Examining Prison
Releases in
Response to COVID:
LESSONS LEARNED FOR REDUCING THE
EFFECTS OF MASS INCARCERATION

Kelly Lyn Mitchell
Julia Laskorunsky
Natalie Bielenberg
Lucy Chin
Madison Wadsworth
July 2022

Acknowledgements
This report was prepared with support from Arnold Ventures. We are grateful for their financial support
and for the work of their Criminal Justice program staff for recognizing the pandemic as an opportunity
to further study back-end release discretion and in helping us to conceive of this project.
Thanks to our three law student research assistants from the University of Minnesota: Natalie Bielenberg,
Lucy Chin, and Madison Wadsworth. It was a journey to figure out what information to survey and how
to make sense of the treasure trove of information that was ultimately collected. We are grateful for their
hard work and insightful observations, and for serving as sounding boards as we reasoned through the
analysis.

The Robina Institute of Criminal Law and Criminal Justice is a research institute within the
University of Minnesota Law School. We perform interdisciplinary research, bringing legal
and social science research together with policy analysis and practice to develop policyfocused solutions to issues within the field of criminal justice.

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

i

Table of Contents
Executive Summary.................................................................................................................................................................................... iii
Introduction..................................................................................................................................................................................................... 1
Prison Release Mechanisms................................................................................................................................................................... 2
Routine Prison Release Mechanisms..........................................................................................................................................................2
Non-Routine Prison Release Mechanisms..............................................................................................................................................4
Assumptions About How COVID Releases Would Occur............................................................................................................6
Understanding COVID Releases.......................................................................................................................................................... 7
Prevalence of Release..............................................................................................................................................................................................7
Legal Mechanisms Used to Accomplish Release..............................................................................................................................10
Criteria for Release.....................................................................................................................................................................................................14
Revisiting Our Assumptions About Release................................................................................................................................ 21
Releases in Determine vs. Indeterminate Jurisdictions................................................................................................................21
Releases Due to Compassionate/Geriatric Provisions....................................................................................................................22
COVID as a Political Issue.....................................................................................................................................................................................26
Effect on Prison Population Size......................................................................................................................................................... 28
Lessons Learned and Recommendations..................................................................................................................................... 29
Conclusion........................................................................................................................................................................................................ 33
Appendix A. Prison Releases and Institutional Features by Jurisdiction................................................................... 34
Appendix B. Release Mechanisms..................................................................................................................................................... 36
Appendix C. Combinations of Release Criteria by Jurisdiction and Release Group................................................... 45
Appendix D. Compassionate Release and Geriatric Parole Provisions by Jurisdiction................................................50
Appendix E. Narrative Descriptions of COVID Prison Releases by Jurisdiction ..................................................... 64

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

ii

Executive Summary
In response to the global pandemic in 2020, states and the federal government began to
make non-routine releases from prison in order to reduce prison populations to allow for
social distancing in prison facilities. This report is aimed at describing where such prison
releases occurred, the legal mechanisms used to achieve these releases, and the factors within
jurisdictions that made non-routine prison releases more or less likely to occur. We write this
report, not to examine the national response to the pandemic, but to better understand when
and how extraordinary measures may be used to effect prison release, and to determine
whether there are lessons from this experience that can be applied to reducing the effects of
mass incarceration.

Prevalence of
Release:

∙ We estimate that a total of 80,658 people were released from prisons in 35
jurisdictions (34 states and the federal prison system) due to COVID-related
policies, which was equivalent to about 5-1/2% of the total state and federal
prison population in 2019.
∙ Most COVID-related releases were quite modest, amounting to the
equivalent of less than 10% of the 2019 prison populations in 27 of the 35
jurisdictions in which releases occurred (Figure 2).

Legal
Mechanisms:

∙ The legal mechanisms used most frequently to release people from
prison during the pandemic were parole (11 jurisdictions), compassionate
release (10 instances in 9 jurisdictions), home confinement (8 jurisdictions),
commutation (7 jurisdictions), and good time or earned time credits (6
jurisdictions) (Figure 3).

Criteria for
Release:

∙ Type of crime, COVID health risk, and time left to serve on one’s sentence
were the criteria most frequently used—either alone or in combination—to
determine eligibility for release due to COVID-related policies.
» Most release groups (39 of 73) required that a person had to have been
convicted of a non-violent offense (Figure 4).
» COVID health risks—addressing both medical vulnerability and age—were
used as criteria in 38 of 73 release groups (Figure 6).
» Most release groups (37 of 73) required that a person have a short time left
to serve on their sentence (Figure 7). Though the amount of time varied
from 30 days in New Mexico to 5 years in Kentucky, the average was 9
months, and the most frequently used time period was 12 months.

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

iii

Examining Prison Releases in Response to COVID 		

Political and
Structural
Influences:

Executive Summary

∙ Gubernatorial leadership played a larger role in whether the jurisdictions
made releases, with fewer jurisdictions with Republican leadership making
releases. However, determinacy may have affected how many releases
were possible, with indeterminate jurisdictions making larger releases than
determinate jurisdictions regardless of political leadership.
» All but three Democratic-led jurisdictions (21 of 24) made COVID-related
prison releases while only about half of Republic-led jurisdictions (14 of
27) did so (Table 4).
» Nearly all of the jurisdictions (7 of 8) with the largest COVID-related
releases—those greater than 10% of the 2019 prison population—were
indeterminate in structure.

Lessons Learned and Recommendations:

1

States and the federal government have the tools to make large-scale
releases, but some modifications are needed. When willing to do so, states
and the federal government can accomplish large-scale prison releases –
mainly using mechanisms already available to them. However, during the
pandemic, this required addressing barriers to making these mechanisms
effective for large-scale releases, such as expanding eligibility requirements,
taking a top-down rather than case-by-case approach, or assigning resources
to expediting considerations for parole and other pre-release reviews.
Jurisdictions could consider making these modifications permanent.

2

The pandemic forced jurisdictions to think differently about technical
violations of supervision. Several jurisdictions reduced sanctions for technical
violations of community supervision or refused to admit people to prison
for technical violations. All jurisdictions could use this as an opportunity to
rethink whether prison is a necessary response to such violations.

3

Jurisdictions could reduce prison populations by increasing resources for
back-end release procedures. In many jurisdictions, all that was required to
achieve release was a more concerted effort to complete the steps needed
for release. Thus, jurisdictions could reduce prison populations simply by
putting more resources into make release processes flow faster.

4
A .

Jurisdictions were risk averse in their approach to identifying people for
release. Most releases focused on people who were convicted of “nonviolent” crime or were very close to their release date. Rather than relying
on this false dichotomy, corrections officials should focus on enhancing
rehabilitative programming, and legislatures should fund those programs
at the levels needed to address crime. Additionally, jurisdictions should
learn more about public attitudes towards crime and public safety to better
understand the feasibility of continuing or expanding second look provisions.
ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

iv

Examining Prison Releases in Response to COVID

5

Efforts at using back-end release powers to reduce prison populations may
have been hampered by politics. Our analysis showed that the political party
of the governor in each state bore a relationship to whether the state chose
to make COVID-related prison releases. Criminal justice is a hot button issue
often characterized by the false dichotomy of being “soft on crime” or “tough
on crime.” Thus, in order to address mass incarceration, it may be necessary to
redefine what it means to be “tough on crime.”

6

Outside pressure may be needed to encourage back-end releases. In several
jurisdictions, back-end prison releases were triggered by external forces. Thus,
outside pressure may be needed to encourage, or make available, the broader
use of back-end release discretion to reduce prison population size.

Conclusion:

A .

Executive Summary

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

Though jurisdictions had back-end release powers, they were hampered
by procedural barriers in using them on a large scale. Finding ways to
overcome those barriers was largely a product of political will. And in some
cases, outside pressure in the form of litigation was necessary to prompt
government and corrections officials to act. Even when they overcame the
barriers to using back-end release mechanisms, jurisdictions took a very
conservative approach to back-end releases, focusing on areas where they
believed there was less risk to public safety or where they thought they could
garner more public support, such as releasing people who committed nonviolent offenses or who had very little time left to serve on their sentence. As
a result, while some jurisdictions were able to release a sizeable number of
people due to the pandemic, the people released tended to be individuals
that were close to being released anyway. Thus, the experience from the
pandemic informs us that jurisdictions are unlikely to tackle the issue of mass
incarceration by using their discretionary back-end release authority unless
we address their risk aversion by redefining what it means to be “tough on
crime.” Some suggested ways to do this include assessing public attitudes for
institutionalizing second look processes and eliminating the false dichotomy
of “non-violent” and “violent” offending and instead focusing on increasing
the availability of rehabilitative programming and incentives for people in
prison to engage in that programming.

v

Introduction
In 2020, life across the globe changed dramatically when the COVID-19 virus took root and began to
spread, eventually growing into the most significant pandemic in the past century. As the number of
people falling ill and dying from the virus began to grow, it quickly became apparent that society would
have to make significant changes to slow its spread. Early evidence indicated that the virus spread more
quickly when people were in close contact for 15 minutes or more and that people who were older
or who had certain health conditions were more vulnerable to sickness, severe disease, and death.
In response, the nation began to shut down. If close contact facilitated virus spread, then an action
that governments and private companies could take would be to disrupt the systems that resulted in
close contact. Schools and churches were closed, workers were sent home, and hospitals and nursing
homes were closed to visitors, all in an effort to disrupt social gatherings and facilitate what came to
be known as “social distancing.” Despite the widespread concern, however, prison was one place where
people remained closely gathered, and where people could neither be easily home sent nor could social
distancing be easily achieved.
At the same time, prison populations remained historically high. Though prison populations have been
decreasing since their all-time high in 2009, they have not yet returned to the levels experienced prior to
the onset of mass incarceration.1 And in many places, individual prison facilities were also overcrowded.
Thus, the prison environment presented dramatic risk to those who were incarcerated. It is no wonder
that activists began to call for people to be released from prison to reduce the population significantly
enough to allow for some form of social distancing even within the prison walls. As some activists put it,
without action, every prison sentence could become a death sentence.2
But as this paper will explain, prison release is a complicated business. The laws that govern prison
sentences are designed to ensure that people are held accountable for their offenses by serving a fair
portion of the sentence pronounced by the court. They are not necessarily designed to facilitate release
from prison, especially prior to a person having served the minimum time required by state and federal
laws. Nevertheless, 34 states and the federal government did manage to facilitate non-routine prison
releases during the height of the pandemic. This report is aimed at describing where such prison releases
occurred, the mechanisms used to achieve these releases, and the factors within jurisdictions that made
non-routine prison releases more or less likely to occur. We write this report, not to examine the national
response to the pandemic, but to better understand when and how extraordinary measures may be
used to effect prison release, and to determine whether there are lessons from this experience that can
be applied to reducing the effects of mass incarceration.

1.
2.

E. Ann Carson, Prisoners in 2019, Bureau of Justice Statistics, https://bjs.ojp.gov/content/pub/pdf/p19.pdf.
Lauren-Brooke Eisen and Ruth Sangree, Brennan Center for Justice, COVID-19 is Turning Prison Terms into Death Sentences,
https://www.brennancenter.org/our-work/analysis-opinion/covid-19-turning-prison-terms-death-sentences#:~:text=By%20
failing%20to%20act%2C%20Abbott,in%20prison%20from%20Covid%2D19, last visited March 24, 2022.

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

1

Prison Release Mechanisms
In order to understand how extraordinary COVID releases were, it is important to first understand
how people are ordinarily released from prison. It is rarely the case that the prison sentence
pronounced at sentencing is the length of time a person will actually serve in prison. Instead, all
jurisdictions have complex rules for determining the first date at which a person is eligible for
release and for ultimately releasing that person into the community.3 In this section, we describe
how people are ordinarily released from prison as well as several non-routine release mechanisms
that commonly exist in state law. We then explain which mechanisms we assumed would be used
to effect non-routine releases during the pandemic.

Routine Prison Release Mechanisms
Generally, prison release processes vary depending on whether the jurisdiction is determinate or
indeterminate. The concept of determinacy refers to how definite the amount of time to be served
in prison is based on the sentence pronounced at the time of conviction. As explained in the Robina
Institute’s Degrees of Indeterminacy project,
“Indeterminacy” means “unpredictability of time served.” … If actual time-that-will-be-served
is highly unpredictable based on the pronounced judicial sentence, then the sentence is
highly indeterminate. If actual time-to-be-served is knowable within a relatively small range
of possibility, then the sentence has a low degree of indeterminacy—or, we might say—it has a
high degree of determinacy. “Determinacy” means “predictability of time served” at the time of
judicial sentencing.4
As shown in Table 1, 34 states have an indeterminate sentencing system while 16 states and the federal
government have a determinate sentencing system.
For those jurisdictions that are indeterminate, a parole board determines release for a majority of people
in prison. State law will usually contain provisions requiring a person to serve a certain proportion of their
pronounced sentence before being considered for release. The proportion of time that must be served
may vary depending on the type of crime. For example, in Texas, people convicted of lower level offenses
are eligible for release when their time served and any good time credit equals 25% of their pronounced
sentence whereas people who have been convicted of more severe offenses, including kidnapping, sex
trafficking, and assault, are required to serve 50% of their pronounced sentence before being eligible
for release.5 Once a person reaches this eligibility date, the person will undergo a review by the parole
board to determine whether they can be released from prison to parole, which is a period of post-prison
supervision.

