Alaska Criminal Justice System Assessment Alaska ...ajc.alaska.gov/acjc/docs/resources/reinvestment/pewpresent8-2015.pdfSystem Assessment Alaska Commission on Criminal Justice August

8/2/2015

1

Alaska Criminal Justice

System Assessment Alaska Commission on Criminal Justice

August 3, 2015

Anchorage, AK

1

Presentation Outline

System assessment

Pretrial

Sentencing

Community supervision

Prison growth and costs

Next steps

Subgroups and schedule

8/2/2015

2

2

Sources

Data analysis

Alaska Department of Corrections probation/parole data

Alaska court case file review

Statutory and policy review

Interviews with system stakeholders

Department of Corrections, judiciary, Department of Law, public

defenders, Division of Behavioral Health, Mental Health Trust

3

PRETRIAL

8/2/2015

3

4


System assessment

Pretrial

o Pretrial research

o Pretrial practices in Alaska

Sentencing



Next steps


Summary Takeaways

5

PRETRIAL RESEARCH

8/2/2015

4

6

Pretrial Risk Assessment

Pretrial

Risk = the likelihood of a negative future outcome

Pretrial risk = the likelihood of failure to appear in court or

new criminal activity during the pretrial period

Actuarial risk assessment tools are more accurate than

professional judgment alone

Have higher predictive validity

Reduce disparities across judicial districts

Source: Mamalian (2011), State of the Science of Pretrial Risk Assessment

7

Risk Assessment Tools Predict Likelihood of Pretrial

Failure

Pretrial

Source: Lowenkamp & VanNostrand (2013), Assessing Pretrial Risk Without a Defendant

Interview

6% 9%

13% 15%

20%

5%

9%

15%

20% 23%

0%

5%

10%

15%

20%

25%

30%

35%

40%

45%

50%

Low Low-moderate Moderate Moderate-high High

Rate

of

Failu

re

Risk Level

Likelihood of Pretrial Misconduct, by Risk Level (Kentucky PRA-S)

Failure to Appear New Criminal Activity

8/2/2015

5

8

Use Risk Scores to Guide Decisions About Release

Conditions

Pretrial

Courts should use risk scores to guide decisions about

release conditions

Examples of release conditions: Third-party custodian, electronic

monitoring, drug and alcohol testing, etc.

When applied to higher-risk defendants, restrictive release

conditions lead to better pretrial outcomes

When applied to lower-risk defendants, restrictive

conditions lead to worse pretrial outcomes

Source: VanNostrand (2009), Pretrial Risk Assessment in the Federal Court

9

Third-Party Custodians Helpful For Higher-Risk

Defendants, Not For Lower-Risk Defendants

Pretrial

Source: VanNostrand (2009), Pretrial Risk Assessment in the Federal Court

56%

30%

8%

-6%

-17%

-30%

-20%

-10%

0%

10%

20%

30%

40%

50%

60%

Risk Level 1 Risk Level 2 Risk Level 3 Risk Level 4 Risk Level 5

% C

han

ge i

n L

ikelih

oo

d o

f P

retr

ial

Failu

re

Percent Change in Likelihood of Pretrial Failure for Federal Defendants with Third Party Custodian Condition, by Risk Level, 2009

8/2/2015

6

10

Pretrial Detention Can Lead to Worse Outcomes

Pretrial

Pretrial detention can lead to worse outcomes, particularly for

low-risk defendants

Low-risk defendants detained longer than 24 hours are:

Less likely to appear for court

More likely to engage in new criminal activity during the pretrial

period

More likely to recidivate long-term

Source: Lowenkamp, VanNostrand & Holsinger (2013), The Hidden Cost of Pretrial

Detention

11

For Low-Risk Defendants, Detention Longer than 24

Hours Associated with Increased Criminal Activity

Pretrial

Source: Lowenkamp, VanNostrand & Holsinger (2013), The Hidden Cost of Pretrial

Detention

+39% +50%

+56% +57%

+74%

0%

10%

20%

30%

40%

50%

60%

70%

80%

2-3 days 4-7 days 8-14 days 15-30 days 31+ daysIncre

ased

Lik

elih

oo

d o

f N

ew

Cri

min

al A

cti

vit

y

Co

mp

are

d t

o D

ete

nti

on

Un

der

24 H

ou

rs

Days Detained Pretrial

Increased Likelihood of New Criminal Activity for Low-Risk Defendants Compared to Those Detained Under 24 Hours

