Statement for the Record of The ACLU, Justice Roundtable, and The Leadership Conference in Response to Department of Justice (DOJ) April 3 and 5 Listening Sessions

View a PDF of this letter here.

April 12, 2019

David B. Muhlhausen, Ph.D.
Director
National Institute of Justice
Department of Justice
950 Pennsylvania Ave NW
Washington, DC 20530

Submitted electronically

Re: Statement for the Record of The ACLU, Justice Roundtable, and The Leadership Conference in Response to Department of Justice (DOJ) April 3 and 5 Listening Sessions

Dear Director Muhlhausen,

On behalf of the American Civil Liberties Union, the Justice Roundtable, The Leadership Conference on Civil and Human Rights[1] and the undersigned organizations, we submit this statement for the record regarding the Department of Justice’s (DOJ) Listening Sessions on the development of a risk and needs assessment system as required by the First Step Act of 2018. Many of these organizations attended the listening sessions held on April 3 and 5 and made public statements consistent with the views set forth in this letter.

Transforming the criminal justice system is one of the most important civil and human rights issues of our time. Our organizations have long advocated for policy changes to advance racial justice, equality, and fairness within the criminal justice system and were closely involved in the negotiations and advocacy efforts to advance a criminal justice package that ultimately became The First Step Act. Through that process, we were committed to ensuring that reform would meaningfully address the front-end drivers of mass incarceration (i.e. mandatory minimum sentences), improve the lives of currently incarcerated individuals through increased rehabilitative programming, provide pathways for early release, safeguard against exacerbating existing disparities within the federal system, and ultimately yield a positive impact on as many federal prisoners as possible.

The First Step Act made some modest steps toward these goals, but in order to ensure that the legislation has the greatest positive impact, the Department of Justice must be committed to implementing the bill in a manner consistent with the text of the statute and Congressional intent. Therefore, we urge the Department to ensure that: (1) The NIJ appoints a new and appropriate “nonpartisan nonprofit” organization to host the IRC and select the membership as required by the statute; (2) Neither the Bureau of Prisons (BOP) security classification system nor the current version of PCRA is adopted as a substitute for the new Risk and Needs Assessment System required by the statute; and (3) The Bureau of Prisons immediately begins providing rehabilitative programming to everyone in federal prison.

1. The Department Must Identify a New Nonpartisan Nonprofit Organization to Host and Appoint the Independent Review Committee Before the Risk and Needs Assessment System Can be Developed.

Title I of The First Step Act is supposed to assist incarcerated individuals in success upon release and to reduce recidivism by providing rehabilitative programming and incentives for early release. The law mandates the use of a risk and needs assessment system in an unconventional and untested manner to determine appropriate programming and ability to receive earned time credits toward early transition to halfway houses, home confinement, or supervised release. This unconventional use, combined with other concerns about risk assessments in general, such as racial biases in risk factors, inaccuracies in identifying risk, and lack of independent testing and validation of tools, gave advocacy groups great concern. Our organizations worked diligently through the legislative process to ensure that the risk and needs assessment system as outlined in the First Step Act would not undermine the overall impact that the legislation could have.^[2] In order to mitigate some of those concerns, the First Step Act required the National Institute of Justice (NIJ) to select a “nonpartisan nonprofit organization with expertise in the study and development of risk and needs assessment tools” to appoint and host an Independent Review Committee (IRC) within 30 days of enactment and to begin implementation of the risk and needs assessment system no later than 210 days after enactment.^[3] The statute states that the NIJ  “shall” first select a nonpartisan nonprofit organization, and that organization “shall” then appoint the IRC’s members. The appointing organization must have “expertise in the study and development of risk and needs assessment tools,” and it must appoint not fewer than six members to the IRC, each of whom shall have expertise in “risk and needs assessment systems.”[4]

