-Director of CPPR• Rudy Haapanen, Ph.D.—Research Director

rahaapanen@ucdavis.edu• Ingrid Cordon, Ph.D.—Quantitative Analyst• Shay O’Brien, MSW—Research Writer

http://cppr.ucdavis.edu/