3.
4.
5.

Kevin R. Reitz, Edward E. Rhine, Allegra Lukac, Melanie Griffith, American Prison Release Systems: Indeterminacy in Sentencing
and the Control of Prison Population Size, Robina Institute of Criminal Law and Criminal Justice, forthcoming 2022.
This definition is included in a series of state reports on prison release discretion. See, e.g., Kevin R. Reitz, Allegra Lukac, &
Edward E. Rhine, Prison Release Discretion and Prison Population, State Report: California at iii (2021), https://robinainstitute.
umn.edu/sites/robinainstitute.umn.edu/files/california_doi_report_10_8_21.pdf.
Tex. Gov’t Code § 508.145 (d), (f) (2021).

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

2

Prison Release Mechanisms

Examining Prison Releases in Response to COVID 		

Table 1. Sentencing and Prison Release System by State
Indeterminate

Determinate

Alabama

Kentucky

Nevada

South Dakota

Arizona

Minnesota

Alaska

Louisiana

New Hampshire

Tennessee

California

New Mexico

Arkansas

Maryland

New Jersey

Texas

Delaware

North Carolina

Colorado

Massachusetts

New York

Utah

Federal

Ohio

Connecticut

Michigan

North Dakota

Vermont

Florida

Oregon

Georgia

Mississippi

Oklahoma

West Virginia

Illinois

Virginia

Hawaii

Missouri

Pennsylvania

Wyoming

Indiana

Washington

Idaho

Montana

Rhode Island

Kansas

Wisconsin

Iowa

Nebraska

South Carolina

Maine

*Washington, D.C., which is a determinate system, is not represented on the table because all prisoners from that jurisdiction
are housed by the Federal Bureau of Prisons.

The parole board will consider a variety of factors in determining whether a person is ready for release,
including the circumstances and severity of the crime, the person’s record of behavior in prison and any
disciplinary actions for misconduct, whether the person has participated in or completed any treatment
while in prison, risk assessment scores, and mental health evaluations.6 Some states have instituted
more routinized processes such as administrative parole, which allows for release without a hearing,7 but
for the most part, parole decisions generally involve discretionary decisions made on an individual basis.
In contrast, release in determinate jurisdictions tends to be more routinized. In these jurisdictions,
the law establishes the proportion of the pronounced sentence that must be served prior to release
to post-prison supervision. When a person reaches the minimum required term, the person is usually
released without any further review. For example, in Minnesota, a person must serve two-thirds of the
pronounced sentence in prison, and one-third of the pronounced sentence on supervised release, which
is Minnesota’s version of parole.8 In certain circumstances, the time served in prison may be extended
in these determinate sentencing systems. For example, in Wisconsin, if a person violates any prison
regulation or refuses or neglects to perform required or assigned duties, the Department of Corrections
may extend the term of confinement in prison by 10 days for the first offense, 20 days for a second
offense,9 or 40 days for a third or subsequent offense. Most prisons also require development of a release
plan, including an initial housing placement, so release can sometimes be delayed if a person is unable
to finalize this plan before the expected release date.10 But for the most part, within determinate systems,
the release date is certain.
No system is purely indeterminate or determinate. For example, states that are considered determinate may
have a group of offenses for which indeterminate life sentences are imposed, and release in those cases may

6.
7.
8.
9.
10.

See, e.g., Alaska Stat. § 33.16.110 (2021); Neb. Rev. Stat. § 83-1,115 (2022); Ark. Admin. Code 158.00.1-2 (2021).
See, e.g., S.D. Codified Laws §§ 24-15A-38 to -39 (2022).
Minn. Stat. § 244.01, subd. 8 and 244.05, subd. 1b (2021).
Wisc. Stat. § 302.113(3)(a) (2022).
See, e.g., Alaska Stat. § 30.30.011(a)(9) (2021) (requiring reentry plans that address housing, employment, treatment or
counseling, and education or job training services).

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

3

Examining Prison Releases in Response to COVID

Prison Release Mechanisms

closely resemble the parole process described above.11 And both types of systems tend to have good time or
earned time policies, which allow a person to earn time off from the pronounced prison sentence for good
behavior or for participating in or completing programming or other requirements.12 These good
time or earned time credits will usually advance the date of release, allowing for earlier release than would
otherwise be possible without the application of such credits.
Both indeterminate and determinate jurisdictions also frequently use work release. This type of release is
typically a form of transition from prison to the community, allowing the individual to leave incarceration for
employment purposes. Some forms of work release allow the person to leave the prison facility only during
work hours. Other forms of work release involve transitioning from prison to a residential correctional facility or
to home confinement and attending work from that location each day.13 In indeterminate jurisdictions, work
release may be a specific form of parole, requiring the same parole hearing process as any other form of parole.14
Additionally, both indeterminate and determinate jurisdictions frequently develop additional programs,
incentives, and mechanisms for early release to address prison overcrowding or to provide further
incentive for people in prison to engage in programming. For example, Colorado allows individuals who
are within six months of their release date and who meet certain requirements to qualify for release
into the Intensive Supervision Program.15 Minnesota has a program allowing early release of people who
have completed a chemical dependency treatment boot camp.16 And Kansas has a provision allowing
a person to be placed on home confinement, sometimes as an alternative placement to prison when
a person has violated the conditions of probation or parole, and sometimes as a form of early release.17

Non-Routine Prison Release Mechanisms
Though most prison releases occur according to the ordinary mechanisms described in the section above,
there are also occasions for more extraordinary forms of release. These mechanisms recognize that changes in
circumstances occurring sometime after sentencing may overtake the need for deterrence or incapacitation
and justify early or temporary release from prison. In this section, we identify and briefly describe some of the
most prevalent non-routine release mechanisms. However, this is not an exhaustive list.

Compassionate
Release:

11.
12.
13.
14.
15.
16.
17.

This type of release, which may also be called medical parole, is for people who
are suffering from a debilitating or terminal medical condition. For those with a
terminal condition, release may be granted within a year or two of the person’s
expected death. For those with a debilitating condition, release may be permitted
either because the person’s condition is so severe that they are no longer considered
a threat to public safety or because the prison is unable to provide appropriate
medical care. When release is granted for the purpose of obtaining medical care,
however, the release may be temporary, and the person may be required to return
to prison once that care has been provided and their condition has been stabilized.

See, e.g., Minn. Stat. § 244.05, subd. 4 (2021) (requiring a person sentenced to life imprisonment to serve at least 30 years before
being considered for release from prison).
See Reitz et al., fn 3.
See, e.g., Iowa Stat. § 906.4 (2022) (authorizing the Iowa Parole Board to order work release and requiring certain individuals
to begin their work release from a residential facility).
See, e.g., Iowa Stat. §§ 904.901-904.909 (2022) (establishing work release as a specific type of parole).
Colo. Stat. § 17-27.5-101 (2022); see also Colorado Department of Corrections, Intensive Supervision Program, https://cdoc.colorado.
gov/parole-and-re-entry-services/supervision/community-corrections/intensive-supervision-program-inmate#:~:text=A%20
residential%20offender%20who%20has,Parole%20Eligibility%20Date%20(PED).
Minn. Stat. § 244.0513 (2021).
Kan. Stat. § 21-6609 (2022).

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

4

Examining Prison Releases in Response to COVID 		

Prison Release Mechanisms

Geriatric
Parole:

This type of release is for people who have reached an advanced age in
prison—often mid-60’s or older. In many jurisdictions, the person must
also have served a significant portion of their sentence (i.e., 10 to 15 years),
though people who have committed certain crimes, such as sex crimes or
those resulting in a life sentence, are often not eligible for this type of release.
Some jurisdictions also require that the person demonstrate a record of good
behavior while in prison18 or that the person must be suffering from health
conditions associated with advanced age.19

Furloughs:

This type of release is temporary. A person may be granted permission to
leave the prison facility for a defined period of time and for a specific purpose.
For example, a person may be granted permission to attend a close relative’s
funeral or to receive medical treatment that is not available in prison.20

Clemency /
Commutation:

This type of release involves reducing a person’s sentence length.21 The person’s
conviction remains on the record, but the sentence is reduced because it is no
longer deemed appropriate or because the person has shown extraordinary
growth and change during their prison stay, meriting an early release.22 The
power to commute sentences is traditionally held by the governor, however,
some parole boards can also grant sentence commutations or make
commutation recommendations to the governor.

Prison
Overcrowding
Valves:

While all of the other release mechanisms highlighted in this section address
release mechanisms that operate on an individual basis, some states also
have prison overcrowding valves that allow them to reduce the prison
population on a more systemic level. For example, in Kansas, if prisons reach
90% or more of their overall capacity, the Kansas Sentencing Commission
must propose for consideration by the legislature and governor changes to
the sentencing guidelines that will result in reduced sentences and alleviate
overcrowding.23 Similarly, Wisconsin has special action parole, which allows
for larger-scale releases to relieve overcrowding when the prison population
equals or exceeds capacity.24

18.
19.
20.
21.
22.
23.
24.

See e.g., La. Stat. Ann. § 15:574.4(A)(4) (requiring that applicants for geriatric parole are age 60 or older, have served at least 10
years of their sentence, and have not had major disciplinary offenses for twelve consecutive months).
See e.g., S.D. Codified Law § 24-15A-55-68 (allowing an individual to qualify for one type of geriatric parole if they are at least
65 years old, have served 10 years of their sentence, and have a medical condition for which care will be at least double the
average annual medical cost).
See, e.g., 28 Vermont Stat. Ann. § 808 (2022) (authorizing furloughs to visit a critically ill relative, attend a funeral, or to obtain
medical treatment, employment, or housing).
Another type of clemency is a pardon, in which a person’s conviction is sealed or effectively erased. Pardons usually occur after
a person has served their full sentence and are less often a type of release directly from prison.
See, e.g., Colo. Stat. § 16-17-102 to -103 (2022) (establishing authority and process for commutation); Executive Clemency
Advisory Board Application Eligibility Criteria Commutation of Sentence, https://drive.google.com/file/d/1f1zkG8AFNvsu3t1UN4S0NVwpX0BiKSV/view.
Kan. Stat. § 21-6822 (2022).
Wisc. Stat. § 304.02 (2022).

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

5

Examining Prison Releases in Response to COVID

Prison Release Mechanisms

Assumptions About How COVID Releases Would Occur
Given how prison releases ordinarily occur, we embarked on this study with two main assumptions about
how prison releases would likely work in the context of the COVID-19 pandemic. Both assumptions were
grounded in the belief that states and the federal government would be largely bound by the release
mechanisms that were already in place, whether routine or non-routine.

Assumption
#1:

Jurisdictions with Indeterminate Sentencing Systems would be More Likely
to Make Releases Than Jurisdictions with Determinate Sentencing Systems.

Our assumption was that jurisdictions with indeterminate sentencing systems would be more likely to
make releases from prison due to COVID than jurisdictions with determinate sentencing systems. We had
this expectation because we thought that the laws and structures that governed ordinary releases from
prison in determinate systems were more tightly structured, and that there would be less opportunity
for jurisdictions to refine or enact procedures to enable the swift release of prisoners. In other words, we
assumed that the very structures that made prison release more definite in these jurisdictions would
serve as barriers to making releases that did not follow those structures. In contrast, because there is more
discretion to effect release in indeterminate sentencing systems, we thought indeterminate sentencing
jurisdictions would have more flexibility to use that discretion to facilitate releases from prison.

Assumption
#2:

Jurisdictions would Rely on Existing Compassionate Release and Geriatric
Parole Provisions to Effect Releases.

Our second assumption was that because the factors that increase the risk of severe illness from COVID
are primarily related to health conditions and age, most jurisdictions would leverage their already
existing compassionate release and geriatric parole mechanisms to reduce prison populations.25 These
statutory schemes already have built-in eligibility criteria and release procedures. Thus, as part of our
second assumption, we assumed that the eligibility and exclusion requirements for people released
from prison due to the pandemic would generally mirror the eligibility and exclusion requirements
for compassionate release and geriatric parole. For that reason, we completed a detailed survey of the
statutory frameworks for compassionate release and geriatric parole in the United States. The results of
our survey are included in Appendix D.

25. People with Certain Medical Conditions, Ctr. for Disease Control & Prevention, https://www.cdc.gov/coronavirus/2019ncov/need-extra-precautions/people-with-medical-conditions.html#:~:text=More%20than%2081%25%20of%20
COVID,medical%20conditions%20they%20have%20increases (last updated May 2, 2022).

