1. Sentencing Reform in California and Public Safety
Steven Raphael
Goldman School of Public Policy
University of California, Berkeley

3. What determines a nation’s incarceration rate?
Rate of admission
Offending rate x Apprehension rate x Conviction rate x Prison admission rate
Time served
Front-end sentencing
Release policy

6. U.S. vs. Australia?

8. Comparison of California Inmate Populations to the Design Capacity of State Institutions

9. Summary of Prison Overcrowding Litigation in California
1990: Coleman v. Brown
2001: Plata V. Brown
2006: Federal three-judge court established with authority over prison growth
2009: Three-judge court orders state to immediately offer plan to reduce prison population to percent of capacity. State appeals to the Supreme Court.
2011: Supreme Court upholds the population limit order by the three- judge panel.

10. Policy changes pushing towards lowering the prison population
2009
Performance incentives for counties to reduce probation revocation (Assembly Bill 678)
Introduction of non-revocable parole
2011
Corrections realignment
2012
California voters pass proposition 36 revising the scope of three-strikes
2014
California voted pass proposition 47 downgrading low level felonies to misdemeanors.

11. Provisions of Assembly Bill 109 (California Realignment)
Non-sexual, non-violent, non-serious (triple nons) felons will serve terms in county jail rather than state prison.
Triple nons currently serving time are monitored post-release under the Post Release Community Supervision program administered by counties. Technical violation will result in short jail terms but not returns to state prison.
Nearly all parole revocations are now served in county jail with a maximum of 180 days.
Enables greater use of non-custodial alternatives for both pre-trial and convicted jail inmates (primarily electronic monitoring).

15. Provisions of proposition 47
Passed by 60 percent of California voters
Reclassified various drug offenses from felonies or wobblers (crimes that can be charged as felonies or misdemeanors) to misdemeanors.
Requires misdemeanor sentencing for petty theft, receiving stolen property, and forgery/writing a bad check (offenses with amounts less than $950).

16. Prop 47 and Felony and Misdemeanor Drug Arrests

17. Prop 47 and All Drug Arrests

18. Prop 47 and Felony and Misdemeanor Property Crime Arrests

19. Prop 47 and Total Property Crime Arrests

20. Effects on prison and jail population
Realignment: large decline in prison population with partial offsetting increase in jail populations.
Proposition 47: declines in both prison and jail populations.

32. Comparison of Arrest Rates Pre-Post Prop 47 By Race/Ethnicity
Twelve months prior
Twelve months following
Change
Black Arrests per 100,000
884
840
-44
Black Booked Arrests per 100,000
672
610
-62
White Arrests per 100,000
278
266
-12
White Booked Arrests per 100,000
202
179
-23
Hispanic Arrests per 100,000
355
337
-18
Hispanic Booked Arrests per 100,000
253
233
-20

33. Avenues through which decarceration in California may impact crime rates (and factors that may mitigate these effects)
Incapacitation
General deterrence
Rehabilitation/specific deterrence, hardening/criminogenic influence
Diminishing crime-fighting returns to scale

39. Preferred estimates of the effect of a realignment-induced one-unit change in prison incarceration rates on part I crime rates

40. And proposition 47?

41. What explains small effects on crime?
Change in offending propensity
Diminishing returns to incarceration
Criminogenic heterogeneity among those who criminally offend
Expansion of the use of prison along the extensive margin
Net less serious offenders for less serious offenses
Expansion of the use of prison along the intensive margin
More likely to incarcerate offenders beyond the age of desistance

42. Evidence from other setting of diminishing marginal returns
Italy’s collective clemency
Dutch sentencing enhancement
Heterogeneity in prison-crime effect estimates at different points in time for the U.S.

43. The 2006 Italian Collective Clemency Bill
Passed on July 1, 2006
Reduced sentences of inmates convicted of certain felony offenses prior to May 2, by three years.
Led to the immediate release of one-third of the prison population on August 1,
Ineligible inmates include those convicted of offenses involving organized crime, sexual assault, terrorism, kidnapping, and exploiting a prostitute.
No post-released supervision.
Pardoned inmates who re-offend have their residual sentence tagged on to any new sentence for offenses occurring within five years.

44. Scatter plot of monthly incarceration rates against month measured relative to August 2006

46. Estimated Annual Crime Effects per Prison Year Served

47. Estimates of the Annual Effects of a Pardoned Inmate on Province Level Crime Totals

48. Estimated Impact of a Pardoned Inmate on Local Crime by Key Percentiles of the Per-Pardon Incarceration Rate

49. Analysis of Habitual Offender Sentence Enhancement in Holland 2001 (Study by Ben Vollard in the Economic Journal, 2012)
The Dutch pass a sentence enhancement in 2001 targeted at offenders with 10 or more prior felony offenses.
Increases sentence for theft from two months to two years.
Limited initially to 10 largest cities, but later (2004) expanded nationwide.
Ability to apply the sentence limited through a centralized rationing process of available prison space.
Resulted in cross-area differences in the extent to which localities were able to apply the sentence enhancement to their population of usual suspects.
Based on police records, Vollard estimates that there were roughly three to four thousands offenders whose criminal history made them eligible for the sentence enhancement.
Dutch incarceration rate at the time low by U.S. standards (124 per 100,000)

50. Relationship between recorded offenses in years preceding the policy change among those receiving the sentence enhancement and the proportion of eligible offenders sentenced under the new law

51. Vollard results
Large average annual incapacitation effects on the order of 50 to 60 offenses prevented per year served.
Moving from the 25th to the 75th percentile of the incapacitation ratio results in a 25 percent reduction in the incapacitation effect.

52. Average prison incarceration rate
Time period
Average prison incarceration rate
Joint incapacitation/deterrence effect per prison year served
1977 through 1988
171
9.93
1989 through 1999
349
1.246
2000 through 2010
449
2.21
Estimates from Raphael, Steven and Michael Stoll (2013), Why are So Many Americans in Prison?, Russell Sage Foundation Press, New York, NY.

53. Concluding thoughts
Much more room for selective incapacitation, especially in the United States where the incarceration rate is so high. Many states have incarceration rates that exceed that of California (whether measured before or after this slate of reforms).
While a sizable share of U.S. prison growth is attributable to changes in drug sentencing policy (roughly one-fifth) the lion’s share in state prisons is attributable to longer sentences for violent offenders (roughly one half of growth).
Many violent offenders are often thought to be low-risk of recidivism. Creates tension between utilitarian and just-desert objectives of corrections.