On April 8, approximately two and a half months after the IRC was to have been established, NIJ announced the organization it has appointed to appoint and host the IRC: the Hudson Institute. The selection of the Hudson Institute appears to be inconsistent with the requirement that the NIJ appoint “a nonpartisan nonprofit organization with expertise in the study and development of risk and needs assessment tools.” The Hudson Institute is known and described as a “politically conservative” think tank, whose research and analysis promotes global security, freedom and prosperity.[5] More specifically, its policy work and publications related to “legal affairs and criminal justice” seem to be solely focused on antitrust and national security public policy. There is no evidence on its website, in the form of research publications or otherwise, which remotely suggests the organization has any expertise or experience in the study and development of risk and needs assessment systems.[6] The only relevant evidence appears to be a single blog post, written on January 18, 2019.[7] Absent other evidence, the suggestion that the Hudson Institute has “expertise in the study of development of risk and needs assessment tools” strains credulity, especially given the variety and number of organizations that have this exact expertise. Further, the Hudson Institute has selected six members for the IRC, only three of whom may meet the required criteria for membership in the IRC outlined in the statute. The NIJ must immediately appoint a new nonpartisan nonprofit organization with expertise and experience in the study and development of risk and needs assessments systems to host the IRC and select its members.

Following the selection by the NIJ of a host organization to appoint expert IRC members, Congress next required the IRC to then provide an unbiased and independent review and evaluation of existing risk and needs assessment systems, and best practices with respect to design, testing and validation of these systems.[8] The law does not permit an alternate to the IRC. The IRC is also responsible for providing recommendations to the Attorney General to inform the final development, adoption and implementation of the new risk and needs assessment system by the BOP.^[9] The law mandates that the Attorney General consult with the IRC to develop risk and needs assessments in order to determine the amount and type of evidence based recidivism programming for each prisoner, and classify individuals into “risk” levels to be used to permit or deny incentives and rewards for successful participation.[10] The law does not allow the Attorney General to review and develop the assessment system independent of the IRC, which is what appears to have happened here.[11] This process is key to ensuring that only evidence and unbiased perspectives are used to develop the risk and needs assessment system. Without a truly non-partisan IRC, there is no way to ensure that the risk and needs assessment system created by BOP will operate in a fair and equitable manner.

2. The Department Cannot Use an Existing Tool as a Substitute for the Risk and Needs Assessment System Required by the First Step Act.

The BOP should not attempt to develop its own risk assessment tool internally or use any existing tool it may have at its disposal without consulting the IRC, as it seems to suggest in recent statements. A DOJ official is reported to have said that the Department of Justice “expects to meet the July deadline” because it is “using resources it has on hand to work on the risk assessment tool internally, in the absence of the committee.”[12] And a summary of DOJ policies on the First Step Act, dated February 7, 2019, stated:

At the outset, the Attorney General is required to develop a risk and needs assessment system to evaluate the recidivism of each inmate, who will be classified as presenting a low, medium, or high risk of recidivism. Alternatively, the Attorney General may “use existing risk and needs assessment tools as appropriate.” (If the Attorney General develops a new system, he must do so within 210 days of enactment of the Act, but that deadline may be subject to delay due to the lapse in appropriations that began on December 21, 2018.)

This policy suggests that the DOJ considers using an existing risk and needs assessment tool as a satisfactory substitute for compliance with the First Step Act. But the statute does not permit the Attorney General to independently adopt an existing tool outside the process established by the law. That process requires the Attorney General to consult with the IRC in carrying out each of his duties under sections 3631(b), 3632 and 3633, and requires the IRC to assist the Attorney General in carrying out those duties, including:

(1) conducting a review of the existing prisoner risk and needs assessment systems in operation on the date of enactment of this Act;

(2) developing recommendations regarding evidence-based recidivism reduction programs and productive activities;

(3) conducting research and data analysis on—(A) evidence-based recidivism reduction programs relating to the use of prisoner risk and needs assessment tools; (B) the most effective and efficient uses of such programs; and (C) which evidence-based recidivism reduction programs are the most effective at reducing recidivism, and the type, amount, and intensity of programming that most effectively reduces the risk of recidivism; and

(4) reviewing and validating the risk and needs assessment system.[13]