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

6

Understanding COVID
Releases
In conducting this survey, our goal was to understand what factors allowed for the large-scale
release of people from prison as a response to the pandemic, not to understand the effect of
the pandemic, but to understand what lessons can be learned that might inform modernday responses to the issue of mass incarceration. In this section, we begin by describing the
number and scale of releases, and then dig into those releases by examining which legal
mechanisms were used to effect release and what characteristics of incarcerated individuals
were generally targeted in setting the parameters for release. We close this section by returning
to our assumptions for release and examining whether releases to prison generally aligned
with or ran in opposition to our expectations for release.

Prevalence of Release
In order to determine the size and scale of releases from prison due to the pandemic, we conducted
a survey of publicly available information,26 including data collected by UCLA’s COVID Behind Bars
Data project, news stories, court orders, executive orders, and information on agency websites. From
these sources, we documented whether each state made releases from prison, the number of people
released, whether release was prompted by external forces such as legal action, the legal mechanism(s)
utilized, and the eligibility criteria for release. As will be shown in the next sections, we further parsed and
analyzed the mechanisms and criteria for release. The information in this report represents the results of
our public survey conducted through December 2021; some releases may have occurred after that date.
Additionally, our survey focused only on releases from prison. Many jurisdictions also made releases from
city and county jails and local correctional facilities, which are not chronicled or examined in this report.
In some jurisdictions, publicly available information did not clearly differentiate between releases from
prisons and jails. In those situations, we utilized what information we had to develop our best estimate
of the size and scale of prison releases as distinct from any releases made from local jails.
We also worked to separate COVID-related releases—that is, non-routine releases—from releases that
occurred as a result of routine parole release or sentence expirations. This proved to be more difficult in
some states than others. Jurisdictions would often report all prison releases within a certain timeframe,
lumping together both routine and non-routine releases. In some cases, we had to determine typical
release numbers based on historical data and compare it against the release numbers reported during
the pandemic. For the purposes of this report, we are only focused on prison releases that can be directly
tied to unique efforts, court orders, governor directives, etc. that were undertaken specifically to reduce
prison populations due to the pandemic. We do not count releases that would have normally occurred
during a non-pandemic year.
26. For each state, we first reviewed any data about prisons releases collected by UCLA’s COVID Behind Bars Data project. We then
conducted a Google search including the state’s name, “prison,” “release,” and “COVID” (or “pandemic” or “corona” if the search
with “COVID” did not yield results) for any additional source materials or information that might shed light on the releases,
and to identify any releases that may not have been recorded by the UCLA project. If a state did not have an entry on the
UCLA project website, we conducted the same Google search, and if no releases were located, we conducted further searches
looking for articles stating or implying that the state had not released any prisoners. We also cross-checked our findings with
the Prison Policy Initiative’s list of pandemic-related criminal justice policy changes and the Crime and Justice Institute’s list
of how criminal justice systems were responding to COVID-19 to ensure we were not missing release information.

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

7

Examining Prison Releases in Response to COVID

Understanding COVID Releases

During the period covered by our survey, we estimate that a total of 80,658 people were released from
prisons in 35 jurisdictions (34 states and the federal prison system) due to COVID-related policies. The
total number of people released within each jurisdiction ranged from three in Florida to 39,588 in the
federal system (see Appendix A. Prison Releases and Institutional Features by Jurisdiction for a complete
list of releases by jurisdiction). As shown in Figure 1, 2000 or fewer people were released in 30 of the
35 jurisdictions that made releases due to the pandemic. Only five jurisdictions, including the federal
government, made what we might classify as large releases of 3000 or more people. Instead, the most
common number of releases in our survey was a 150 or fewer people (9 jurisdictions). Though we might
have expected the numbers of people released to correlate with each jurisdiction’s prison population
size, when we ranked the states in order of prison release size compared to 2019 prison population
size, there appeared to be no relationship (Appendix A. Prison Releases and Institutional Features by
Jurisdiction).

Figure 1. Frequency of Total Prison Releases by Jurisdiction
* California and Federal BOP, not shown, were outliers, releasing 11,104 and 39,588 people, respectively
9

9

8

6

6

5

4
3

3 3

2

3

6151-6300

6301-6450

6001-6150

5850-6000

5551-5700

1
5701-5850

5201-5350

5350-5500

5051-5200

4751-4900

4901-5050

4601-4750

4451-4600

4151-4300

4301-4450

4001-4150

3851-4000

3551-3700

1
3701-3850

3401-3550

3101-3250

3251-3400

2951-3100

2801-2950

2651-2800

2351-2500

2501-2650

2201-2350

1
2051-2200

1
1651-1900

1 1

1901-2050

1201-1350

751-900

901-1050

601-750

301-450

451-600

1-150

151-300

0

1

1351-1500

1

1501-1650

2

1051-1200

NUMBER OF JURISDICTIONS

7

NUMBER OF PEOPLE RELEASED

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

8

Examining Prison Releases in Response to COVID 		

Understanding COVID Releases

Though there was great variation in the absolute numbers of people released from prison in each
jurisdiction, we surmised that this variation might have been due to each jurisdiction’s varying prison
population size. In order to gain some sense of scale for the releases that were made, we used the
2019 prison population for each jurisdiction as a baseline estimate of the population at the start of
the pandemic. We then divided the total number released in each state by that baseline to arrive at
a rough proportion of the 2019 prison population that was released over the course of the pandemic.
This measure was imperfect because though most prison releases occurred in the spring of 2020, some
jurisdictions continued to make releases well into 2021. However, this measure, imperfect though it may
be, allowed us to level the playing field among states to determine which states released a greater or
fewer number of individuals relative to the size of their overall prison population.
The majority of releases were quite modest, amounting to the equivalent of less than 10% of the 2019
prison population in 27 of the 35 jurisdictions in which releases occurred (Figure 2). Seven states made
releases that amounted to less than 1% of their 2019 prison population, and three of those states—Florida,
Kansas, and Montana—released fewer than 10 people each. Only eight jurisdictions were bolder, releasing
the equivalent of more than 10% of the 2019 prison population. New Jersey released the largest group
proportionally, at 34%. And federal prisons released the equivalent of 23% of their 2019 prison population
(Appendix A).

Figure 2. COVID Releases as Proportions of 2019 Prison Populations
COVID RELEASES AS A PROPORTION OF 2019 PRISON POPULATIONS

16

No Release
7

<1%

11

1-3%
5

4-6%
4

7-9%
1

10-12%

3

13-15%
2

16-18%

2

>18%
0

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

NUMBER OF JURISDICTIONS

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

9

Examining Prison Releases in Response to COVID

Understanding COVID Releases

Legal Mechanisms Used to Accomplish Release
While all jurisdictions continued to release individuals through routine means during the pandemic,
34 states and the federal government, made a number of non-routine releases directly due to the
pandemic. These releases had two goals: to reduce the prison populations in order to facilitate more
social distancing inside prison facilities, and to release individuals who were at an increased risk of serious
complications from the COVID-19 virus. The legal mechanisms for these releases varied substantially. In
some jurisdictions, executive power was used to relax requirements for work release or medical furlough.
In other jurisdictions, the parole board simply made a concerted effort to process cases more quickly. In
this section, we provide a high-level summary and analysis of the release mechanisms that were used
to make non-routine releases during the pandemic. More detailed information about each jurisdiction’s
efforts can be found in the release events narrative (Appendix E. Narrative Descriptions of COVID Prison
Releases by Jurisdiction).
Appendix B identifies the legal mechanisms employed by states to effect releases during the pandemic.
In some jurisdictions, multiple mechanisms were used, so jurisdictions may appear in the table more than
once. Again, our purpose here is not to provide a detailed survey of each non-routine release but to instead
identify the administrative, statutory, and legal elements that allowed non-routine releases to occur. Most
importantly, we note whether the mechanisms used required a significant modification to be effective. This
allows us to determine what barriers and opportunities exist for future decarceration efforts.
Many of the prison releases chronicled during this study were logistically complicated, multi-step events
that required the cooperation of multiple agencies. For this reason, it was sometimes difficult to identify
which legal mechanism or agency was ultimately responsible for the release. For example, in many states,
Executive Orders broadened eligibility for existing legal mechanisms, but it was unclear if individuals
were ultimately released due to the actions of the department of corrections or through discretionary
parole release (or both). Similarly, it was sometimes difficult to determine how many individuals were
released through each mechanism, as reports often concentrated on release events, without providing a
breakdown of who was released through each mechanism. We do our best to summarize these events
accurately based on multiple sources.

Overview of Mechanisms Used
Across all jurisdictions, there were 10 main legal mechanisms used to effect release during the pandemic
(Figure 3). The most frequent mechanism used was parole, having been utilized in 11 jurisdictions.
Generally, this meant that the parole board—sometimes in unison with officials from other criminal
justice agencies—put resources toward releasing more people. For example, by increasing the pace of
parole reviews or holding special panels to identify eligible individuals for release.

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

10

Examining Prison Releases in Response to COVID 		

Understanding COVID Releases

Figure 3. Frequency of Legal Mechanisms Used to Effect Prison Releases During COVID
*Compassionate release includes medical parole and medical furlough.
12

11

INSTANCES OF USE

10

10

8
6

8
7

6

4

4

2

2

4

2

3

2

2
U
nc
le
ar

O
th
er

re
le
as
e
W
or
k

re
du
ct
io
n

Se
nt
en
ce

re
du
ct
io
n

va
lve

Sa
nc
tio
n

ov
er
cr
ow
di
ng

Pa
ro
le

Pr
iso
n

co
nfi
ne
m
en
t

tim
e

H
om
e

tim
e/
ea
rn
ed

Fu
rlo
ug
h

Go
od

Re
le
as
e*

Co
m
pa
ss
io
na
te

Co
m
m
ut
at
io
n

0

LEGAL MECHANISMS

Compassionate release was used in 10 instances, spanning nine jurisdictions,27 largely without
modification to existing programs. Where this mechanism was used, the release numbers were quite
modest compared to when states used other means. For example, the Oklahoma Department of
Corrections recommended a group of individuals for medical parole that were at an increased risk
of COVID-19 complications and who were not serving sentences for violent or sex crimes.28 The parole
board ultimately released 12 people from the list. The Louisiana Department of Corrections created a
special panel consisting of representatives from various criminal justice agencies to identify individuals
for medical furlough, prioritizing individuals serving time for nonviolent offenses and those with a
short time left on their sentence. It ultimately released 68 individuals.29 For an in-depth analysis of why
compassionate or geriatric release mechanisms were underutilized as a response to the pandemic, see
the section below revisiting our assumptions about Releases Due to Compassionate/Geriatric Provisions.

27. There were two distinct instances of compassionate release being used at the federal level.
28. 2001 OK. HB 2924, https://plus.lexis.com/document/?pdmfid=1530671&crid=963d660e-8557-4a0d-9685-d2a96306b197&pd
docfullpath=%2Fshared%2Fdocument%2Fstatutes-legislation%2Furn%3AcontentItem%3A4J1B-8GH0-0033-41Y8-0000000&pdcontentcomponentid=125155&pdworkfolderlocatorid=NOT_SAVED_IN_WORKFOLDER&prid=798894bf-b319-45878f64-108c9d0e480d&ecomp=-t4hk&earg=sr4; Chris Polansky, Board Recommends Special Medical Parole For 12 State
Inmates, Public Radio Tulsa (May 13, 2020), https://www.publicradiotulsa.org/local-regional/2020-05-13/board-recommendsspecial-medical-parole-for-12-state-inmates#stream/0.
29. Julie O’Donoghue, Louisiana Prisons Need To Do More COVID-19 Releases, Advocates Say, La. Illuminator (Sept. 3, 2020),
https://lailluminator.com/2020/09/03/louisiana-prisons-need-to-do-more-covid-19-releases-advocates-say/; Lea Skene, Review
panel to consider medical release for some Louisiana state prison inmates due to coronavirus, The Advoc. (Apr. 14, 2020),
https://www.theadvocate.com/baton_rouge/news/coronavirus/article_62e9f822-7e79-11ea-bfb3-933881495eb6.html.

