Dear House Judiciary Committee Member:
On behalf of The Leadership Conference on Civil and Human Rights and the 74 undersigned organizations, we urge you to vote “No” on the FIRST STEP Act that will be considered during the mark up. Any effort to pass prison reform (or “back-end” reform) legislation without including sentencing reform (or “front-end” reform) will not meaningfully improve the federal system. Across the country, states that have enacted legislation containing both front and back end reforms have reduced rates of incarceration and crime. Any legislation that addresses only back end reforms is doomed to fail in achieving these goals. Without changes to sentencing laws that eliminate mandatory minimums, restore judicial discretion, reduce the national prison population, and mitigate disparate impacts on communities of color, the FIRST STEP Act alone will have little impact. The Leadership Conference on Civil and Human Rights plans to include your position on The FIRST STEP Act in our voting scorecard for the 115th Congress.
Moreover, proposals referred to by the White House and others as “prison reform,” including the FIRST STEP Act and S.1994, the CORRECTIONS Act, would do little to reform prisons or the federal justice system. The FIRST STEP Act attempts to allow people to participate in reentry and rehabilitation programs and earn time credits that would permit them to serve the end of their prison sentence in home confinement or halfway houses. However, currently there are not enough of these programs available in the Federal Bureau of Prisons (BOP) to serve those currently in prisons. Furthermore, BOP more recently has reduced the number of residential reentry centers it contracts with to provide halfway house programming.
In addition, many people would not be eligible to earn credits by participating in rehabilitation or reentry programs merely based on their criminal convictions. Even if a person is deemed eligible to participate in BOP programming based on the required risk and needs assessment system, after participating in programming, the BOP warden could deny the transfer of prisoners to a halfway house or home confinement if they find “by clear and convincing evidence that the prisoner should not be transferred into prerelease custody based on evidence of the prisoner’s actions after the conviction of such prisoner” and provides “a detailed written statement regarding such finding.” The federal criminal justice system is deeply flawed and needs to adopt a top to bottom overhaul. The FIRST STEP Act alone does not come close to achieving the desperately needed reforms to create a fair and just system.
The FIRST STEP Act is Unlikely to Achieve Meaningful Prison Reform.
The FIRST STEP Act, as introduced, would exclude too many in people in federal prisons from receiving time credits for participating in the rehabilitative programs authorized by the bill. The long list of exclusions in the bill sweep in, for example, those convicted of certain immigration offenses and drug offenses. Because immigration and drug offenses account for 53.3 percent of the total federal prison population, many people could be excluded from utilizing the time credits they earned after completing programming. Furthermore, these exclusions could also have a disparate impact on racial minorities, since the majority of those held in federal prison for immigration and drug offenses are people of color. Any person who will return to our communities from prison someday should get time credit incentives for completing rehabilitative programs. Any reforms enacted by Congress should impact a significant number of people in federal prison and reduce racial disparities or they will have little effect on the fiscal and human costs of incarceration.
Moreover, the purported incentives towards rehabilitation are not real or meaningful. The FIRST STEP Act’s earned time credits are not real time off a sentence, but more time in a halfway house or home confinement. This is inadequate. Limited space already reduces the amount of time individuals can spend in halfway houses. Recent closures of residential reentry centers have further exacerbated the problem, making it unlikely that people will be able to use all the “time” they earn under the bill.  Existing programs have been successful precisely because they provide real time off of an individual’s sentence. For example, the residential drug abuse program (RDAP) provides a one-year sentence reduction to those who complete the program. However, RDAP currently faces a 5,000 person wait list due to limited resources. Additionally, home confinement is rarely used by the Bureau of Prisons and the bill eliminates the option of community supervision altogether. For the incentive structure to be real, earned time credits must equate to an actual reduction in sentence to encourage individuals to engage in rehabilitative programming. Such a real incentive structure would result in fiscal savings. For example, if only one in nine individuals earned 60 days of credit in a year, $100 million in savings would be realized. Congress should be following the model of RDAP in any prison reform and give people real time off their sentences, not a promise of more of something they already cannot get.
Finally, the bill does not include any funding for the recidivism reduction programming it seeks to expand, already grossly underfunded. The bill authorizes only $50 million in funding per year for five years for rehabilitative programs in federal prisons. However, there is currently no guarantee that such funding will ever actually be appropriated. Even if such funding is actually appropriated, the authorization level will not adequately support the expanded programming necessary to implement the system. Further, while the bill says any savings should be reinvested in programming, it retains a provision permitting the Attorney General to recommend how any savings should be spent, including on “law enforcement,” effectively creating the option for savings not to be reinvested. Further, the bill also provides that in order to expand programming, BOP shall enter into partnerships with private organizations and companies under policies developed by the Attorney General, “subject to appropriations.” This could privatize what should be public functions and could allow private entities to unduly profit from incarceration. In the end, any positive reform contemplated by the FIRST STEP Act is heavily contingent upon sufficient funding to create or expand evidenced based recidivism reduction programming and access to a halfway house or home confinement, which is highly unlikely. The bill, as drafted, is therefore an empty promise, unlikely to achieve meaningful prison reform and unlikely to reduce crime or rates of mass incarceration.