The only tool BOP currently uses is its security level classification system, which was primarily designed to assist in housing decisions upon entry into prison, not determine the likelihood that a person will recidivate upon release. Like many risk assessment systems, the BOP’s security classification system classifies as “high risk” many people who do not go on to reoffend. In addition, the BOP’s system heavily weights static factors such as age, past criminal history, and current offense—factors that cannot change while a person is in prison, therefore making it nearly impossible to lower risk categories in order to receive the new earned time credits. Under this system, males are placed in one of four security levels based on their number of points:  minimum (0-11 points); low (12-15 points); medium (16-23 points); high (24+).[14] The maximum base score is 45, with 42 points for the unchangeable factors of age, past criminal history, and current offense, and 3 points for the changeable factors of educational level and substance abuse. Similarly, the PCRA, a tool used by probation officers only to improve the reentry success of people on probation or supervised release in the community,[15] has a maximum score of 18 points, consisting of 9 points for the static factors of criminal history and age at intake, and 9 points for dynamic factors that could potentially change in the community, but many of which cannot change during incarceration.[16]

There is no existing tool that does what the First Step Act requires. Among other things, the statute requires development of tools that are both objective and statistically validated based on such factors as indicators of progress and regression and dynamic factors that can change while in prison. Decades of criminology research has shown that the factors that carry the most weight in BOP’s classification system and the PCRA are primarily the result of the behavior and decisions of police officers and prosecutors, rather than the individuals or groups that the data is claiming to describe.^[17] Because these factors “can exacerbate unjust disparities,” the Colson Task Force recommended that risk assessment tools be “employed solely to guide the individualized delivery of treatment and programming to improve reentry success.”[18] Likewise, one of the PCRA’s creators concluded that use of risk assessments for purposes other than to inform risk reduction efforts, such as determining the length of prison sentences, may “exacerbate racial disparities in incarceration.”[19]

Using a risk and needs assessment system to determine time credits is novel and untested. State correctional systems typically award time credits based on performance and/or disciplinary record, not a risk assessment.^[20] Risk assessments alone do not predict the recidivism risk of any person; they only roughly group people into a limited number of categories.^[21] When risk and needs assessment evaluations are adopted, they are typically used by states to identify programming for people in prison, rather than to award time credits.^[22]

Dynamic factors (i.e. those that can change over time), such as work history, family ties, and pro-social networks are nearly impossible to change while in prison and therefore make it very difficult for a person to lower his or her risk score during incarceration. Therefore, the use of any existing risk and needs assessments tool would result in a large number of people in prison being unable to earn early release credits from programming by decreasing their risk categories—contrary to the law’s mandate to provide all individuals in prison the incentive and opportunity to actively participate in programming throughout their entire term of incarceration.[23] A needs-based assessment should be used to identify the criminogenic needs of each individual and develop a program of interventions to address those needs to lower the individual’s risk of recidivating and actually help people succeed in their communities upon release, as the law requires.[24]

In addition, because communities of color are persistently over-policed across the nation and a person’s “criminal history” may not consist of actual criminal convictions, consideration of the static factors used by risk assessment systems bias the results against persons of color. Studies have shown that these tools can produce results that are heavily biased against Black defendants and have a disparate negative impact on Black people because the factors considered and the criminal justice data used by these systems are biased.^[25] Risk assessments rely on static factors, including criminal history and age at the time of the offense, and dynamic factors, including work history and educational achievement. Both static and dynamic factors tend to correlate with socioeconomic class and race, and studies show that Black people are more likely to be misclassified as high risk than White or Hispanic offenders.^[26]

For these reasons, the BOP security classification and the PCRA are not appropriate substitutes for the risk and needs assessment tools required by the First Step Act. They were not designed to identify the specific criminogenic needs of incarcerated individuals and heavily rely on static factors that would undermine the effectiveness of the system.

3. The Bureau of Prisons must immediately begin providing rehabilitative programming.

The core intent of the First Step Act is to provide rehabilitative and re-entry programming, as well as residential re-entry centers (i.e. halfway houses) and home confinement. The BOP does not currently provide minimally sufficient recidivism reduction programs, nor does it have sufficient halfway house capacity so that those released from prison can successfully transition to the community. Since 2017, BOP has relentlessly cut rehabilitative programming, staff, and halfway houses. There are 25,000 people in federal prison waiting to be placed in prison work programs,[27] at least 15,000 people waiting for education and vocational training,[28] and at least 5,000 people are awaiting drug abuse treatment.[29] There is nowhere near enough programming to help prisoners succeed in their communities upon release and thereby reduce recidivism overall.  We therefore urge BOP to begin rebuilding rehabilitative services now.