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

11

Examining Prison Releases in Response to COVID

Understanding COVID Releases

Commutation—shortening the period of time to be served in prison—was utilized in seven jurisdictions.
In six jurisdictions, the governor commuted sentences while in Georgia, the parole board exercised its
commutation power. Traditionally, commutation is based on case-by-case review of individual cases, and
for a variety of reasons, such as to correct an unreasonably harsh punishment. The process varies from
state to state, but generally requires multiple steps such as preparing lengthy individual applications,
seeking input from criminal justice officials such as the prosecuting attorney, and multiple hearings
to determine suitability for release. It is a bottom-up approach where each commutation is driven by
each individual case. In the cases we chronicled in Appendix B. Release Mechanisms, this process was
modified significantly by having the system drive the process (i.e., a top-down approach). In these cases,
the governor or parole board commuted the sentences of large groups of people based on general
membership in a group (e.g., individuals who were at an increased risk of complications from the virus).
For example, Kentucky Governor Andy Beshear filed several Executive Orders in direct response to the
pandemic, which commuted the sentences of 1,184 people in jails and prisons.30 All of these individuals
had been identified as being medically vulnerable to COVID-19 and/or were nearing the end of their
sentence, as well as fulfilling other requirements, such as serving time for non-violent, non-sexual offenses.
A variation on the theme of commutation was sentence reduction, which was used in two jurisdictions.
Here, courts were the main actors, revisiting and reducing previously imposed sentences. In Rhode
Island, the Supreme Court took an active role, suspending the time limit for motions to resentence, and
authorizing lower courts to reduce sentences for people within 90 days of release.31
Eight jurisdictions used existing authority to move incarcerated persons from prison to home
confinement. Interestingly, nearly all of the home confinement releases—seven of eight instances—
occurred in determinate jurisdictions, suggesting this mechanism is one way determinate jurisdictions
can exercise back-end release discretion. Another seven used good time or earned time credit awards
to speed up the release of eligible people. In these instances, jurisdictions often had to relax existing
requirements, such as Colorado, which removed its cap on the number of credits that could be earned.
Most notably, in New Jersey, the legislature enacted a new law establishing public health emergency
credits, which moved up the release date of 5,181 individuals.32 Though similar to good time or earned
time credits in other states, New Jersey was the only state to create credits that would specifically be
triggered by a public health emergency.
A handful of jurisdictions utilized other mechanisms as well. Four jurisdictions exercised what we refer
to as “sanction reductions.” In these jurisdictions, departments of correction utilized the power they
already had to impose sanctions on people who violated parole to do the reverse; that is, to reduce the
prison sanction previously imposed for the violation, and therefore release the person back onto parole.
Two other jurisdictions—Arkansas and Ohio—used the power of prison overcrowding statutes that were
already on the books to declare an emergency and identify individuals for release.

30. Nearly 1,000 Kentucky prison sentences to be commuted, Beshear says, FOX19 (Apr. 2, 2020), https://www.fox19.
com/2020/04/02/watch-live-gov-beshear-provides-update-covid-kentucky/; Chris Williams, Lawmakers question DOC
officials on Governor Beshear’s COVID-19 commutations, WHAS11 (Dec. 3, 2020), https://www.whas11.com/article/news/
kentucky/kentucky-governor-prisoners-released-covid-19-lawmakers-face-off-corrections-officials/417-84c107ff-50e3-452db050-31a7d3b5db05.
31. In re Req. for Prison Census Control In Resp. to COVID-19 (Apr. 3, 2020), https://www.courts.ri.gov/PDF/In%20re%20Request%20
for%20Prison%20Census%20Control%20(Order).pdf; Mark Reynolds, R.I. Supreme Court OKs release of 52 inmates,
Providence J. (Apr. 4, 2020), https://perma.cc/3JUV-NFLY.
32. Suzette Parmley, Has 'COVID Time' Legislation Worked, and What Does It Mean for NJ's Criminal Justice Reform Future?,
Law.com: N.J. L. J. (May 11, 2021), https://www.law.com/njlawjournal/2021/05/11/has-covid-time-legislation-worked-and-whatdoes.

12

Examining Prison Releases in Response to COVID 		

Understanding COVID Releases

Modifications To Existing Mechanisms
Almost all jurisdictions relied on existing mechanisms of release to effect pandemic specific releases; however,
many required significant modifications to be useful. Below, we highlight three common themes: expansion
of eligibility criteria or suspension of specific rules or limits, expedited processes, and large-scale use.
Expansion of Eligibility Criteria or Suspension of Specific Rules or Limits
Across multiple different types of release mechanisms, jurisdictions needed to enact modifications in
order to facilitate their use under tight time constraints. In each case, the goal was to increase the pool
of individuals who would be eligible for release under the mechanism used. For example, both Maryland
and Illinois suspended caps or limits on the amount of good time or earned time credits that could be
accrued, thereby advancing release dates for a number of people in prison. And Minnesota broadened
its work release eligibility criteria by including individuals who were within 3 months of their end of
confinement date (versus 8 months prior to the modification), raising the risk assessment score cut off,
and removing exclusions for specific offenses (e.g., weapons offenses).33 A few jurisdictions were also able
to expand their definition of medical vulnerability for compassionate release. For example, Oklahoma’s
compassionate release statute allowed the corrections department’s medical director to request parole
for any medical reason, thereby opening the door to develop a protocol for evaluating people who were
more likely to be vulnerable to the effects of the virus.34
Expedited Processes
A second change to the functioning of existing mechanisms was to simply expedite their use. Often
states would create review committees, empanel parole boards more frequently, or task personnel
with identifying and reviewing individuals for release. In most cases, these activities were spurred by
an executive order or court litigation. While existing mechanisms were used for release – such as parole
– it does not appear the mechanisms would have been used to accomplish these releases without
the issuance of a governor’s order or court action. We counted nearly two dozen instances where an
executive order or court order spurred the corrections department or parole board to either move faster
to release people or to use a particular mechanism for release. For example, Maryland’s governor issued
an Executive Order instructing the parole board to accelerate consideration of parole for people 60 years
of age and older (as long as they met other offense and re-entry planning requirements).35 And in New
Jersey, the governor instructed the Department of Corrections to expedite the release to emergency
medical home confinement for older, medically vulnerable individuals who had a short time left on their
sentence, resulting in the release of 300 individuals.36 When executive orders started the chain of events,
this probably represented one executive action among many taken within the jurisdiction to address
the danger represented by the virus. In contrast, when court orders started the chain of events, this likely
represented the presence of outside pressure to force corrections and/or the administration to act.
33. Mgmt. Analysis and Dev., Research summary: Prison population management (Dec. 21, 2020), https://mn.gov/obfc/assets/
Appendix%20A%20Ombuds%20for%20Corrections%20COVID%20Report_tcm1157-470275.pdf.
34. 57 Ok. Stat. § 57-332.18 (2022).
35. Luke Broadwater, With coronavirus spreading, Maryland Gov. Hogan signs order for expedited release of hundreds of
prisoners, The Balt. Sun (Apr. 19, 2020), https://www.baltimoresun.com/coronavirus/bs-md-pol-hogan-prisoners-202004197mzvooaoxfbyngowb2xdeucrme-story.html?utm_source=The+Marshall+Project+Newsletter&utm_campaign=f1efaf562fEMAIL_CAMPAIGN_2020_04_20_11_35&utm_medium=email&utm_term=0_5e02cdad9d-f1efaf562f-166145513;
36. N.J. Exec. Order No. 124 (Apr. 10, 2020), http://d31hzlhk6di2h5.cloudfront.net/20200410/c0/64/ce/2c/0ef068b5d2c6459546c33a46/
EO-124.pdf; Joe Atmonavage, First wave of 50 inmates approved for release from N.J. prisons under Murphy’s order, NJ.com
(Apr. 27, 2020), https://www.nj.com/coronavirus/2020/04/first-wave-of-50-inmates-approved-for-release-from-nj-prisonsunder-murphys-order.html; Daniel Israel, Curbing the spread of COVID-19 in state prisons, Hudson Rep. (Oct. 20, 2020),
https://hudsonreporter.com/2020/10/20/curbing-the-spread-of-covid-19-in-state-prisons/.

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

13

Examining Prison Releases in Response to COVID

Understanding COVID Releases

Large-scale Use
A third theme was large-scale use of procedures that had previously only been used on a case-by-case
basis. The use of nearly all mechanisms fit this theme, but of particular note were the changes made to
processes for commutation, compassionate release, home confinement, and accrual of good time or
earned time credits. Commutation and compassionate release usually require a lengthy application and
hearing process, with multiple layers of review. But instead, jurisdictions often tasked the department
of corrections with identifying a cohort of individuals who fit specific criteria, and releases under these
mechanisms were made en masse. Eligibility for release to home confinement or upon accrual of
sufficient good time or earned time credits are determinations that jurisdictions routinely make as they
process people through the final months of their prison term. But in order to respond to COVID, these
reviews had to be completed on an expedited basis, and again, often involved review of a specifically
identified cohort rather than on a case-by-case basis.

Criteria for Release
As noted above, in order to effect early release from prison, it was necessary to utilize some legal
mechanism to permit release from prison prior to the time that would otherwise have been defined
by law. In almost all cases, these mechanisms already existed, though they may have been sped up or
modified through use of a triggering event such as an executive order or court order. In a few cases,
early release was a new process developed whole cloth for the purpose of reducing prison populations
in response to the pandemic. In each case, however, government officials typically took a targeted
approach, identifying specific criteria to use in identifying people whose could be released from prison.
In this section, we analyze those criteria in an effort to understand what they might tell us about the
appetite and/or tolerances for back-end releases.

Overview of Release Criteria
Though 34 states and the federal government made releases during the pandemic, several jurisdictions
established multiple pathways for release by either using more than one release mechanism, as detailed
in the prior section, or by opening multiple channels for release based on different criteria. For example,
when New Jersey instituted its expedited parole process, the state’s eligibility criteria included those
older than 60, those with high-risk medical conditions, those denied parole within the previous year, and
those with short amounts of time left on their sentences. Thus, New Jersey had multiple release groups,
each with different eligibility criteria.37 In this section then, the unit of analysis is the release group rather
than the jurisdiction. Looked at in this way, there were 73 distinct release groups across 35 jurisdictions.
In order to examine the criteria for releasing individuals from prison due to the pandemic, we combed
through the source materials that comprised our survey of the states to find greater detail about each
release group. The source materials included executive orders, court orders, news stories, agency websites,
and other documents and sources detailing responses to the pandemic. From these source materials,
we captured the text describing each release group, and then parsed that text to determine the unique
sets of criteria governing each potential release from prison within the jurisdictions.

37. Lauren del Valle & Leah Asmelash, New Jersey releases more than 2,200 eligible inmates under nation's first public health
crisis sentencing law, CNN (Nov. 4, 2020), https://www.cnn.com/2020/11/04/us/new-jersey-prisoners-covid-trnd/index.html.

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

14

Examining Prison Releases in Response to COVID 		

Understanding COVID Releases

It quickly became apparent that no two jurisdictions were alike in their release criteria. Some jurisdictions
established wholly unique criteria. For example, though many jurisdictions used age and health as criteria
for release, only North Carolina specifically targeted women over the age of 50 who had particular health
problems.38 And even where criteria were similar, there were small variations. For example, though both
Georgia and Arkansas identified people who had committed “non-violent” crimes for release, Georgia
did not further define that term while Arkansas specifically noted that non-violent included non-sexual
and non-domestic abuse offenses.39 Thus, it was necessary to develop a schema for comparison. We did
this by first parsing the text into all possible criteria based on the actual wording of our sources, and then
refining similar criteria into broader categories until we were satisfied that the final categories reflected
the full range of release criteria. As a final step, we grouped the final criteria into high-level categories
reflecting the overarching purposes of the release criteria within each grouping. Table 2 shows the
schema we developed and our full analysis of criteria for release is shown in Appendix C.

Table 2. Schema for Comparing Release Criteria
High-Level Category

Individual Criteria
Non-violent / Low-level offenses

Type of crime

Non-sexual offenses
No crimes against a person
Low risk to public safety

Risk to public safety

Low risk of recidivism
Conduct in prison
Technical violations of supervision

COVID health risks

Medically vulnerable
Age based
Accrual of sufficient good time credits

Time Served

Fulfilled a certain percentage/amount of sentence
Short time left on sentence
Already eligible for release

Reentry

Housing plan

Our schema includes five overarching categories capturing a total of 14 individual criteria: type of crime, risk
to public safety, time served, COVID health risks, and reentry. Four of the five categories included criteria that
are traditionally considered in parole release. The fifth category, related to COVID health risks, while unique
to this situation, was also not that unusual since most states have some form of compassionate release or
medical parole and are accustomed to considering health concerns in some manner.

38. Pamela Walker, Pandemic Prompts Department of Public Safety to Transition Some Offenders to Supervision in the
Community, N.C. Dep’t of Pub. Safety (Apr. 13, 2020), https://www.ncdps.gov/news/press-releases/2020/04/13/pandemicprompts-department-public-safety-transition-some-offenders.
39. Ninette Sosa, ADC: 1,243 inmates considered for early release due to COVID, KNWA Fox24 (Apr. 29, 2020), https://www.
nwahomepage.com/lifestyle/health/coronavirus/adc-1243-inmates-considered-for-early-release-due-to-covid/; Press Release,
State Bd. of Pardons and Paroles, Board Considering Releases to Address COVID-19 in Georgia Prisons (March 31, 2020), https://
pap.georgia.gov/press-releases/2020-03-31/board-considering-releases-address-covid-19-georgia-prisons.