8/2/2015

7

12

Money Bond Not The Most Effective Tool To Protect

The Public During The Pretrial Period

Pretrial

Ability to pay money bond ≠ low-risk

There are low-risk defendants who are unlikely to engage in new criminal

activity but who can’t afford bail

And there are high-risk defendants who are likely to engage in new criminal

activity who can afford bail

Source: Schnacke (2014), Money As a Criminal Justice Stakeholder: The Judge’s

Decision to Release or Detain a Defendant Pretrial

13

Unsecured Bonds Are As Effective At Achieving

Court Appearance As Secured Bonds

Pretrial

Source: Jones (2013), Unsecured Bonds: The As Effective and Most Efficient Pretrial

Release Option

97%

87%

80%

43%

93%

85%

78%

53%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Low Low-Moderate Moderate High

% A

pp

ear

for

Co

urt

Risk Level

Court Appearance Rates By Bond Type, by Risk Level

Unsecured Secured

8/2/2015

8

14

Unsecured Bonds Are As Effective At Achieving

Public Safety As Secured Bonds

Pretrial


Release Option

93%

84%

69% 64%

90%

79%

70%

58%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Low Low-Moderate Moderate High% C

om

ple

te P

retr

ial

Peri

od

Wit

ho

ut

Arr

est

Risk Level

Public Safety Rates by Bond Type, by Risk Level

Unsecured Secured

15

Defendants With Unsecured Bonds Released At

Higher Rates Than Those With Secured Bonds

Pretrial


Release Option

93% 95% 96%

88% 83%

65%

54%

46%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Low Low-Moderate Moderate High

% R

ele

ased

Pre

tria

l

Risk Level

Pretrial Release Rates by Bond Type, by Risk Level

Unsecured Secured

8/2/2015

9

16

Appearance Rates Increase When Courts Make

Accommodations For Defendants

Pretrial

Court date reminders (mail, email, phone call, text)

Night court for defendants who work during the day

Remote participation in court hearings by teleconference

Source: Bechtel, Holsinger, Lowenkamp & Warren (2015), A Meta-Analytic

Review of Pretrial Research: Risk Assessment, Bond Type, and Interventions

17

Research Summary

Pretrial

Pretrial risk assessment can help predict likelihood of

pretrial failure

Courts should use risk scores to guide release decisions, and focus release

conditions on higher-risk defendants

Pretrial detention can lead to worse outcomes, particularly

for low-risk defendants

Secured financial bond increases pretrial detention, without increasing

pretrial success

8/2/2015

10

18

PRETRIAL PRACTICES IN ALASKA

19

817

1,479

0

200

400

600

800

1,000

1,200

1,400

1,600

2005 2006 2007 2008 2009 2010 2011 2012 2013 2014

Pretrial Defendants on July 1, by Year

Number of Pretrial Detainees Up 81% in Last Decade

Pretrial

Source: Alaska Department of Corrections

8/2/2015

11

20

Pretrial Admissions Down 13%, But Still High

Numbers of Nonviolent Misdemeanor Admissions

Pretrial

1,865 2,014

4,058 2,748

4,615 4,106

12,623

11,171

0

5,000

10,000

15,000

20,000

25,000

2005 2014

Number of Pretrial Admissions, 2005 and 2014, by Type

Nonviolent Misd.

Violent Misd.

Nonviolent Felony

Violent Felony


21

Pretrial Admissions: Citation vs. Arrest

Pretrial

Officers permitted by statute to issue citations for nonviolent

misdemeanors rather than arrest

However, lack of statutory guidance on when to cite vs.

arrest

8/2/2015

12

22

34

24

10 6

67

44

16

9

0

10

20

30

40

50

60

70

80

Felony Violent Felony Nonviolent Misd. Violent Misd. Nonviolent

Days

Mean Length of Stay for Pretrial Detainees, 2005 and 2014 (Days)