Further, any savings resulting from the First Step Act should be reinvested in rehabilitative programing so that as many people as possible can improve their lives and benefit from the new earned time credits. We also urge the Department of Justice to ensure that there is no privatization of public functions and to prevent private entities from unduly profiting from incarceration while implementing this legislation. In the end, any positive reform contemplated by the First Step Act is contingent upon sufficient funding to expand and improve evidenced-based recidivism reduction programming, and the availability of halfway house placements and home confinement.

Conclusion

For these reasons, we urge you to: (1) Immediately appoint a new nonpartisan nonprofit organization to host the IRC and select its members; 2) Ensure that the department adheres to the statute and does not use the current BOP security classification system or the PCRA as a substitute for the independently tested and validated risk and needs assessment tool required; and (3) Immediately direct resources to begin expanding rehabilitative programming in all federal prisons as required by the First Step Act. Thank you for your attention to these matters. If you have any questions, please feel free to contact Jesselyn McCurdy, American Civil Liberties Union Washington Legislative Office, Deputy Director at [email protected] (202) 675-2307 or Sakira Cook, Leadership Conference on Civil and Human Rights, Program Director, at [email protected] or (202) 263-2894.

Sincerely,

American Civil Liberties Union
Bread for the World
CURE (Citizens United for Rehabilitation of Errants)
Defending Rights & Dissent
Drug Policy Alliance
The Justice Roundtable
Justice Strategies
The Leadership Conference Education Fund
The Leadership Conference on Civil and Human Rights
NAACP Legal Defense and Educational Fund
The National Council for Incarcerated and Formerly Incarcerated Women and Girls (The Council)

[1] The American Civil Liberties Union (ACLU) is a nationwide organization working in courts, legislatures, and communities to defend and preserve the individual rights and liberties that the Constitution and the laws of the United States guarantee everyone in this country. The Justice Roundtable (Justice Roundtable) is a national coalition of legal, civil rights, criminal justice, human rights and faith-based organizations dedicated to advocating for a fairer federal criminal justice system. The Leadership Conference on Civil and Human Rights (The Leadership Conference) is the nation’s oldest and largest civil rights coalition representing people of color, women, children, older Americans, people with disabilities, gays and lesbians, major religious organizations, labor unions, and civil and human rights groups. For almost a half century, The Leadership Conference has led the fight for equal opportunity and social justice.

^[2] Vote “No” on The FIRST STEP Act. 2018. https://www.aclu.org/letter/vote-recommendation-first-step-act.

^[3] FIRST Step Act, Pub. L. No. 115-391, § 107. 2018.

[4] First Step Act, Sec. 107(b)-(d).

[5] https://en.wikipedia.org/wiki/Hudson_Institute

[6]https://www.hudson.org/topics/101-crime-and-justice

[7] Walters, John P. & Tell, David “Criminal Justice Reform and the First Step Act’s Recidivism Reduction Provisions: Preliminary Issues for Policymakers,” Hudson Institute, Jan. 18, 2019 available at https://www.hudson.org/research/14776-criminal-justice-reform-and-the-first-step-act-s-recidivism-reduction-provisions-preliminary-issues-for-policymakers.

[8] The risk and needs assessment system shall be used to-(1) determine the recidivism risk of each prisoner as part of the intake process, and classify each prisoner as having minimum, low, medium, or high risk for recidivism;(2) assess and determine, to the extent practicable, the risk of violent or serious misconduct of each prisoner; (3) determine the type and amount of evidence-based recidivism reduction programming that is appropriate for each prisoner and assign each prisoner to such programming accordingly, and based on the prisoner’s specific criminogenic needs, and in accordance with subsection (b);(4) reassess the recidivism risk of each prisoner periodically, based on factors including indicators of progress, and of regression, that are dynamic and that can reasonably be expected to change while in prison;(5) reassign the prisoner to appropriate evidence-based recidivism reduction programs or productive activities based on the revised determination to ensure that- all prisoners at each risk level have a meaningful opportunity to reduce their classification during the period of incarceration to address the specific criminogenic needs of the prisoner; and all prisoners are able to successfully participate in such programs; (6) determine when to provide incentives and rewards for successful participation in evidence-based recidivism reduction programs or productive activities in accordance with subsection (e) (7) determine when a prisoner is ready to transfer into prerelease custody or supervised release in accordance with section 3624; and (8) determine the appropriate use of audio technology for program course materials with an understanding of dyslexia. 18 U.S.C. § 3632 (a) (b).