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

15

Examining Prison Releases in Response to COVID

Type of crime

Understanding COVID Releases

Figure 4. Use of Type of Crime Criteria in

Criteria falling into the “type of crime” category set
Release Groups
parameters regarding which conviction offenses,
if any, would impact one’s eligibility for release.
Number of Release Groups Utilizing These Criteria
Crimes described as “nonviolent” or “low-level”
Type of Crime (total for overarching category)
generally made individuals eligible for release while
crimes described as sex offenses or person offenses
45
generally excluded individuals from eligibility
Non-violent / Low-level offenses
39
for release. “Nonviolent” and “low level offenses”
had varying definitions across jurisdictions. Some
Non-sexual offenses
jurisdictions targeted specific offenses, such as
19
Oregon and Oklahoma, which specifically targeted
No crimes against a person
those convicted of drug and property offenses.40
9
Others defined non-violent offenses by exclusion.
For example, Kentucky excluded anyone convicted
of any crime higher than a class C or D felony, which
are generally sexual or violent offenses,41 and at least eight states excluded people convicted of domestic
violence offenses.42 The criterion for nonviolent / low-level offenses was the most prevalent in this category,
appearing in 39 of 73 release groups. Nineteen release groups required that the person not have been
convicted of a sex offense, and nine release groups required that the person not have been convicted of a
person offense (Figure 4; Appendix C).
Risk to public safety
“Risk to public safety” included criteria that were
aimed at identifying people who were deemed
safe to release into the community. The criteria
included low risk to public safety, low risk of
recidivism, conduct in prison, and technical
violations of supervision. “Low risk to public
safety” was either broadly defined, representing a
potentially qualitative judgement of the person’s
risk, or defined by the person’s placement in
prison. For example, Maine explicitly required that
the person being considered for release must be
in a “minimum security” placement.43 In contrast,
“low risk of recidivism” was usually expressly to

Figure 5. Use of Risk to Public Safety
Criteria in Release Groups
Number of Release Groups Utilizing These Criteria
Risk to public safety (total for overarching category)

25

Low-risk to public safety

15
Low-risk of recidivism

5
Conduct in prison

5
Technical violations of supervision

3

40. Shane Kavanaugh, Oregon Gov. Kate Brown approves
early release of 57 inmates vulnerable to coronavirus, The
Oregonian (June 25, 2020), https://www.oregonlive.com/
coronavirus/2020/06/gov-kate-brown-approves-early-release-of-57-inmates-vulnerable-to-coronavirus.html; Hicham Raache, Gov.
Stitt approves hundreds of prison commutations to mitigate coronavirus spread, KFOR (Apr. 10, 2020), https://kfor.com/news/
coronavirus/gov-stitt-approves-hundreds-of-prison-commutations-to-mitigate-coronavirus-spread/?eType=EmailBlastContent&eId
=07a90610-fdf3-4e1b-994a-712a02c5065c.
41. Ky. Exec. Order. No. 2020-267 (Apr. 2, 2020), https://governor.ky.gov/attachments/20200402_Executive-Order_2020-267_
Conditional-Commutation-of-Sentence.pdf.
42. The jurisdictions include California, Colorado, Iowa, Maine, Maryland, New Mexico, Ohio, and Rhode Island.
43. Megan Gray, Maine prisons pressured to release more inmates, and information, during pandemic, Portland Press Herald
(May 3, 2020), https://www.pressherald.com/2020/05/03/maine-prisons-pressured-to-release-more-inmates-and-moreinformation-during-pandemic/?rel=related

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

16

Examining Prison Releases in Response to COVID 		

Understanding COVID Releases

be determined by use of a risk assessment tool. “Conduct in prison” also tended to be broadly stated,
either referring to people who had good conduct for a specified period of time or who did not have any
disciplinary violations. Oregon, for example, required that an individual have a record of good conduct for the
12 months preceding their release.44 Finally, “technical violations of supervision” appeared to be a proxy for
risk, referring to the reason the person was incarcerated at the time of consideration for release rather than
the offense for which the person was serving time in prison. Overall, none of these criteria were very prevalent,
with only 25 of 73 release groups featuring these criteria (Figure 5; Appendix C).
COVID health risks
Within the category of “COVID health risks” we
included criteria that mirrored the risk factors
Figure 6. Use of COVID Risk Criteria in
that made individuals particularly vulnerable to
Release Groups
severe disease if they were to contract COVID:
Number of Release Groups Utilizing These Criteria
age and medical vulnerability. The minimum
age for release eligibility ranged from 30 to 65
COVID Risk (total for overarching category)
years old, though 60 and 65 were the ages most
38
frequently utilized by jurisdictions. Only Ohio
Medically vulnerable
considered age in the abstract, without listing
29
a number of years.45 Medical vulnerability was
Age-based
often broadly defined, likely due to the evolving
16
nature of the CDC guidelines. A few jurisdictions
defined medical vulnerability to include chronic
and serious conditions such as diabetes, cancer, or
asthma. And a few jurisdictions defined medical
vulnerability to include pregnancy.46 COVID health risks was the third most prevalent category, with one
or more of these criteria appearing in 38 of 73 release groups (Figure 6; Appendix C).
Time served
The “time served” category includes several criteria aimed at setting parameters around how much of
their sentence a person must have served to be considered for release. The criteria are that a person
served a certain percentage or number of years of the sentence, had a short time remaining on the
sentence, was already eligible for release, or accrued sufficient good time credits to be released. Most
jurisdictions that used the “percentage / amount served” criterion required the person to have served at
least half of their sentence, though the percentages ranged from 25-50%. Wisconsin sought to address
people with lengthy sentences by establishing a release group for people sentenced before 1999.47 The
most represented criterion in this category was “short time left on sentence,” which appeared in 37 of 73
44. Conrad Wilson, Oregon prisons to release more inmates as COVID-19 outbreaks continue, East Oregonian (Dec. 15, 2020),
https://www.eastoregonian.com/coronavirus/oregon-prisons-to-release-more-inmates-as-covid-19-outbreaks-continue/
article_7f423606-3f12-11eb-b1ee-572e37fe342a.html.
45. 2 high-profile prisoners have sentences commuted by DeWine amid coronavirus crisis, WTOL 11 (Apr. 17, 2020), https://www.
wtol.com/article/news/health/coronavirus/ohio-prisoners-sentences-commuted/512-3786f7da-1d88-4d6f-b57c-6e18ce89e6c6;
46. For example, New York and Illinois each had a small release group focused on new mothers and those who were pregnant.
Justin Bey, 8 pregnant women to be released from New York prison over virus fears, CBS News (May 6, 2020), https://
www.cbsnews.com/news/coronavirus-new-york-prisons-pregnant-women-freed-covid-19; Annie Sweeney, Facing growing
coronavirus threat, Illinois prison officials release moms jailed with their babies: ‘Oh my goodness, there was no words’, Chi.
Trib. (Mar. 27, 2020), https://www.chicagotribune.com/coronavirus/ct-coronavirus-woman-babies-released-prison-20200327t6rfew4m6jbuxmw4lrw5v47dfi-story.html.
47. Emily Hamer, Wisconsin DOC has released nearly 1,600 inmates so far to combat COVID-19 spread, Wis. State J. (May
8, 2020), https://madison.com/wsj/news/local/crime-and-courts/wisconsin-doc-has-released-nearly-1-600-inmates-so-far-tocombat-covid-19-spread/article_03537daa-e1ec-5fe8-ac68-f5cf38ce8be5.html (https://perma.cc/58WK-99UW

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

17

Examining Prison Releases in Response to COVID

Understanding COVID Releases

release groups. Jurisdictions used this to describe
the amount of time left on a person’s sentence
Figure 7. Use of Time Served Criteria in
before eligibility for release under established
Release Groups
rules. The amounts varied from just 30 days in
Number of Release Groups Utilizing These Criteria
New Mexico to five years in Kentucky, which was
48
an outlier. Most jurisdictions required the person
Time Served (total for overarching category)
to be within one year of release, though a fair
45
number required as few as 90 days remaining on
Accrual of sufficient good time credits
an individual’s sentence. A third criterion in this
6
category was that a person was “already eligible
Fulfilled a certain %/amount of sentence
for release.” This criterion referred to situations
6
where an incarcerated individual had already met
Already eligible
the conditions of release, such as those who had
6
served their minimum sentence but who had
not yet gone before a parole board or individuals
Short time left on sentence
37
who had recently been denied parole, as in New
49
Jersey. Finally, “accrual of sufficient good time
credits” also served to identify people who had
served a sufficient proportion of their sentence.
Colorado removed the caps on good time credits to increase eligibility for release50 while New Jersey
created an entirely new category of public health emergency credits in order to expand eligibility for
release.51 Time served was the second most prevalent category. Criteria within this group appeared in 45
of 73 release groups (Figure 7; Appendix C).

-f------------

Reentry
The least prevalent category, having been required for consideration in just 18 release groups, was
reentry, which required that the person have a housing plan in place in order to be eligible for release.
This number may be deceptive, however, as housing plans may have been an implicit consideration in
all releases. Jurisdictions that explicitly included housing described it in different ways. Most required
“stable” housing, while others required “safe housing,” “a viable housing plan,” “a plan for housing and
well-being,” or “access to housing and healthcare.” The inclusion of this criterion suggests that, just as
with standard prison release, institutions may have been hesitant to release individuals with nowhere to
go, especially in a climate when many states were experiencing lock downs, and local jurisdictions were
already struggling to implement social distancing standards community-wide, including by temporarily
housing people who were homeless in hotels rather than shelters.52

48. Elise Kaplan, Gov. orders early release of some inmates, Albuquerque J. (Apr. 6, 2020), https://www.abqjournal.com/1440938/
gov-orders-early-release-of-some-inmates-from-prison.html; Kentucky's Response to COVID-19, Kentucky.gov (Oct. 20, 2020),
https://governor.ky.gov/Documents/20201020_COVID-19_page-archive.pdf.
49. Joe Atmonavage, First wave of 50 inmates approved for release from N.J. prisons under Murphy’s order, NJ.com (Apr.
27, 2020), https://www.nj.com/coronavirus/2020/04/first-wave-of-50-inmates-approved-for-release-from-nj-prisons-undermurphys-order.html.
50. Colo. Exec. Order No. D 2020 016 (Mar. 25, 2020), https://www.colorado.gov/governor/sites/default/files/inline-files/D%20
2020%20016%20Suspending%20Certain%20Regulatory%20Statutes%20Concerning%20Criminal%20Justice_0.pdf
51. Lauren del Valle & Leah Asmelash, New Jersey releases more than 2,200 eligible inmates under nation's first public health
crisis sentencing law, CNN (Nov. 4, 2020), https://www.cnn.com/2020/11/04/us/new-jersey-prisoners-covid-trnd/index.html.
52. See, e.g., Nina Moini, For Ramsey County Homeless Hotels Officer Safe Haven and Hope Amid the Pandemic, MPR News
(Feb. 18, 2021), https://www.mprnews.org/story/2021/02/18/for-ramsey-county-homeless-hotel-rooms-offer-safe-haven-andhope-amid-the-pandemic.

A .

ROBINA INSTITUTE

OF CRIMINAL LAW ANO CRIMINAL JUSTICE

18

Examining Prison Releases in Response to COVID 		

Understanding COVID Releases

Combinations of Release Criteria
Having parsed out the individual criteria used by jurisdictions in making release decisions, we then
analyzed the relationship between criteria to see if certain combinations were more prevalent than
others in establishing who might be eligible for release. Figure 8 shows how jurisdictions combined
criteria from each category when defining eligibility for release. The information is organized by release
group, which we define as each group within the jurisdiction that had distinct eligibility requirements
for release from prison (see also Appendix C. Combinations of Release Criteria for combinations by
jurisdiction).

