1. 1 of 15 Steve Aos Director Washington State Institute for Public Policy Phone: (360) 586-2740 E-mail: saos@wsipp.wa.gov Institute Publications: www.wsipp.wa.gov An “Evidence-Based” Investment Strategy Identifying & Implementing Policies that Improve Outcomes and Save Money: The Washington State Approach Canadian Congress on Criminal Justice Vancouver, British Columbia October 4, 2013

2. 2 of 15 Washington State Institute for Public Policy Nature of the Institute Non-partisan, created by 1983 Legislature General purpose legislative research unit Projects assigned by legislative bills Legislative & Executive Board Directions to WSIPP from the WA Legislature What works? What doesn’t? What are the costs & benefits of policies to improve… Crime (1994, 1999, 2003, 2005, 2009, 2012, 2013) Education, Early Ed (2003, 2006, 2009, 2012, 2013) Child Abuse & Neglect (2003, 2009, 2012) Substance Abuse (2003, 2005, 2009, 2012) Mental Health (2005, 2009, 2012) Health Care (2012) Developmental Disabilities (2008) Teen Births (1994) Employment, Workforce Training (2009) Public Assistance (2009) Public Health (2009) Housing (2009) WSIPP CapitolOlympia Seattle My House

3. Results Application in WA The ROI Numbers Big Picture 3 of 15 Change in Crime Rates United States and Washington: 1980 to 2011 Crime Rates : -45% -46% United StatesWashington Homicide Rates: -48% -49%

4. Results Application in WA The ROI Numbers Big Picture 4 of 15 Adult Prison Incarceration Rates: 1930 to 2012 1930 1940 1950 1960 1970 1980 1990 2000 2010 2020 *Incarceration Rate *The incarceration rate is defined as the number of inmates in prisons per 1,000 resident population in Washington or the United States. 1 2 3 4 5 6 0 United States Washington

5. Results Application in WA Big Picture The ROI Numbers 5 of 15 1.Evidence:. Are There Evidence-Based Policies that Improve Outcomes with a Positive Return on Investment? Our 3-Step Research Approach 2.Economics: 3.Portfolio: What works to improve outcomes; what does not? We analyze all rigorous evaluations on policies to improve public outcomes of legislative interest. What is the return on investment? We compute benefits, costs, and risk to the people of Washington State using a consistent framework. How would a combination of options affect statewide outcomes? What is the risk?

6. Results Application in WA Big Picture The ROI Numbers 6 of 15 WSIPP “Consumer Reports” Lists Evidence-based policy options ranked by return on investment

7. Results Application in WA Big Picture The ROI Numbers 7 of 15 Change In Crime (# of EB Studies) Benefits Minus Costs, per-person, life cycle (Probability: you lose $) Cog-Behavioral Treatment-7% (38) $9,283 (<1%) ISP: surveillance only0% (14)-$4,718 (89%) ISP: treatment focus-14% (17)$7,295 (4%) Multisystemic Therapy -13% (11)$24,751 (2%) Aggression Repl. Training -20% (4)$29,740 (4%) Adult Offender Programs Pre-School* (low income) -21% (11) $14,934 (<1%) Nurse Family Partnership*-17% (3) $13,182 (20%) Prevention* Functional Family Therapy -22% (8)$30,706 (<1%) Drug Tx in Prison (TC or out-patient) -12% (21) $10,974 (<1%) Scared Straight+8% (10)-$9,887 (100%) What Works to Reduce Crime? * Programs have a number of other non-crime benefits; all benefits reported here. (Examples from our latest results) Incarceration Per Capita-13% (11)$24,751 (2%) Prison & Policing Juvenile Offender Programs * Police Per Capita-22% (8)$30,706 (<1%) We located and meta-analyzed 38 rigorous outcome evaluations conducted in the United States, Canada, and UK. We find, on average, the program reduces recidivism 7 percent. Without CBT, an moderate-to-high risk offender has a 69% chance of being reconvicted for a new felony or misdemeanor; With CBT, the odds drop to about 64%. The reduced recidivism = a NET gain of $9,283 per participant. We estimate CBT, which is done in groups, costs $412 per participant; benefits of reduced recidivism total $2,308 to taxpayers (lower criminal justice costs) and $7,387 to crime victims and others (reduced victimization). A total benefit-to-cost ratio of $21 to $1 Risk: Less than 1% of the time you lose money (costs exceed benefits). We ran the model 1,000 times testing the expected bottom line for the risk and uncertainty in our findings. Updated results available soon