The FIRST STEP Act Provides No Incentive to Those Most in Need of Rehabilitative Programming and is Unlikely to Reduce Recidivism
The FIRST STEP Act is unlikely to reduce recidivism because it focuses time credit incentives for completing rehabilitative programs on minimum- and low-risk category prisoners who need less rehabilitation and intervention and who likely shouldn’t have been sentenced to a term of imprisonment in the first place, not on medium- and high-risk prisoners who are more in need of the incentives to complete programs. Only minimum- and low-risk category prisoners can “cash in” the time credit incentives they earn, and these prisoners are able to earn more time credits than medium- and high-risk prisoners. This approach is not evidence-based. Data has demonstrated that effectively reducing recidivism requires focusing programs, jobs, and real and meaningful incentives on those most likely, not least likely, to reoffend. While the bill rightly attempts to incentivize participation in rehabilitation programs, it wrongly ties those credit incentives to an individual’s risk category. Conversely, state correctional systems typically award time credits based on performance and/or disciplinary record, not a risk and needs assessment and according to research these systems should be used to identify appropriate correctional interventions, not to set the length of prison sentences. Ultimately, the bill takes a flawed and untested approach that is unlikely to reduce recidivism – a result that could dissuade future prison reforms and embolden critics of reform who believe that “nothing works.”
Instead of this approach, earned time credits should vest at the end of each year, which will enable BOP to adjust sentences automatically and incentivize participation in recidivism reduction programs. In particular, we recommend that individuals at any risk level should be eligible to earn ten days of earned time credit for every 30 days of programming completed, vesting at the end of each year. Earned time credits that lead to actual sentence reductions – earlier release from confinement altogether – are powerful incentives for participation and meaningful rewards for individuals committed to their personal rehabilitation and reentry. Automatic adjustments would eliminate delays in prerelease, decrease court costs, and allow BOP and courts to focus their time elsewhere.
One important aspect of the FIRST STEP Act is that it relies on the use of an undeveloped profile-based algorithmic risk assessment evaluation. Within 180 days after the enactment, the Attorney General would be required to develop and adopt a risk and needs assessment system that would be used to categorize federal prisoners as minimum, low, medium, or high risk. However, the bill also gives the Attorney General and BOP Director overly broad discretion to “use existing risk and needs assessment tools as appropriate” to implement the system. This is problematic, as it could allow the Attorney General to use the BOP’s current security classification system, a system that is not designed to identify specific criminogenic needs and heavily relies on static factors, as proxy for the risk and needs assessment tool, which would ultimately undermine the purpose and effectiveness of the system. Moreover, the FIRST STEP Act fails to mandate the implementation of safeguards that are necessary whenever such tools are used. For example, the bill does not require that the risk and needs assessment tool be statistically validated prior to adoption, doesn’t account for the time needed to complete one in advance, and does not require that these tools be validated by an independent authority that has no stake in the outcome of each validation analysis. It is also absolutely vital that algorithmic-based tools, if they are used at all, are designed and calibrated with input from the community and those who would be impacted by their use, rather than input from Administration officials alone. Finally, any use of a risk and needs assessment tool must establish a mechanism by which every assessed individual has a meaningful opportunity to review and challenge their designation as high, medium, low or minimum risk. Assessment instruments like the one proposed by this bill can be expensive to design, implement, and validate, and can unfortunately result in unintended, harmful consequences for individuals and communities, especially communities of color.
The FIRST STEP Act Uses Risk Assessment Tools in an Unconventional Manner, and These Assessment Tools are Often Unreliable and Exacerbate Racial and Socioeconomic Disparities.
Using a risk 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. Research shows that risk assessments often do not accurately predict risk and risk assessments often classify many people as high risk who do not reoffend. One study showed that only 52 percent of those assessed as moderate or high risk by risk assessment tools went on to commit any offense, meaning that almost half of all persons classified as moderate or high risk were actually low risk. Another study found that risk assessments were no better at predicting recidivism than regular human beings provided with the same information. An assessment of the tool used by the Administrative Office of the United States Courts, the Federal Post Conviction Risk Assessment (PCRA), found that 58 percent of offenders on probation or supervised release classified by the PCRA as high risk are not re-arrested. Risk assessments do not predict the recidivism risk of any person; they only roughly group people into a limited number of categories. 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.