Figure 8. Frequency of Release Criteria Combinations
8

8

8

T = Type of Crime
R = Risk to Public Safety

NUMBER OF RELEASE GROUPS

7

C = COVID Health Risks
S = Short Time Left to Serve on Sentence

6

6

5

H = Housing Plan

5

4

4

3

2

4

4

3
2

4

4

3

2

2

2

2

2

1

1

1

2

1

1

1

1
+H

+H

+S

+R
T

T

+C

+C

+S

+H

+H
T

+R

+S

+S

+C
+R

T

+H
T

+R

+C

+H

+S
C

+H

+C
R

+S

+C
T

+H

+C
T

+S

+S

+R
T

T

+C

+H

+R
T

S

+H
C

+S
C

+H
R

+S
R

+C
R

S

+S

R

T

C

T

T

+C

0

RELEASE CRITERIA CATEGORIES

About three-fifths of the release groups utilized eligibility criteria in just one or two categories. Type of
crime, COVID health risks, and time served were the most common categories for release criteria, each
appearing as a component in more than half of all release groups. But while COVID health risks and time
served were frequently used on their own to determine eligibility for release, type of crime almost never
was. Instead, it was often paired with criteria in these two other categories.
Only 17 of the 73 release groups utilized criteria in just one category, most frequently COVID health risks
and time served. For release groups in which eligibility focused on COVID health risks alone, this indicates
that jurisdictions were focused on preventing people from contracting serious disease, possibly out of
concern for the individuals’ health, but also likely out of concern for their ability to care for people if they

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

19

Examining Prison Releases in Response to COVID

Understanding COVID Releases

became seriously ill. In contrast, release groups in which eligibility focused solely on time served may
have been prompted by those same health concerns tempered by risk aversion. In many cases, persons
falling into the time-served category were already eligible for release under some existing mechanism—
i.e., people who had earned sufficient good-time credits, who had served their minimum sentence but
had not yet had a parole hearing—so there was little risk in expediting their release.
Twenty-five of the 73 release groups utilized criteria in two categories, most frequently adding a
consideration for the type of crime for which the person was in prison to considerations for COVID health
risks or time served. Because type of crime generally singled out people who had been convicted of
nonviolent, non-sexual, non-person offenses, the addition of this criteria indicates a further concern for
risk. Though people convicted of low-level crimes such as drug and property offenses are often more
likely to recidivate,53 they are also typically more likely to be perceived as less of a danger to public
safety.54 Thus, in terms of public perception, these were the individuals whom officials could be more
likely to gain public support for releasing early.
Sixteen release groups required consideration of eligibility criteria in three categories, while 14 release
groups required eligibility criteria in four categories. Only Kentucky had release groups requiring
eligibility criteria from all five categories. Interestingly, type of crime appeared in all but two of these
multi-category release groups, suggesting that these jurisdictions were less willing to consider releasing
people convicted of crimes that might be labeled as “violent.” Many of the release groups requiring
criteria in four or five categories also included risk to public safety as an explicit consideration. For these
release groups then, type of crime was not serving as a sole proxy for dangerousness. About half required
use of a risk assessment tool to determine that the person had a low risk of recidivism, while about half
relied on the less precise judgement that a person was “low risk to public safety.” COVID health risks were
still a prominent consideration in many of these release groups but layered within a set of criteria that
required consideration of risk on multiple dimensions.
A more detailed table showing combinations of release by criteria by jurisdiction is shown in Appendix
C. Because most jurisdictions developed multiple sets of criteria for release, they appear in the table
multiple times for distinct release groups. This is significant because it shows that these jurisdictions
were working multiple angles to facilitate the release of individuals from prison. For example, Illinois
appears in the table with four different release groups, some of which focused on releasing women who
were pregnant, people 55 years or older with less than a year left on their sentence, and others who were
within 9 months of release.55 Thus, Illinois was considering multiple types of criteria, including COVID
health risks and time served. Over half of the jurisdictions (19) took this multi-pronged approach while 16
jurisdictions that made releases had just one release group.56

53. Nancy La Vigne & Ernesto Lopez, Recidivism Rates: What You Need to Know, Council on Criminal Justice (Sept. 1, 2021),
https://counciloncj.org/recidivism_report/.
54. Joel M. Caplan, What Factors Affect Parole: A Review of Empirical Research, 71 Fed. Prob. (2007).
55. Ill. Exec. Order No. 2020-11 (Mar. 23, 2020), https://www2.illinois.gov/sites/coronavirus/Resources/Pages/ExecutiveOrder2020-11.
aspx; Annie Sweeney, Facing growing coronavirus threat, Illinois prison officials release moms jailed with their babies: ‘Oh
my goodness, there was no words’, Chi. Trib. (Mar. 27, 2020), https://www.chicagotribune.com/coronavirus/ct-coronaviruswoman-babies-released-prison-20200327-t6rfew4m6jbuxmw4lrw5v47dfi-story.html; Ill. Exec. Order No. 2020-21 (Apr. 6,
2020), https://www2.illinois.gov/Pages/Executive-Orders/ExecutiveOrder2020-21.aspx; Money v. Pritzker, 453 F. Supp. 3d 1103
(N.D. Ill. 2020).
56. The states with one release group were Florida, Georgia, Hawaii, Kansas, Louisiana, Maine, Michigan, New Mexico, Rhode
Island, Utah, Vermont, Virginia, Washington, and West Virginia.

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

20

Revisiting Our Assumptions
About Release
In this section, we revisit the assumptions we made about how releases from prison would
work during the pandemic. As discussed below, for the most part, releases ran contrary to our
assumptions, suggesting that other factors were at play as jurisdictions worked to reduce their
prison population size. These factors will be addressed more fully in the next section, examining
lessons learned from the experience of the pandemic.

Releases in Determine vs. Indeterminate Jurisdictions
Our first assumption was that indeterminate systems,
Table 3. Release Decisions by System Type
which have more built-in discretion and therefore
more flexibility in their release processes, would be
No Release
Release
Total
more likely to release people from prison due to
#
%
#
%
the pandemic than determinate systems. However,
Indeterminate 14
41%
20
59%
34
we found this assumption to be at least partially
incorrect. All but two jurisdictions with determinate
Determinate
2
12%
15
88%
17
sentencing (88%) made releases from prison in
response to the pandemic. In contrast, nearly 60% of indeterminate jurisdictions made releases and just over
40% made no releases (Table 3). For more detailed information about which jurisdictions made releases by
system type, see Appendix A. Prison Releases and Institutional Features by Jurisdiction.

Number of Jurisdictions

But though a greater share of determinate rather than
indeterminate jurisdictions made releases, the scale
Figure 9. COVID Releases by Determinate/
of releases was larger in indeterminate sentencing
Indeterminate Jurisdictions
systems. For both determinate and indeterminate
■ Indeterminate ■ Determinate
jurisdictions, the majority of the releases were modest,
15
amounting to the equivalent of 1-9% of the 2019
14
14
12
13
prison population. Only two determinate jurisdictions
made releases that were greater than 10% of the 2019
9
prison population, compared to six indeterminate
6
jurisdictions. In this sense then, the difference
6
3
between structures may have been the reason why
2
2
indeterminate jurisdictions were able to release
0
No Releases
1-9%
10% or greater
more people. The more modest release proportions
COVID Releases as a Propotion of 2019 Prison Population
in determinate jurisdictions may have been reflective
of the fact that determinate jurisdictions have a more
definite structure for prison release, so they may have
experienced two limitations: 1) they may have been limited by existing criteria for release when using existing
release mechanisms; and 2) it may have been more difficult to develop procedures outside of those laws to
effect release. At the same time, the greater discretion available in indeterminate jurisdictions may have more
easily allowed for larger releases. Thus, while the system type may not have impacted whether the jurisdiction
made releases, it may have affected how many releases were possible.

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

21

Examining Prison Releases in Response to COVID

Revisiting Our Assumptions About Release

Releases Due to Compassionate/Geriatric Provisions
Our second assumption was that existing compassionate release and geriatric parole mechanisms would be
the primary means of releasing people from prison during the pandemic. We made this assumption because
the persons most at risk for serious illness from COVID were people who were older or who had certain
preexisting health conditions.57 Thus, it seemed natural that jurisdictions would lean on these procedures,
using existing eligibility and exclusion criteria, when making releases. However, that was largely not the
case. Compassionate release and geriatric parole were use in just 10 instances spanning 9 jurisdictions. As
explained in this section, this was largely due to an incompatibility in eligibility and exclusion requirements
as well as procedural barriers.

Incompatibility in Eligibility and Exclusion Requirements
The statutes establishing compassionate release and geriatric parole provisions include both eligibility
requirements—i.e., factors that must be present to qualify for consideration—and exclusion requirements—
categorical factors that disqualify a person from eligibility to even apply for consideration. But as will be
shown in this section, these factors were both simultaneously too narrow and too broad. Eligibility factors for
compassionate release and some forms of geriatric release were too narrow to encompass the health risks for
COVID, and even if they had not been, the exclusion factors would have resulted in a potential pool of people
in prison that was too broad for serious consideration by state and federal corrections leaders and officials.
• Eligibility factors for compassionate release and the forms of geriatric release based on medical
condition were too narrow to encompass the health risks for COVID.
Compassionate release programs generally require applicants to suffer from severe, terminal, debilitating,
and/or chronic incapacitation, disease, or illness. Some require that the condition be so severe as to render the
person with only a short time left to live. These statutes are reactive, intending to be used only after a severe
medical diagnosis has been established. In contrast, COVID releases needed to be forward looking, intending
instead to prevent severe illness.58 Thus, instead of requiring severe incapacitation, jurisdictions that included
medical vulnerability in their criteria for release focused on individuals having underlying health conditions or
high risk of complications from COVID-19. Pennsylvania for example, considered risk to COVID based on age,
autoimmune disorders, pregnancy, or other serious chronic medical conditions like heart disease, diabetes,
cancer, and other medical conditions that placed people at higher risk for coronavirus.59 Existing provisions for
compassionate release or geriatric parole based on age and medical condition simply did not fit the situation.
Where such statutes were used to effect release, they had to be altered in some way. For example, in Colorado,
the governor issued an executive order that, among other actions, suspended the criteria for compassionate
release and gave the department of corrections discretion to determine appropriate interim criteria.60

57. People with Certain Medical Conditions, Ctr. for Disease Control & Prevention, https://www.cdc.gov/coronavirus/2019ncov/need-extra-precautions/people-with-medical-conditions.html#:~:text=More%20than%2081%25%20of%20
COVID,medical%20conditions%20they%20have%20increases (last updated May 2, 2022).
58. It should be noted that some jurisdictions actually disallowed otherwise approved and/or eligible individuals from leaving
prison if they tested positive for COVID-19 prior to release in order to protect the general public. (See e.g., Ninette Sosa, Approx.
800 inmates approved for early release; COVID-19 precaution, KNWA Fox24 (May 14, 2020), https://www.nwahomepage.com/
lifestyle/health/coronavirus/approx-800-inmates-approved-for-early-release-covid-19-precaution/; Ky. Exec. Order. No. 2020267 (Apr. 2, 2020), https://governor.ky.gov/attachments/20200402_Executive-Order_2020-267_Conditional-Commutation-ofSentence.pdf).
59. Tom Wolf, Order of the Governor of the Commonwealth of Pennsylvania Regarding Individuals Incarcerated in State
Correctional Institutions, Commonwealth of Penn. Off. of the Governor (Apr. 10, 2020), https://famm.org/wp-content/
uploads/2020.4.10-TWW-SCI-reprieve-release-order-COVID-19.pdf.
60. Colo. Exec. Order No. D 2020 016 (Mar. 25, 2020), https://www.colorado.gov/governor/sites/default/files/inline-files/D%20
2020%20016%20Suspending%20Certain%20Regulatory%20Statutes%20Concerning%20Criminal%20Justice_0.pdf.

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

22

Examining Prison Releases in Response to COVID 		

Revisiting Our Assumptions About Release

Even if the statutory provisions focusing on medical conditions had been broad enough to include the
types of conditions that made people in prison vulnerable to severe disease from COVID, there was one
other aspect to these laws that would have been problematic. Most jurisdictions maintain the ability to
revoke compassionate release and geriatric parole if the individual’s health status no longer makes them
eligible for this extraordinary form of release. Ohio, for example, states that compassionate parole may be
revoked “if an individual's health improves to the point that they are no longer medically incapacitated,
in imminent danger of death, or terminally ill.”61 Another state, Kansas, takes this even further stating
that compassionate parole can be revoked if, among other things, an individual’s illness or condition
significantly improves or if the person does not die within 30 days of release.62 Such revocation provisions
would have defeated the purpose of releases during COVID, which were designed to reduce the prison
populations in order to facilitate some form of social distancing.
Given that the medical conditions in these statutes did not comport with the conditions surrounding
COVID releases, one might have thought that states would turn to the other types of geriatric provisions
based on age and time served. While 24 jurisdictions had geriatric parole provisions on the books, they
did not appear to utilize them or even draw inspiration from them in setting release criteria. Many
jurisdictions used age and time served as criteria for release. But while many geriatric parole provisions
focus on time served, most often requiring that the person had served at least 10 years of their sentence,
the time-served component of prison releases due to COVID focused on time left on the sentence, usually
requiring that the person had less than a year left on their sentence. Geriatric time-served provisions
were about accountability whereas COVID release time-served provisions were about minimizing risk.
And to be fair, most geriatric parole statutes that had a time-served requirement were quite lengthy,
suggesting the statutes were designed to address individuals serving lengthy sentences for more serious
and violent offenses, which, as shown in the next section, were the very groups jurisdictions were looking
to avoid releasing during COVID.
• The exclusion groups for compassionate release and geriatric parole would have resulted in a
potential pool of people in prison that was too broad for serious consideration.
Both compassionate release and geriatric parole statutes have clear exclusion groups, that is, people
who are not eligible to apply for such release because of their sentence or crime of conviction. Similar to
the schema we developed for examining release criteria during the pandemic, exclusion groups in this
context could be thought of as layering a consideration for the type of crime on top of the other eligibility
criteria for compassionate release and geriatric parole.

61. Ohio Rev. Code §§ 2920 (A)(5), (N)-(S); Ohio Rev. Code § 2967.05 (2021).
62. Kan. Stat. §§ 22-3728-3729 (2021).

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

23

Examining Prison Releases in Response to COVID

Revisiting Our Assumptions About Release

Figure 10. Visualization of Crime Type Eligibility and Exclusions

NONVIOLENT
OFFENSES

VIOLENT
OFFENSES
Sex Offenses
Person Offenses

Most nonviolent offenses except some
nonviolent person and nonviolent sex
offenses were typically considered for
COVID releases.