2005

2014

Pretrial Misdemeanor Defendants Defendants Staying Longer Pretrial Than They

Used To


23

Alaska Court File Review

Pretrial

Purpose: To examine pretrial release conditions and

time to first release

Sampled court case files from Anchorage, Fairbanks,

Juneau, Bethel, and Nome

Randomly selected from DOC release cohort for each jurisdiction

Of case files reviewed, 310 files had bail conditions for analysis

Data entry and analysis by Pew and Alaska Judicial Council

8/2/2015

13

24

Few Defendants Released On Their Own

Recognizance Or On Unsecured Bond

Pretrial

Statute presumes release on recognizance or unsecured

bond

In practice, only 12% of defendants sampled were

released on their own recognizance, and an additional

10% had unsecured bond

Source: Alaska Court File Review

25

Less than Half of Sampled Defendants Are Released

From Prison Pretrial

Pretrial

Released 48%

Not Released 52%

Percent of Sampled Defendants Released Pretrial


8/2/2015

14

26

Release Linked to Ability to Pay Rather Than

Defendant’s Risk

Pretrial

Pretrial risk assessment not used in decisions about

whether to release or detain, or in setting conditions of

release

Because secured bond is ordered in the majority of cases,

release is often linked to ability to pay rather than the

defendant’s risk of pretrial failure


27

Money Bond Set In Two-Thirds Of Cases

Pretrial

Secured Bond

Required 67%

Secured Bond Not Required

33%

Percent of Sampled Defendants With Secured Bond Requirements


8/2/2015

15

28

41% of Bonds Set At $2,500 or More

Pretrial

11%

18%

31%

41%

0%

5%

10%

15%

20%

25%

30%

35%

40%

45%

Under $500 $500-$999 $1,000-$2,499 $2,500+

Percentage of Sampled Defendants with Secured Bond Amounts, by Category


29

Lower Release Rates For Higher Bond Amounts

Pretrial

64%

43%

38%

33%

0%

10%

20%

30%

40%

50%

60%

70%

Under $500 $500-$999 $1,000-$2,499 $2,500+

Percent of Sampled Defendants With Secure Bond Released, By Amount Category


8/2/2015

16

30

Longer Detention Before First Release For

Defendants With Higher Bond Amounts

Pretrial

4 Days 6 Days

52 Days 50 Days

0

10

20

30

40

50

60

Under $500 $500-$999 $1,000-$2,499 $2,500+

Avera

ge T

ime t

o R

ele

ase (

Days)

Average Time to First Release For Released Sampled Defendants With Secure Bond, by Dollar Amount


31

Other (Non-Financial) Release Conditions Not Tied

To Risk

Pretrial

While courts have statutory authority to order non-financial

release conditions

Examples: Third-party custodians, drug and alcohol monitoring,

home arrest, etc.

No actuarial tool used to guide decisions on release

conditions

Unknown whether more restrictive conditions are focused on

higher-risk defendants

8/2/2015

17

32

Third-Party Custodian Conditions Required In

Addition to Money Bond

Pretrial

Third-party custodian required for defendant’s release in

23% of cases sampled

Of those defendants with a third-party custodian condition,

all also had money bail conditions


33

Three-Quarters of Defendants With Third Party

Custodian Conditions Not Released Pretrial

Pretrial

Released 25%

Not Released 75%

Percentage of Sampled Defendants with Third Party Custodian Condition Who Were Released


8/2/2015

18

34

Accommodations To Increase Court Appearance

Pretrial

No statewide court date reminder system (mail, email,

phone call, text) to increase court appearance rates

Reports of some courts making ad hoc accommodations to

increase court appearance such as conducting hearings by

teleconference

35

Less Serious Cases Not Prioritized For Speedier

Trial

Pretrial

No statute on speedy trial timelines

Court rule on speedy trial:

Prioritizes scheduling of cases for defendants in custody

over defendants who have been released

But does not prioritize scheduling of misdemeanor

cases over felony cases

Speedy trial time limit for felonies, misdemeanors, and

violations is 120 days

8/2/2015

19

36

Questions For Policy Development

Pretrial

What guidance can be provided to law enforcement regarding

cite vs. arrest for misdemeanor offenses?

How can Alaska incorporate evidence-based risk tools into

pretrial decision-making?

What accommodations can be made to increase court

appearance rates?

How can court processes be streamlined to reduce pretrial

lengths of stay?