^[9] Ibid.

[10] Ibid.

[11] Ibid.

[12] Roscoe, Ayesha. “3 Months into New Criminal Justice Law, Success For Some And Snafus For Others”, NPR, Apr. 1, 2019. https://www.npr.org/2019/04/01/708326846/3-months-into-new-criminal-justice-law-success-for-some-and-snafus-for-others.

[13] First Step Act, Sec. 107(a), (e); see also 18 U.S.C. §§ 3632(a)-(b), 3633.

[14] Program Statement P5100.08, Ch. 5, Pg.2.

[15] See “An Overview of the Federal Post-Conviction Risk Assessment.”. Office of the United States Courts, Office of Probation and Pretrial Services, 2011. Pgs.1-3.

[16] The PCRA is scored as follows:  GED or less than high school (1 point); currently unemployed even if a student, homemaker, or retired (1 point); unstable work history over the past 12 months (1 point); current alcohol problem (1 point); current drug problem (1 point); single, divorced, separated (1 point); unstable family situation (1 point); lack of positive pro-social support (1 point); attitude toward supervision and change (1 point).  Supervisees are assigned to one of four risk categories based on their number of points:  low (0-5 points); low/moderate (6-9 points); moderate (10-12 points); high (13-18 points).  See Thomas H. Cohen et al., Does Change in Risk Matter? Examining Whether Changes in Offender Risk Characteristics Influence Recidivism Outcomes, 15 Criminology & Pub. Pol’y 263, 271-72, 294 (2016); PCRA Officer Section, http://www.ned.uscourts.gov/internetDocs/jpar/RGK-FSR2014-PCRA%20Officer%20Section.pdf.

^[17] Vote “No” on The FIRST STEP Act. 2018. https://civilrights.org/resource/vote-no-first-step-act/

[18] See “Transforming Prisons, Restoring Lives: Final Recommendations of the Colson Task Force on Federal Corrections.” Jan. 2016. Pg.34.

[19] Skeem, Jennifer L. & Lowenkamp, Christopher, “Risk, Race, and Recidivism:  Predictive Bias and Disparate Impact”, 54 Am. Soc’y of Criminology 680, 703-05, 2016.

^[20] Ibid.

^[21]James, Nathan “Risk and Needs Assessment in the Criminal Justice System”, U.S. Congressional Research Service, No. R44087, June 22, 2015.

^[22] See e.g., Tex. Gov’t Code § 498.002 (classifying inmates’ “time-earning category” based on factors other than risk assessment); R.I. Gen. Laws § 42-56-24 (determining amount of sentence credit based on factors other than risk assessment); Okla. Stat. § 57-138 (same); N.C. Gen. Stat. § 15A-1340.18 (same); Ohio Rev. Code Ann. § 2967.193 (same).

[23] 18 U.S.C. §§ 3621(h)(6), 3632(d).

[24] See 18 U.S.C. §§ 3633, 3635(3).

^[25] See, e.g., Skeem, J. & Lowenkamp, C. (2015). Risk, Race & Recidivism: Predictive Bias and Disparate Impact. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2687339.

^[26]Whiteacre, Kevin. “Testing the Level of Service Inventory-Revised (LSI-R) for Racial/Ethnic Bias”, 17 Crim. Just. Pol’y Rev. 330, 2006.; see also Fennessy, Matthew & Huss, Matthew T., “Predicting Success in a Large Sample of Federal Pretrial Offenders: The Influence of Ethnicity”, 40 Crim. Just. & Behav. 40. Jan. 2013. 53.

[27] See Federal Bureau of Prisons, UNICOR, https://www.bop.gov/inmates/custody_and_care/unicor_about.jsp.

[28] See “Oversight of the Federal Bureau of Prisons”, U.S. House of Representatives, Com. on the Judiciary, Subcom. On Crime, Terrorism, Homeland Security, and Investigations. Apr. 17, 2018. Pg.50

[29] See Dep’t of Justice, Bureau of Prisons, Drug Abuse Treatment Program, 81 Fed. Reg. Apr. 16, 2016,.24484, 24488,