The most serious violent offenses may
receive a sentence of death, life
without parole, or life with the
possibility of parole. These are typically
the only offenses excluded from
compassionate and some forms of
geriatric release. Other forms of
geriatric release also exclude some
additional violent crimes.

All violent offenses plus some
nonviolent person and nonviolent
sex offenses were typically
excluded for COVID releases.

Compassionate release exclusions are typically narrow, only preventing people who have been sentenced
to death, life without parole, or life from applying for release due to having a terminal illness or serious
medical condition. The same types of exclusions are often present for geriatric parole when eligibility
is based on age plus a serious medical condition (Appendix D). These severe sentences are most often
imposed for murder or repeat violent offenses, or for people determined to be predatory or habitual
offenders. Thus, for these types of statutes, the only people who are excluded from applying are people
who would not be eligible for parole anyway. That is, people for whom the expectation at the time of
sentencing was that they would serve the remainder of their life in prison.
In contrast, exclusions for geriatric parole when paired with time served are often broader, encompassing
people who have been convicted of sex offenses or crimes of violence in addition to those with sentences
of death, life without parole, or life. Thus, in these statutes, other considerations such as accountability for
the offense, likely play a larger role in determining who can and cannot apply for release. In other words,
for a person convicted of kidnapping and sexual assault, lawmakers in the jurisdiction may have thought
it more appropriate to ensure that the person serve a significant portion of the pronounced sentence,
even if that means the person would be quite elderly upon release.
In comparison to these statutory exclusion groups for compassionate release and geriatric parole, the
exclusion groups for releases from prison due to COVID were even broader (Figure 5). The offenses for
which a person could be considered for release due to COVID were primarily those considered nonviolent,
and in many jurisdictions, also had to be considered non-sexual and non-person, thereby excluding all
violent offenses as well as some person and sex offenses that might be labeled as nonviolent.63 Thus,
one of the reasons states may not have relied on their compassionate release and geriatric parole
provisions may have been because those statutes would have permitted consideration of too many
people who had been convicted of crimes labeled as “violent.” Had jurisdictions not focused solely on
nonviolent offenses, they may have thought they would be overwhelmed with applications from people
in prison seeking release through these mechanisms. Or jurisdictions may have been risk averse, not
wanting to be placed in the position of releasing people they or the public might perceive as being more
dangerous. In either case, notwithstanding the eligibility criteria discussed above, the exclusion groups
24

Examining Prison Releases in Response to COVID 		

Revisiting Our Assumptions About Release

in compassionate release and geriatric parole provisions would not have been sufficiently narrow to
address decision makers’ concerns—and likely, what they perceived the concerns of the public to be64—
for accountability and public safety.

Procedural Barriers
Additional reasons compassionate release and geriatric parole were not widely used during the pandemic
may have been procedural. Compassionate release and geriatric parole provisions did not lend themselves
to the scale of response needed to significantly reduce prison populations. Under these statutes, release
is considered on a case-by-case basis after significant scrutiny and individualized assessment. In most
jurisdictions there are at least three different approvals and/or checkpoints that individual applicants
must survive – any of which have the discretion to take an individual out of consideration.65
Statutory provisions also vary as to the timing of review. In Oklahoma, for example, after approval of
one’s initial application materials, a final decision must be rendered within four business days,66 while in
South Dakota, the statute ensures only that a hearing will take place within three months.67 The threemonth timeline would have been excruciating for individuals in prison who had no control over their
circumstances, especially when the virus was spreading quickly, and so little was known about treatment
and prevention. But even the seemingly quick turnaround in Oklahoma may have been challenging
because the clock starts after initial approval and thus fails to account for slowdowns in starting the
application process and gathering the requisite application materials.
Thus, the procedural provisions inherent in compassionate release and geriatric parole that make
review highly layered, individualized, discretionary, and slow create a tool that is useful in some, but
not all (or even most) situations, and which is difficult to deploy for making largescale rather than
case-by-case releases.

63. Examples of nonviolent person offenses might include some forms of theft or burglary and violation of an order for protection.
Examples of nonviolent sex offenses might include possession of child pornography or failure to register as a sex offender.
64. See, e.g., Research Summary: Prison Population Management, Responses to Covid-19 in Minnesota Prisons at 52 (Dec. 21, 2020),
https://mn.gov/obfc/assets/Appendix%20A%20Ombuds%20for%20Corrections%20COVID%20Report_tcm1157-470275.pdf
(documenting concerns by corrections officials that there was a public perception that the Department of Corrections was
being reckless with early release).
65. See e.g., N.C. Gen. Stat. § 148-4 (which includes review by five different individuals within the DOC system); Vt. Stat. Ann. tit. 28,
§ 502a (d) and Vt. Stat. Ann. tit. 28, § 808 (which requires review by three different departments within the DOC system before
an application is submitted to the Board of Parole).
66. See, e.g., Ok. Stat. § 332.16 (2021).
67. See, e.g., S.D. Codified Laws §§ 24-15A-56 (2022).

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

25

Revisiting Our Assumptions About Release

Examining Prison Releases in Response to COVID

COVID as a Political Issue
One assumption we did not make was that politics
Table 4. Prison Releases by Gubernatorial
might affect releases. This was due in part to the
Political Party
fact that at the start of the pandemic in 2020,
political leadership across the United States was
No Release
Release
Total
nearly evenly split, with 24 states being headed by
Democratic
3
21
24
Democrats, and 26 states and the U.S. Presidency
Governor
being headed by Republicans (Appendix A. Prison
Republican
13
14
27
Releases and Institutional Features by Jurisdiction).
Governor*
Moreover, given that the point of releasing people
*Includes federal prisons, which are under the jurisdiction of
from prison due to COVID was to facilitate social the U.S. President, and which was Republican at the start of
distancing for those that remained, we expected that the pandemic.
the jurisdictions with larger prison populations would
be more inclined to pursue releases as a strategy for protecting the health of the individuals in their charge.
And we anticipated that this public health concern would be great enough to transcend politics.

But though politics may have played a role in which
jurisdictions chose to make releases to reduce their
prison population size, it does not appear to have
played as great of a role in the proportion of people
released. For both Democratic and Republican
jurisdictions, the majority made releases amounting
to less than 10% of the 2019 prison population.
Only eight jurisdictions released the equivalent of
more than 10% of their 2019 prison populations,
and these favored jurisdictions with Republican
leadership (Figure 11).

Number of Jurisdictions

But over the course of the pandemic, the management of COVID itself became a highly polarized
political issue. States headed by Democratic governors were more likely to impose statewide mask
mandates and other more extreme measures to manage the spread of the virus while states headed by
Republican governors were less likely to impose such measures.68 These differences in approach may have
also affected prison releases. Of those jurisdictions that released people from prison, three-fifths were
in jurisdictions headed by Democratic governors
and two-fifths were in jurisdictions headed by
Figure 11. Prison Release Proportions by
Republican governors (or the Presidency). In
Gubernatorial Political Party in 2020
contrast, of those states that did not release people
from prison due to the pandemic, all but three
■ Democractic Governor ■ Republican Governor*
20
were headed by Republican governors (Table 4).
19

15

13

10

8

5
0

3
No Releases

6
2

1-9%

10% or greater

COVID Releases as a Propotion of 2019 Prison Population

*Includes federal prisons, which are under the jurisdiction of
the U.S. President, and which was Republican at the start of
the pandemic.

68. Brian Neelon et al., Associations Between Governor Political Affiliation and COVID-19 Cases, Deaths, and Testing in the U.S.,
61 Am. J. Preventative Med. 115, 116 (2021).

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

26

Examining Prison Releases in Response to COVID 		

Revisiting Our Assumptions About Release

When we overlay gubernatorial leadership with the system type (i.e., indeterminate vs. determinate), a
further pattern develops. As shown in Figure 12, jurisdictions with Democratic leadership were nearly
evenly split between determinate and indeterminate sentencing. Yet nearly all of these jurisdictions
had modest releases of just 1-9%. In contrast, almost all of the jurisdictions with Republican leadership
were indeterminate systems. In these states, the more common response was not to make a release at
all. But for those that did, the indeterminate jurisdictions were evenly split between modest and large
releases while the determinate systems made mostly modest releases. But across all groups, 7 of the
8 jurisdictions with the largest releases—those greater than 10% of the 2019 prison population—were
indeterminate in structure. Thus, overall, we can conclude that whether a system was determinate or
indeterminate did not affect whether a jurisdiction made releases due to COVID. Instead, gubernatorial
leadership played a larger role in that decision, and fewer jurisdictions with Republican leadership made
releases. However, determinacy appears to have contributed to the scale of releases with indeterminate
jurisdictions making larger releases than determinate jurisdictions regardless of political leadership.

Figure 12. COVID Releases as a Function of Gubernatorial Leadership and System Type
12

Number of Jurisdictions

12

10

10

9

8

No Releases

6

5

■

5

■ 10% or greater

4
2

2

1-9%

3
2
1

1

1

0

0

Indeterminate

Determinate

Democratic Governor

Indeterminate

Determinate

Republican Governor*

COVID Releases as a Proportion of 2019 Prison Populations

*Includes federal prisons, which are under the jurisdiction of the U.S. President, and which was Republican at the start of the
pandemic.

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

27

Effect on Prison
Population Size
Several studies have examined the effects of the pandemic on the size of the incarcerated population in
the US. Some estimates show a reduction of 71,030 in the number of individuals incarcerated in state and
federal prison during the first half of 2020, when states made the majority of releases.69 Another estimate
shows that that during the first year of the pandemic, roughly March 2020 to July 2021, the number of
people in state prison decreased by about 16.3%, from approximately 1.23 million to 1.03 million people.70
Given the sheer size of the U.S. prison population, this decrease is quite dramatic – resulting in 200,000
fewer individuals incarcerated in state prisons by the middle of 2021.71 Similar reductions likely occurred
in federal prisons.
These reductions were the result of both decreased admissions and increased prison releases compared
to previous years. During the pandemic, states also continued to make routine releases from prison, as
people became eligible for parole or reached their mandatory release dates. We conclude that nonroutine COVID-19 releases had a moderate effect on reducing the prison population. Our analysis shows
that COVID-19-related releases—most of which occurred during the first year of the pandemic—allowed
approximately 80,658 individuals to be released from prison. Of this number, we estimate that 39,588
were released from federal prisons, and 41,070 were release from state prisons. Since the explicit goal
of these release events was to reduce prison populations in order to facilitate social distancing and/
or release individuals who were at an increased risk of complications from the virus, these releases
were unlikely to happen during a non-pandemic year. We approximate that at the state level, COVID-19
releases accounted for about one-fifth (20.5%) of the overall reduction in state prison populations across
the U.S. Thus, COVID related prison releases made up a modest, but not insignificant, percentage of all
prison population decreases during the pandemic. The 80,658 releases represent about 5-1/2% of the
total state and federal prison population in 2019.
As other research has concluded,72 the “largest, most rapid single-year decrease in prison population
in American history”73 mainly occurred as a result of a reduction in prison admissions. The reduction in
prison admissions was due to decreases in police-citizen interactions resulting in fewer arrests, temporary
court closures that led to a disruption in court proceedings,74 and state-level policies severely curtailing
imprisonment by refusing transfers from jails to prisons for newly sentenced individuals and prohibiting
or limiting imprisonment for technical violations of probation and parole.75 These factors reduced prison
populations by a significantly greater extent than back-end efforts to release incarcerated individuals.

69. Franco-Paredes, C., Ghandnoosh, N., Latif, H., Krsak, M., Henao-Martinez, A. F., Robins, M., ... & Poeschla, E. M. (2021). Decarceration
and community re-entry in the COVID-19 era. The Lancet Infectious Diseases, 21(1), e11-e16.
70. Klein, B., Ogbunugafor, C. B., Schafer, B. J., Bhadricha, Z., Kori, P., Sheldon, J., & Hinton, E. (2021). The COVID-19 pandemic
amplified long-standing racial disparities in the United States criminal justice system. medRxiv
71. Id.
72. Council of State Governments, More Community, Less Confinement, https://csgjusticecenter.org/publications/morecommunity-less-confinement/national-report/.
73. Klein, et al., supra n. 70.
74. Julie Marie Baldwin, John M. Eassey, and Erika J. Brooke. “Court operations during the COVID-19 pandemic”. In: American
Journal of Criminal Justice 45.4 (2020), pp. 743–758.
75. Sharma, D., Li, W, Lavoie, D, and Lauer, C. Prison Populations Drop by 100,000 During the Pandemic. The Marshall Project,
https://www.themarshallproject.org/2020/07/16/prison-populations-drop-by-100-000-during-pandemic.