37

SENTENCING

8/2/2015

20

38

2,303

2,627

1,500

1,700

1,900

2,100

2,300

2,500

2,700

2,900

2005 2006 2007 2008 2009 2010 2011 2012 2013 2014

Sentenced Offenders on July 1, by Year

Number of Sentenced Inmates Up 14% in Last

Decade


Sentencing

39


System assessment

Pretrial

Sentencing

Prison admissions and recidivism

Prison length of stay and recidivism



Next steps


8/2/2015

21

40

PRISON ADMISSIONS AND

RECIDIVISM

41

Prison Admissions and Recidivism: Current Practice

in Alaska Research summary–

Incarceration is not more effective at reducing

recidivism than non-custodial sanctions

Current practices in Alaska—

Large – though declining – number of misdemeanants

sentenced to prison

Number of nonviolent felons sentenced to prison up

slightly

Despite options, limited use of prison alternatives

outside of probation

Sentencing

8/2/2015

22

42

Despite Declines, 82% of Prison Admissions Are

Misdemeanants; Nonviolent Felons Up Slightly

536 554

1,287 1,327

2,235 2,093

7,573 6,569

0

2,000

4,000

6,000

8,000

10,000

12,000

14,000

2005 2014

Number of Admissions for Sentenced Offenders, 2005 and 2014, by Type

Nonviolent Misd.

Violent Misd.

Nonviolent Felony

Violent Felony


Sentencing

43

Sentencing Option Eligibility

Prison/jail (including time served) Felons and some misdemeanants

Therapeutic courts Misdemeanants and some felons

Probation Misdemeanants and some felons

Suspended imposition of sentence Misdemeanants and some felons

Fine Misdemeanants and few felons

In Sentencing Offenders, 5 Primary Options

Sentencing

8/2/2015

23

44

Prison and Probation Used Often; Limited Use of

Other Options

13,885

15,237

1,471

323

Prison Admissions Probation Dispositions SIS Dispositions Therapeutic Court Opt-Ins

Placements, 2014

Source: Alaska Department of Corrections; Alaska Court System; Therapeutic Courts Coordinator

Sentencing

45

Court Type Available Slots

Statewide, FY15

Average

Utilization, FY15

Range

Drug/DUI/Veteran 192 67% 73% - 99%

Mental Health 115 88% 53% - 94%

Felony Offenders Therapeutic Courts Underutilized

Sentencing

And Lack Statewide Practice Standards

Source: Therapeutic Courts Coordinator

High risk/high need participants that could benefit most

from program potentially screened out prior to undergoing

risk assessment

8/2/2015

24

46

Availability Of Therapeutic Courts Differ

Geographically

Court Drug Court DUI Court MH Court

Anchorage X X X

Fairbanks X

Juneau X X

Ketchikan X X

Palmer X

Bethel X

32 other courts

Sentencing

Source: Therapeutic Courts Coordinator

47

LENGTH OF STAY AND RECIDIVISM

8/2/2015

25

48

Length of Stay: Current Practice in Alaska

Research summary: Longer prison stays do not reduce

recidivism more than shorter stays

Current practices in Alaska:

Misdemeanant length of stay down slightly

All felony offense types staying in prison longer over

previous decade

Nonviolent: Property and drug offenders staying a month

longer; alcohol and public order offenders staying about

three months longer

Violent: Person offenders staying about three months

longer; sex offenders staying 15 months longer

Sentencing

49

Felony Offenders In 4 out of 5 Major Categories, Misdemeanor Length

of Stay Down

19

58

45 40

35

18

43 43

36 42

0

10

20

30

40

50

60

70

80

90

Alcohol Person Property Public Order Transportation

Days

Mean Length of Stay for Sentenced Offenders with Misdemeanor Convictions, 2005 and 2014, By Top 5 Offense Category

2005

2014

Sentencing


8/2/2015

26

50

Sentencing Felony Offenders Felony Length of Stay Up Across Nonviolent

Offense Categories

231 238

171

117

260 276 269

224

0

50

100

150

200

250

300

Property Drugs Alcohol Public Order

Days

Mean Length of Stay for Sentenced Offenders Convicted of Nonviolent Felony Charges, 2005 and 2014, by Offense Category

2005

2014


51

472

561 554

1,033

0

200

400

600

800

1000

1200

Person Registerable Sex Offense

Days

Mean Length of Stay for Sentenced Offenders Convicted of Violent Felony Charges, 2005 and 2014, by Offense Category

2005

2014

Felony Offenders Felony Length of Stay up 17% for Person Offenders;

86% for Sex Offenders

Sentencing


8/2/2015

27

52

Sentencing Felony Offenders Questions for Policy Development

Can Alaska increase availability of its alternatives to incarceration?

How can the state ensure its therapeutic courts are achieving the

maximum recidivism reduction?

What opportunities exist to target longer prison stays on more

serious offenders and shorter prison stays on less serious

offenders?