In addition, risk assessments often heavily rely on static factors (those that cannot change) such as criminal history, family members’ criminal history, and the community in which a person lived before entering the criminal justice system. Given that communities of color are persistently over-policed across the nation and that a person’s “criminal” history need not include any actual criminal convictions, consideration of these factors will likely bias the results against persons of color. Dynamic factors (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 their risk score during incarceration. Therefore, the FIRST STEP Act will result in a large number of people in prison unable to earn early release credits from programming by decreasing their risk category. Rehabilitative programs in prison should use a needs-based assessment 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.
Finally, relying on a risk assessment tool for earning time credits could amplify racial disparities and perpetuate other injustices in the criminal justice system. Studies have shown that these tools can produce results that are heavily biased against Black defendants and have a disparate negative impact on African Americans. 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 African Americans are more likely to be misclassified as high risk than White or Hispanic offenders. Therefore, although risk assessments may seem objective or neutral, the data driving many predictive algorithms is profoundly limited and biased. Furthermore, decades of criminology research has shown that such data primarily documents the behavior and decisions of police officers and prosecutors, rather than the individuals or groups that the data is claiming to describe.
The FIRST STEP Act Does Attempt to Fix the “Good Time” Credit Calculation.
The federal prison system’s method of calculating earned credit reduces a prisoner’s sentence to a maximum credit of 47 days per year – below the 54 days that Congress intended. This decision results in unnecessary increases in time served by prisoners, at significant cost. The FIRST STEP Act attempts to fix the good time credit calculation, such that prisoners would receive 54 days of good time credit per year, not 47 days, for following prison rules, which is a prospective fix only. Further, many have suggested that this fix could immediately impact 4,000 individuals if made retroactive, however this number has not been verified. According to a BOP calculation done over a decade ago, approximately 4,000 people could be eligible for release within a year, not on the first day of implementation. While the “good time fix” is a much needed, positive reform, which should become law, this provision alone is not enough to overcome our overwhelming concerns with the core of the bill as outlined above.
It is important to note that while reforms to address back-end drivers of our prison system are needed, they cannot function as a substitute for front-end sentencing reform. Only front-end reforms have the power to significantly stem the tide of incarceration, reduce the exorbitant cost of the prison system, and give redress to those inside who are serving sentences that are disproportionate to the severity of the offense. Any approach that does not include sentencing reform will be insufficient to meet the challenges we face. Our continued progress toward meeting the economic and societal challenges posed by the current system and establishing a fair and more just system depends on a comprehensive approach to reform.
It is up to Congress to continue to advance front end and back end reform designed to improve both federal sentencing laws and the functioning of the federal prison system. If Congress is serious about addressing meaningful prison reform, it will pass legislation that would deal with the conditions of confinement such as reducing the use of solitary confinement, providing adequate medical care to prisoners, and addressing exorbitant prison phone rates. While we appreciate the inclusion of some promising provisions in the introduced bill, such as prohibitions on the shackling of pregnant women, reforms to the federal compassionate and elderly release programs, and an audit of the program several years after its implementation, these changes are not significant enough to overcome our primary concerns with the bill and many could be adopted administratively by the Bureau of Prisons. Furthermore, we remain concerned that the challenges and solutions to reforming our federal prison system have not been fully explored by this committee and that no hearings have been held in order to give due consideration to the FIRST STEP Act in particular.
For the foregoing reasons, we urge you to vote “No” on the FIRST STEP Act. If you have any questions, please feel free to contact Sakira Cook, Senior Counsel at The Leadership Conference on Civil and Human Rights, at (202) 263-2894 or firstname.lastname@example.org, or Jesselyn McCurdy, Deputy Director of the American Civil Liberties Union Washington Legislative Office, at (202) 675-2307 or email@example.com.
The Leadership Conference on Civil and Human Rights
African American Ministers In Action
American Civil Liberties Union
American Federation of Labor-Congress of Industrial Organizations
American Humanist Association
Asian Pacific American Labor Alliance
Association of University Centers on Disabilities (AUCD)
Autistic Self Advocacy Network
Bend the Arc Jewish Action
Buried Alive Project
Campaign for Youth Justice
Center for Responsible Lending
Coalition of Black Trade Unionists
Coalition on Human Needs
Color of Change
CURE (Citizens United for Rehabilitation of Errants)
Defending Rights & Dissent
Disability Rights Education & Defense Fund
Drug Policy Alliance
Equal Justice Society
Evangelical Lutheran Church in America
Faith Action Network – Washington State
Government Information Watch
Harm Reduction Coalition
Hip Hop Caucus
Human Rights Watch
Jewish Council for Public Affairs
Law Enforcement Action Partnership
Let’s Start, Inc.