8. Next Steps Washington State Overview The Numbers Number of Slots Funded 100 Taxpayer Cost per Slot (net) Total Cost Total Benefits Bottom Line: Benefits – Costs Outcome: Crime Program: Functional Family Therapy for Juvenile Offenders $3,300 $330,000 60 47 - 13 Number who recidivate with a new crime WITHOUT FFT WITH FFT Net Change $211,000 $2,743,000 +$2,413,000 Lifetime benefits per avoided recidivist 8 of 15 Functional Family Therapy We studied, carefully, how FFT is done Washington “Base” Population Characteristics: We gathered longitudinal information on juvenile offenders in Washington. Economics: We built a model to estimate the cost of crime in Washington State Costs to the victims of crime Costs to the taxpayers for the criminal justice system Other benefits of juvenile crime avoidance: education outcomes Economics: We built a model to estimate the cost of crime in Washington State Costs to the victims of crime Costs to the taxpayers for the criminal justice system Other benefits of juvenile crime avoidance: education outcomes

9. Results Application in WA Big Picture The ROI Numbers 9 of 15 WSIPP “Consumer Reports” Lists Evidence-based policy options ranked by return on investment

10. Results The ROI Numbers Overview Application in WA 10 of 15 Using Research to Craft Criminal Justice Policy: Washington’s Legislative Milestones 1995 2012 & 2013 2000 2005-6 Juvenile Justice → Budget Change Adult Corrections → Budget Change What policy portfolio reduces crime & limits prison construction? → Budget Change, Silo Translated the Legislature's evidence-based investments into budget drivers for prison. Evidence-based budget requirements (for juvenile justice, adult corrections, children’s mental health, child welfare, adult mental health & substance abuse). 2007 2003 Prevention → Budget Change 1984 Sentencing Reform

11. Results The ROI Numbers Overview Application in WA 11 of 15 Four Evidence-Based Implementation Issues (Things we’ve learned) Formal Assessment Process (Tools) to align participants with the right programs, and to focus resources on higher-risk populations. State-Funded Quality/Fidelity System to assure better adherence to the assessment system and the intervention program models. Swift and Certain Apprehension clear evidence (for crime deterrence) for certainty, but not for severity of punishment. Funding Formulas with the Right Incentives to encourage interest, adherence, and innovations in evidence-based programs.

12. Application in WA The ROI Numbers Big Picture Results 12 of 15 Keeping Track of Results: Prison Beds Avoided Cumulative Effect of Washington’s History of Evidence-based Programming 0 1980 Average Daily Prison Population Year 2,000 4,000 6,000 8,000 10,000 12,000 14,000 16,000 18,000 20,000 22,000 24,000 198519901995200020052010201520202025 As of 2013, there are about 1,500 fewer people in prison as a result of Washington’s evidence- based adult, juvenile, & prevention programs. These effects are in the state prison forecast. Blue Area = Actual Prison PopulationYears beyond 2013 are current state forecast. Orange Area = What ADP would have been without the high ROI programs.

13. Application in WA The ROI Numbers Big Picture Results 13 of 15 Trends in Adult Recidivism in Washington: 1990-2007 Annual prison release cohorts by DOC risk classification level 31% 20% 0% 10% 20% 30% 40% 50% 60% 70% 80% Among moderate risk offenders released from prison in 1990, 31% were reconvicted for a new felony within three years. Among moderate risk offenders released from prison in 2007, 20% were reconvicted for a new felony within three years. Any Felony Re-conviction (within 3 years after release) '90'92'94'96'98'00'02'04'06'90'92'94'96'98'00'02'04'06'90'92'94'96'98'00'02'04'06'90'92'94'96'98'00'02'04'06 Lower Risk (20% of total prison pop ) Moderate Risk (17%) High, Non-violent (19%) High, Violent (44%) Risk Classification Level of Offenders In Prison 0% 10% 20% 30% 40% Violent Felony Re-conviction by year of release from prison Lower Risk 20% Moderate Risk 17% High, Non Violent 19% High, Violent 44%

14. Application in WA The ROI Numbers Big Picture Results 14 of 15 VT MA RI CT WAOR ID CA NM KS TX IA IL NY MS FL 16 other US States are now implementing versions of the “Washington Approach” via the Results First project of the MacArthur Foundation and the Pew Charitable Trusts

15. Caveat 15 of 15 Two Goals of Criminal Justice Policy: Crime Reduction (to achieve less crime in the future) Justice (to address criminal wrongs done in the past) Benefit-cost and recidivism risk findings can help policymakers with the crime reduction goal, but they are pretty much silent on the justice goal.