Can Alaska further focus prison beds on serious violent offenders?

53

COMMUNITY SUPERVISION

8/2/2015

28

54


System assessment

Pretrial

Sentencing


Focus on high risk offenders and target criminogenic needs

Use swift, certain, and proportionate sanctions

Incorporate rewards and incentives

Incorporate treatment into supervision

Frontload resources


Next steps


55

Community Corrections Growth Outpacing Prison

Growth

Community Supervision

-

2,000

4,000

6,000

8,000

10,000

12,000

2005 2006 2007 2008 2009 2010 2011 2012 2013 2014

DOC Populations on July 1, 2005-2014

Prison Population Probation Parole Population

CRC Population EM Population


8/2/2015

29

56

Almost Two-Thirds of Offenders Released Return to

Prison Within Three Years



71%

63%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

FY 2002 FY 2011

Percentage of Offenders Released Who Return to Prison Within 3 Years, FY 2002 and 2011

57

Number of Supervision Violators in Prison Up 15% in

Last Decade

1,013

1,161

0

200

400

600

800

1,000

1,200

1,400

2005 2014

Number of Supervision Violators in Prison, July 1, 2005 and 2014



8/2/2015

30

58

Recidivism Reduction Principles


Focus on high risk offenders and target criminogenic needs

Use swift, certain, and proportionate sanctions for violations

Incorporate rewards and incentives

Incorporate treatment into supervision

Frontload resources

59

Risk, Needs, Responsivity: Research Principle


Risk: Focus resources on higher-risk offenders

Needs: Use supervision and programming to target

criminogenic needs

Responsivity: Address barriers to program success

Source: Andrews (1999), Recidivism Is Predictable and Can Be Influenced: Using

Risk Assessments to Reduce Recidivism

8/2/2015

31

60


Alaska uses risk/needs assessment tools to identify risk of

reoffending and treatment needs, and uses scores to inform

supervision levels and case planning

However, large portion of community supervision resources still

focused on low-risk offenders

Even with lower supervision standards, low-risk offenders make up a large

share of caseloads and require staff resources that would otherwise be

dedicated to offenders who are more likely to reoffend

Significant portion of the halfway house population is not

assessed, potentially leading to mixing of risk levels

Risk, Needs, Responsivity: Current Practices

61

39% of Probation/Parole Population Low-Risk


1,602

2,105

382

-

500

1,000

1,500

2,000

2,500

Minimum (0-18) Medium (19-28) Maximum (29-55)

LSI-R Classifications on July 1, 2014


8/2/2015

32

62

Pretrial Population in Halfway Houses Unassessed

for Risk Level


Sentenced 70%

Pretrial 30%

Halfway House Population On July 1, 2014, by Status


63

Swift, Certain, and Proportional Sanctions:

Research Principle


Respond to problem behavior in a manner that will

change that behavior

Deterrence:

Swift, certain, and proportional sanctions have a stronger

deterrent effect than delayed, random, and severe

sanctions

Source: Nagin & Pogarsky (2001), Integrating Celerity, Impulsivity, and Extralegal

Sanction Threats into a Model of General Deterrence: Theory and Evidence

8/2/2015

33

64

Swift, Certain, and Proportional Sanctions:

Current Practices


PACE program incorporates swift and certain responses

PACE probation imposes swift, certain and proportional jail stays for higher-risk

offenders who violate supervision conditions

However, only applies to a small portion of offenders on community supervision

For standard probation and parole, no system-wide framework

for swift, certain, and proportional sanctions

There is a system in place for revoking offenders to prison, but no statewide

system for intermediate sanctioning

Some sanctioning processes are inconsistent with swift, certain, and

proportionate principles, including long delays between the problem behavior

and the response, and disproportionately long revocation sentences

65

Supervision Violators

35

33

0

5

10

15

20

25

30

35

40

2005 2006 2007 2008 2009 2010 2011 2012 2013 2014

Days

Mean Length of Stay for Unsentenced Supervision Violators, 2005-2014 (Days)

Petitions To Revoke Take A Month to Resolve



8/2/2015

34

66

Nearly Half Of Revocations Staying More Than

One Month; 29% More Than Three Months

Swift, Certain, Proportionate Community Supervision

344

407 423

283 312

153

57

839

675

552

359 333

111

52

0

100

200

300

400

500

600

700

800

900

0-7 Days 8-30 Days 1 to 3 mths 3 to 6 mths 6 mths to 1year

1 to 2 years 2 or moreyears

Number of Supervision Violators Released in 2005 and 2014, by Length of Stay

2005

2014


67

Incorporate Rewards and Incentives:

Research Principle


Provide rewards and incentives for meeting case-specific

goals of supervision to enhance individual motivation

Develop a continuum of rewards to round out the continuum

of sanctions

Higher program completion when rewards outnumber

sanctions

Source: Wodahl, Garland, Culhane & McCarty (2011), Utilizing Behavioral Interventions

to Improve Supervision Outcomes in Community-based Corrections

8/2/2015

35

68


For offenders in prison, statute authorizes good time and

furlough incentives to reward positive behavior and program

participation

However, for offenders on community supervision, no system-

wide framework exists to incentivize and reward positive

behavior and compliance

No statutory mechanism for earned compliance credits

Variation in practices limit the use of early termination as a motivational tool

Incorporate Rewards and Incentives:

Current Practices

69

Incorporate Treatment Into Supervision:

Research Principle


Incorporate treatment into supervision case plans rather

than using surveillance alone

Utilize cognitive-behavioral treatment and community-

based drug and alcohol treatment

Source: Washington State Institute for Public Policy (2012), available at:

http://www.wsipp.wa.gov/BenefitCost?topicId=2

8/2/2015

36

70

Incorporate Treatment Into Supervision:

Current Practices


Division of Probation and Parole uses LSI-R to identify

criminogenic needs with top priority needs forming the

basis of offender case management plans

Significant efforts to increase access to treatment and

programming, but still dramatic unmet treatment needs

Insufficient inpatient and outpatient treatment beds and

qualified treatment providers

Regional disparities in community-based treatment and

programming

71

0 200 400 600 800 1,000 1,200 1,400 1,600

Screening

Assessment/Referrals

ANSAT

LSSAT

RSAT

Aftercare

Department of Correction Substance Abuse Services, FY 2010 – FY 2014

FY10

FY11

FY12

FY13

FY14


DOC Substance Abuse Services Increasing

Incorporate Treatment into Supervision Community Supervision

8/2/2015

37

72 Source: Alaska Department of Corrections

DOC Community-Based Treatment


Community-based intensive outpatient and continuing

care programs in 5 communities Anchorage

Fairbanks

Juneau

Kenai

Palmer

In FY15, 21% of the community supervision

population accessed DOC community-based

treatment

73

Significant Unmet Treatment Need in Alaskan

Population


13%

87%

Percentage of Alaskans With Alcohol Dependance or Abuse Who Recieved Treatment (2009-2013)

11%

89%

Percentage of Alaskans With Illicit Drug Dependance or Abuse Who Received Treatment

(2009-2013)

Source: SAMSHA (2014), Behavioral Health Barometer: Alaska, 2014

8/2/2015

38

74

Frontload Resources: Research Principle


Focus supervision and programming resources during

the initial weeks and months following release from

prison when violations and arrests are most likely to

occur

75

Frontload Resources: Current Practices


Significant recent efforts to improve transition in the first

months following release from prison

Improved re-entry planning policies, creation of re-entry coalitions, hiring re-

entry coordinators

Revocations in Alaska most likely to happen in the first months

following release from prison; however, supervision resources

allocated well beyond these initial months

Moreover, supervision terms have increased over past decade

While probation officers have some discretion to reduce

supervision levels over time, options are limited and potentially

arbitrary

8/2/2015

39

76

Failure Most Likely To Happen In First Three Months


62%

17%

11%

3%

8%

0%

10%

20%

30%

40%

50%

60%

70%

0-3 months 4-6 months 7-12 months 13-24 months 25-36 months

Time Served on Probation / Parole Before Return to DOC, 2014


77

Average Length of Stay on Community Supervision

Up 13%


23.5 months

26.5 months

0

5

10

15

20

25

30

2005 2006 2007 2008 2009 2010 2011 2012 2013 2014

Mo

nth

s

Mean Length of Supervision for Successful Discharge, 2005-2014


8/2/2015

40

78

Limited and Potentially Arbitrary Process To Reduce

Offenders’ Supervision Levels Over Time


Probation and parole officers have the discretion to:

Reduce supervision levels over time

Place low-risk offenders on an administrative caseload

Recommend certain offenders for early termination of probation

However, there are opportunities to streamline and standardize

step down process in order to most efficiently allocate

resources

79

Most Offenders on Active Medium or Minimum

Supervision; 10% Inactive



626

1,179

2,882

526 764

0

500

1,000

1,500

2,000

2,500

3,000

3,500

Inactive/MinimumBank

Minimum Medium Maximum Unclassified

Number of Offenders by Supervision Level, July 2015

8/2/2015

41

80

Questions For Policy Development


What measures can be put in place to focus community

supervision resources more on higher-risk offenders?