Life for Pot
NAACP Legal Defense and Educational Fund, Inc.
National Association of Human Rights Workers
National Association of Social Worker
National Bar Association (NBA)
National Black Justice Coalition
National Center for Lesbian Rights
National Coalition on Black Civic Participation
National Council of Churches
National Employment Law Project
National Hispanic Media Coalition
National Immigration Law Center
National Juvenile Justice Network
National LGBTQ Task Force Action Fund
National Organization for Women
National Religious Campaign Against Torture
NETWORK Lobby for Catholic Social Justice
People For the American Way
Prison Policy Initiative
Service Employees International Union (SEIU)
Sikh American Legal Defense and Education Fund
Southern Poverty Law Center
Students for Sensible Drug Policy
The Decarceration Collective
The National Council for Incarcerated and Formerly Incarcerated Women and Girls
The United Church of Christ
The United Methodist Church – General Board of Church and Society
T’ruah: The Rabbinic Call for Human Rights
Union for Reform Judaism
United Church of Christ
United Church of Christ, Justice and Witness Ministries
United We Dream
Washington Lawyers’ Committee for Civil Rights & Urban Affairs
We Got Us Now
334 East 92nd Street Tenant Association
 See “Most States Cut Imprisonment and Crime.” The Pew Charitable Trusts. Nov. 10, 2014. http://www.pewtrusts.org/en/multimedia/data-visualizations/2014/imprisonment-and-crime.
 See FIRST STEP Act. 115th Cong. § 13632(d)(4)(D)(xlii), (xliii), (2018).
 See “Offenses.” Federal Bureau of Prisons. Last Updated Feb. 24, 2018. https://www.bop.gov/about/statistics/statistics_inmate_offenses.jsp.
 See Taxy, Sam et. Al. “Drug Offenders in Federal Prison: Estimates of Characteristics Based on Linked Data.” Bureau of Justice Statistics. Oct. 2015. https://www.bjs.gov/content/pub/pdf/dofp12.pdf. See also Carson, E. Ann. “Prisoners in 2016.” Bureau of Justice Statistics. Jan. 2018. https://www.bjs.gov/content/pub/pdf/p16.pdf.
 See FIRST STEP Act, 115th Cong. § 404(a)(1)(A) (2018).
 See Lynch, Sarah N. and Harte, Julia. “Exclusive: Trump Administration Reduces Support for Prisoner Halfway Houses.” Reuters. Oct. 13, 2017. https://www.reuters.com/article/us-usa-justice-prisons-exclusive/exclusive-trump-administration-reduces-support-for-prisoner-halfway-houses-idUSKBN1CI2ZA.
 Office of Inspector General. (2016, November) Audit of Federal Bureau of Prisons’ Management of Inmate Placements in Residential Reentry Centers and Home Confinement. U.S. Department of Justice. Retrieved from https://oig.justice.gov/reports/2016/a1701.pdf.
 For example, participation in the BOP’s residential drug abuse program (RDAP) results in a one-year sentence reduction and has had as many as 7,000 people on its waiting list. See American Bar Association. (n.d.). Residential Drug Abuse Program (RDAP). Retrieved from https://www.americanbar.org/content/dam/aba/administrative/individual_rights/Ch 3-RDAP.authcheckdam.pdf.
 See Walrath, Rowan. “Software Used to Make “Life-Altering” Decisions Is No Better Than Random People at Predicting Recidivism.” Mother Jones. Jan. 2018. https://www.motherjones.com/crime-justice/2018/01/compas-software-racial-bias-inaccurate-predicting-recidivism/
 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)
 See, e.g., Skeem, J. & Lowenkamp, C. (2015). Risk, Race & Recidivism: Predictive Bias and Disparate Impact. Retrieved from https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2687339.
 The U.S. Supreme Court upheld BOP’s methodology against a challenge brought by inmate petitioners.41 However, BOP officials told us that the agency was supportive of amending the statute, and had submitted a legislative proposal to Congress such that 54 days would be provided for each year of the term of imprisonment originally imposed by the judge, which would result in inmates serving 85 percent of their sentence.42 BOP provided us estimates in December 2011 showing that if the GCT credit allowance was increased by 7 days, as proposed, BOP could save over $40 million in the first fiscal year after the policy change from the early release of about 3,900 inmates. As of December 2011, the legislative proposal had not been introduced on the floors of the House or Senate.