16. ANNUAL CRIMINAL JUSTICE ISSUE EXCLUSIVE RATINGS Over 200 Crime–Related Programs and Policies Programs programs taxpayer dollars that reduce crime and save taxpayers money. programs BEST 2013 Crime Policies to Adopt S Institute Publications: www.wsipp.wa.gov Reports Are Available on: Criminal Justice Juvenile Justice Child Welfare Education Mental Health Substance Abuse Prevention …More on the way

17. Thank You

18. Appendix

19. Prison is Increasingly Used for Violent Offenders Change in Average Violent Felony Risk Scores Since 1990 Year -5% +0% +5% +10% +15% +20% +25% +30% +35% 19901995200020052010 Since 1990, the average violent felony risk score of offenders released from Washington prisons has increased more than 30%.

20. 0 1 2 3 4 5 6 7 8 9 10 1980198519901995200020052010 * Crime rates are the number of reported crimes to police per 1,000 resident population. Source: WASPC and FBI. 0 10 20 30 40 50 60 70 80 90 100 1980198519901995200020052010 United States Washington Crime Rates: Violent & Property Washington and United States: 1980 to 2010 Violent Crime Rate* Property Crime Rate* United States Washington Violent crime has dropped since the mid-1990s, and Washington’s rate remains lower than the US. Property crime rates have declined, and the gap between Washington and the US has narrowed.

21. 0 10 20 30 40 50 60 1985198719891991199319951997199920012003200520072009 Juvenile Arrest Rates: Closing the Gap Long-Term Trends in Juvenile Arrest Rates (1985 to 2010) Juvenile Arrest Rate (arrests per 1,000 10-17 year olds) Year United States Washington Prior to the mid-1990s, Washington’s juvenile arrest rate was consistently higher than the US rate. The gap started to close in the mid-1990s. Today, the two juvenile arrest rates are virtually identical. In 1995, the Legislature First Tells WSIPP to Study Juvenile Justice!

22. DOC Risk-Level Classifications of Adult Offenders: The Timing of Recidivism in Washington State for… Months After Being At-Risk in the Community …a New Felony Conviction…a Violent Felony Conviction Months After Being At-Risk in the Community Source: WSIPP analysis of data from the Administrative Office of the Courts and the Department of Corrections 0.0% 0.5% 1.0% 1.5% 2.0% 2.5% 3.0% 3.5% 4.0% 161218243036 Classified by DOC as: High, Violent Lower Moderate High,Non-Violent Classified by DOC as: High, Violent 0.0% 0.5% 1.0% 1.5% 2.0% 2.5% 3.0% 3.5% 4.0% 161218243036

23. Caveat Application in WA Principles Big Picture 5 of 14 Keeping Track of Results: the Incarceration-Crime Relationship Washington’s Crime Rate (non-drug crimes per 1,000 pop) 30 35 40 45 50 55 60 65 70 75 Washington’s Incarceration Rate (ADP per 1,000 pop) 1234 If Prison Doesn’t Work If Prison Really Works Each data point is that year's incarceration rate and crime rate. Elasticity: Elasticity: how a percentage change in a state’s incarceration rate leads to a percentage change in its crime rate. Key Development: The long-term link between incarceration and crime in WA appears to have changed, favorably, around 2006. Crime is now falling without expensive increases in incarceration rates. Better public policies have had a role in the improved results. 2011 2007 2008 2009 2010 ‘89 ‘90 ‘91 ‘92 ‘93 ‘94 ‘95 ‘96 ‘97 ‘98 ‘00 ‘99 ‘01 ‘02 ‘03 ‘04 ‘05 2006 ‘85 ‘86 ‘87 ‘88 ‘81 ‘82 ‘83 ‘84 1980

24. Results Application in WA Big Picture The ROI Numbers 5 of 12 Evidence-Based Community Supervision of Adult Offenders: Three Findings from WSIPP Research Reviews +1% Intensive Supervision: Surveillance- Oriented (14)* Intensive Supervision: Treatment- Oriented (17)* Supervision: Focused on Risk, Treatment, & Response (6)* Change in Criminal Recidivism -10% ($1.93 b/c) -16% ($6.83 b/c) *The number of high-quality research studies on which this finding is based.

25. Results Application in WA Big Picture The ROI Numbers 5 of 12 1.Risk More crime can be avoided when policies focus on higher-risk (rather than lower-risk) offender populations. 2.Treatment (delivered with fidelity) Benefit-cost evidence indicates that some policies work and others do not. Careful selection and implementation needed. 3.Swift and Certain Apprehension/Punishment Clear evidence (for crime deterrence) for certainty, but not for severity of punishment. 4.Budget Drivers (aka: incentives) Deliver savings to taxpayers by tying policies to budget drivers (e.g. fiscal notes & caseloads); incentive funding formulas. Evidence-Based Polcies that Reduce Crime and Save Money: —four principles that improve the odds of success—