Are there opportunities to expand use of swift, certain, and

proportional sanctions on supervision?

Are there opportunities to incentivize positive behavior and

reward compliance with supervision conditions?

What opportunities exist for expanded training and quality

assurance processes?

81


System assessment

Pretrial

Sentencing



Next steps


8/2/2015

42

82

Spending on Corrections Up 60% Over Past 2

Decades

Projected Growth and Costs


$126 Million

$327 Million

0

50

100

150

200

250

300

350

Dep

art

me

nt

of

Co

rre

cti

on

s O

pe

rati

ng

E

xp

en

dit

ure

s,,

in

Mil

lio

ns

Department of Corrections Operating Expenditures, FY 1995-2014

*Figures do not include capital expenditures; 60% based on inflation-adjusted numbers

83

Community Supervision Accounts For 43% Of DOC

Population, But Only 6% of DOC Budget


84%

6%

10%

DOC Spending by Placement Type, FY15

47%

43%

10%

DOC Population by Placement Type, July 1, 2014


8/2/2015

43

84

Absent Further Reform, Prison Population Projected

to Grow 27%, Costing At Least $169 Million



3,903

5,095

6,511

-

1,000

2,000

3,000

4,000

5,000

6,000

7,000Historical and Projected Prison Growth, 2004-2024

Historical Average Daily Population Projected Average Daily Population

Current Hard Bed Capacity

85

Methodology for Projected Growth and Costs


Prison population projected to grow by 1,416 beds in 10 years

By 2017, the population will exceed current hard bed capacity,

requiring the reopening of a closed facility (128 beds)

By 2018, the population will exceed expanded capacity,

necessitating out of state contracting

Accommodating the projected growth will cost taxpayers at

least $169 million

8/2/2015

44

86

Questions?

87


System assessment

Pretrial

Sentencing



Next steps


8/2/2015

45

88

A

Sentencing

B

Community

Supervision

C

Pretrial

Alex Bryner

(chair)

Ron Taylor

(chair)

Trevor Stephens

(chair)

Craig Richards Kris Sell Terry Vrabec

Quinlan Steiner Jeff Jessee John Coghill

Wes Keller Stephanie Rhoades Brenda Stanfill

Greg Razo

Policy Subgroups

Next Steps

89

Subgroup A Wed., Sept. 9, 9:00 a.m. – 12:00 p.m.

Wed., Oct. 14, 9:00 a.m. – 12:00 p.m.

Wed., Nov. 18, 9:00 a.m. – 12:00 p.m.

Subgroup B Wed., Sept. 9, 1:00 – 4:00 p.m.

Wed., Oct. 14, 1:00 – 4:00 p.m.

Wed., Nov. 18, 1:00 – 4:00 p.m.

Subgroup C Thurs., Sept. 10, 9:00 a.m. – 12:00 p.m.

Thurs., Oct. 15, 9:00 a.m. – 12:00 p.m.

Thurs., Nov. 19, 9:00 a.m. – 12:00 p.m.

Full Commission Thurs., Oct. 15, 1:00 – 4:00 p.m.

Thurs., Nov. 19, 1:00 – 4:00 p.m.

Thurs., Dec. 10, 10:00 a.m. – 1:30 p.m.

Meeting Schedule

Next Steps

8/2/2015

46

90

Contact Info

Terry Schuster

Office: 202.540.6437

Email: [email protected]

Emily Levett

Office: 202.540.6732


Melissa Threadgill

Office: 617.435.8386


Rachel Brushett

Office: 202.540.6915


Public Safety Performance Project

www.pewtrusts.org/publicsafety

mailto:[email protected]




http://www.pewtrusts.org/publicsafety

Documents

Alaska Criminal Justice System Assessment Alaska ...ajc.alaska.gov/acjc/docs/resources/reinvestment/pewpresent8-2015.pdfSystem Assessment Alaska Commission on Criminal Justice August