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

28

Lessons Learned and
Recommendations
The intersection of the COVID-19 pandemic with the effects of mass incarceration proved to be a real-time
demonstration of the power of back-end prison release discretion, a concept which we expand upon
in the Robina Institute’s report entitled American Prison Release Systems: Indeterminacy in Sentencing
and the Control of Prison Populations.76 Had there not been legal mechanisms in the back end of the
system through which jurisdictions could exercise discretion to effect releases, jurisdictions could not
have taken steps to reduce prison populations in an effort to slow the spread of the virus and protect
individuals from its effects. However, though 34 states and the federal government managed to exercise
these discretionary powers, releases in most places were modest, and this fact, along with other findings
in the report, impart several lessons about the feasibility of using back-end releases to reduce the effects
of mass incarceration.
1. States and the federal government have the tools to make large-scale releases, but some
modifications are needed.
The states and federal government have enormous power to make discretionary back-end release
decisions. Even in jurisdictions with determinate sentencing systems—places where we thought there
would be stronger legal impediments to releasing people from prison—there were numerous avenues
for release. Thus, a lesson learned is that when willing to do so, states and the federal government can
accomplish large-scale prison releases – mainly using mechanisms already available to them. However,
during the pandemic, this required addressing barriers to making these mechanisms effective for
large-scale releases, such as expanding eligibility requirements, taking a top-down rather than caseby-case approach, or assigning resources to expediting considerations for parole and other pre-release
reviews. These modifications represent real opportunity to continue large-scale releases for the purpose
of reducing prison populations, especially in jurisdictions where prisons are already overcrowded. Thus,
jurisdictions could consider making these modifications permanent. For example:
∙ At least four states that included advanced age in one or more of their targeted groups for
COVID-19 releases, do not have a geriatric parole provision in their state code. These and other
jurisdictions could enact such second-look provisions which would allow the courts or the parole
board to consider the release of elderly people serving long sentences.
∙ A few jurisdictions eliminated caps on the number of good time or earned time credits that
could be earned (often based on offense type) to advance the date of a person’s release. Research
has shown that incentives are more powerful at promoting behavioral change than sanctions.
Yet, eligibility rules for one of the most powerful incentives—earlier release from prison—are often
narrowly restricted to people already considered to be less of a risk to public safety. Thus, we
should be thinking about expanding these types of incentives to the people for whom we most
want to see behavioral change—those considered a greater risk to public safety. The counterpoint
76. Reitz, K., Rhine, E., Lukac, A., and Griffith, M. (2022). American Prison-Release Systems: Indeterminacy in Sentencing and the
Control of Prison Population Size. Final Report, Robina Institute of Criminal Law and Criminal Justice. https://robinainstitute.
umn.edu/publications/american-prison-release-systems-indeterminacy-sentencing-and-control-prison-population.

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

29

Examining Prison Releases in Response to COVID

Lessons Learned and Recommendations

to this is that people who have been convicted of more serious offenses should serve longer
terms in prison. But good time and earned time credits allow for both things to occur by allowing
for earlier release only if the person adheres to the requirements necessary to earn the credits.
Expanding the availability of good time or earned time credits on a permanent basis would give
people in prison more agency to determine the length of time served, and incentivize them to
continue participating in rehabilitative programing, which, if properly resourced to provide quality
programming, will in turn foster long-term behavioral change and reduce recidivism.
∙ Similarly, many jurisdictions suspended eligibility requirements for home confinement and work
release in order to broaden the pool of people who could be released using these mechanisms.
All jurisdictions could review these eligibility requirements, which are often unnecessarily narrow
and severely reduce the effectiveness of the early release mechanism.
∙ Most jurisdictions that utilized commutations did so on a large-scale basis by categorically
considering groups of individuals meeting specific criteria rather than requiring individual
applications. Prior to the pandemic, a similar process was set in motion by President Obama to
reduce sentences for nonviolent drug offenders.77 Similar large-scale efforts could continue to be
used to address sentences that appear to be out of step with modern ideals, and doing so could
streamline or bypass the lengthy individualized review required in most states today.
2. The pandemic forced jurisdictions to think differently about technical violations of supervision.
As noted in our findings, four release groups relied on sanction reductions, wherein corrections
departments used their discretion to impose sanctions for violations of parole to do the opposite; that
is, to reduce the sanctions previously imposed and release people back onto parole. And though we
did not chronicle it in this report, many jurisdictions also refused to accepted prison admissions due to
technical violations of probation or parole. This suggests that many jurisdictions did not view prison as a
necessary response to technical violations of supervision. Some jurisdictions are taking the opportunity
to rethink the use of incarceration as a sanction for violations altogether.78 This lesson is one that could
be taken universally.
3. Jurisdictions could reduce prison populations by increasing resources for back-end release
procedures.
One of the most prevalent criteria for release was time-served. In many cases, individuals were already
eligible for release because they had served the minimum required term or had accumulated sufficient
good time or earned time credits. But they had not yet been released because they still needed to go
through a pre-release review or a parole hearing or complete some form of programming. The common
thread here is that in all of these cases, all that was required to achieve release was a more concerted
effort to complete these steps. Thus, jurisdictions could reduce prison populations simply by putting
more resources into make release processes flow faster.

77. See e.g., Lorelei Laird, Clemency Project 2014 has submitted more than 1,000 petitions to the White House, ABA Journal, May
16, 2016, https://www.abajournal.com/news/article/clemency_project_2014_has_submitted_more_than_1000_petitions_to_
the_white_h.
78. See, e.g., Kelly Lyn Mitchell, Lily Hanrath, Erin Harbinson, Understanding Probation Violations and Disrupting the Revocation
Pathway in Ramsey County, Minnesota, Robina Institute of Criminal Law and Criminal Justice (2021), https://robinainstitute.
umn.edu/publications/understanding-probation-violations-and-disrupting-revocation-pathway-ramsey-county.

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

30

Examining Prison Releases in Response to COVID 		

Lessons Learned and Recommendations

4. Jurisdictions were risk averse in their approach to identifying people for release.
Though 35 jurisdictions made prison releases due to COVID, the majority of those releases included
a requirement that the persons being released have been convicted of non-violent crimes. Still more
required that the individuals be very close to having served their required minimum term. Thus,
jurisdictions were risk averse in making their releases. They focused on the people they thought would
carry the least risk of recidivating. Corrections officials in at least one jurisdiction noted that these were
also the releases that were publicly palatable, indicating that jurisdictions may have been concerned not
only about actual risk but also public perception about how such releases would impact public safety.
There are two lessons to be learned here. The first is that more focus needs to be placed on the
rehabilitation of individuals in prison and less emphasis needs to be placed on the offense for which
they were convicted. Often, those who are deemed to be non-violent and low risk to public safety
receive all the rewards in prison. They earn more good time credits and serve shorter proportions of their
pronounced sentences—all built upon the false dichotomy of “non-violent” versus “violent” offending.
But these approaches ignore the fact that regardless of the offense for which a person was convicted,
people in prison have varying criminogenic needs (factors that lead to reoffending), and it is only through
addressing those needs that they can change behavior and prevent future criminal offending. Yet the
programming necessary to address these needs is often under resourced. Thus, rather than relying on
this false dichotomy, corrections officials should focus on enhancing rehabilitative programming, and
legislatures should fund those programs at the levels needed to address crime.
The second lesson learned is that jurisdictions overrode their understanding of these rehabilitative
concepts and made decisions based on their assumptions about public perception. Thus, a second
recommendation would be to learn more about public attitudes towards crime and public safety.
For example, one research study on sentencing found the public had a less punitive attitude towards
considering older criminal history offenses or offenses committed when the person was a juvenile, and
that this was in contrast to how most sentencing guidelines were constructed.79 It may be that the public
has a more nuanced view of crime and public safety than corrections officials give them credit. Better
understanding of public attitudes may allow for the expanded use of existing legal mechanisms to
provide for second looks and permit the release of more people from prison, especially for groups that
have traditionally been excluded from such efforts, such as individuals serving time for violent offenses.
5. Efforts at using back-end release powers to reduce prison populations may have been hampered
by politics.
In order to use existing mechanisms to bring prison populations down to the levels equivalent to those
experienced in the U.S. before the onset of mass incarceration, jurisdictions would have to view mass
incarceration as an emergency in the same way they viewed COVID-19 as an emergency. But politics could
affect this equation. Criminal justice is a hot button issue often characterized by the false dichotomy of
being “soft on crime” or “tough on crime.” These concepts permeate political discourse, and can be central
to the campaign platforms upon which candidates for elected office run. Those who follow the “tough on
crime” line of thinking may see current sentencing and punishment practices as appropriate responses to
crime and criminal behavior. Our analysis showed that the political party of the governor in each state bore a
relationship to whether the state chose to make COVID-related prison releases, thus politics is likely to affect
whether states and the federal government would be willing to continue such releasing practices in order
to reduce the effects of mass incarceration. Thus, in order to address mass incarceration, it may be necessary
to redefine what it means to be “tough on crime” to include concepts that have been proven to reduce
reoffending such as using incentives and rehabilitative programming to promote behavioral change.

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

31

Examining Prison Releases in Response to COVID

Lessons Learned and Recommendations

6. Outside pressure may be needed to encourage back-end releases.
In several jurisdictions, back-end prison releases were triggered by external forces. In our review, these
triggering events were documented as executive orders or directives, or court orders as a result of
litigation. Executive orders served two functions. First, they offered the relief needed to expand eligibility
or suspend requirements, thereby facilitating use of legal mechanisms that otherwise would have served
as barriers. But in some jurisdictions, the orders served to spur action where it might not otherwise
have taken place, by, for example urging parole boards to move faster, setting up committees to review
cohorts of potential releasees, or ordering the department of corrections to identify potential groups for
large-scale releases. Litigation served a similar purpose, by forcing consideration of the health and safety
implications of prisons as congregate living settings. Thus, another lesson learned is that jurisdictions
may not take it upon themselves to view mass incarceration as an emergency requiring a solution, or
they may simply lack the power to address this issue without outside intervention. Outside pressure may
be needed to encourage, or make available, the broader use of back-end release discretion to reduce
prison population size.

79. Rhys Hester, Julian V. Roberts, Richard S. Frase, and Kelly Lyn Mitchell, A Measure of Tolerance: Public Attitudes on Sentencing
Enhancements for Old and Juvenile Prior Records. 3(2) Corrections 137 (2018).

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

32

Conclusion
Although the points above are discretely divided, the fact is that the lessons learned are intertwined.
Though jurisdictions had back-end release powers, they were hampered by procedural barriers in using
them on a large scale. Finding ways to overcome those barriers was largely a product of political will.
And in some cases, outside pressure in the form of litigation was necessary to prompt government and
corrections officials to act. Even when they overcame the barriers to using back-end release mechanisms,
jurisdictions took a very conservative approach to back-end releases, focusing on areas where they
believed there was less risk to public safety or where they thought they could garner more public support,
such as releasing people who committed non-violent offenses or who had very little time left to serve
on their sentence. As a result, while some jurisdictions were able to release a sizeable number of people
due to the pandemic, the people released tended to be individuals that were close to being released
anyway. Thus, the experience from the pandemic informs us that jurisdictions are unlikely to tackle the
issue of mass incarceration by using their discretionary back-end release authority unless we address
their risk aversion by redefining what it means to be “tough on crime.” Some suggested ways to do this
include assessing public attitudes for institutionalizing second look processes and eliminating the false
dichotomy of “non-violent” and “violent” offending and instead focusing on increasing the availability of
rehabilitative programming and incentives for people in prison to engage in that programming.

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

33

Appendix A
Prison Releases and Institutional Features by Jurisdiction
Jurisdiction

Prison
Pop. at
End of
2019

2019
Prison
Pop.
Rank

Total
number
of people
released

Release
Rank

Ratio of
number
released to
2019 prison
population

System Type

Gov Party
2020

Alabama

28,304

15

-

36

0%

Indeterminate

Republican

Alaska

4,475

44

-

36

0%

Indeterminate

Republican

Arizona

42,441

9

-

36

0%

Determinate

Republican

Arkansas

17,759

28

730

17

4%

Indeterminate

Republican

California

122,687

3

11,014

2

9%

Determinate

Democratic

Colorado

19,785

22

310

24

2%

Indeterminate

Democratic

Connecticut

12,823

31

357

8

3%

Indeterminate

Democratic

Delaware

5,692

40

-

36

0%

Determinate

Democratic

Federal

175,116

1

39,588

1

23%

Determinate

Republican

Florida

96,009

4

3

35

0.003%

Determinate

Republican

Georgia

54,816

5

918

15

2%

Indeterminate

Republican

Hawaii

5,279

42

-

16

0%

Indeterminate

Democratic

Idaho

9,437

34

-

36

0%

Indeterminate

Republican

Illinois

38,259

10

644

18

2%

Determinate

Democratic

Indiana

27,180

16

27

36

0.1%

Determinate

Republican

Iowa

9,282

35

197

19

2%

Indeterminate

Republican

Kansas

10,177

32

6

33

0.1%

Determinate

Democratic

Kentucky

23,082

21

190

11

1%

Indeterminate

Democratic

Louisiana

31,609

14

68

31

0.2%

Indeterminate

Democratic

Maine

2,185

49

95

29

4%

Determinate

Democratic

Maryland

18,595

27

2,000

7

11%

Indeterminate

Republican

Massachusetts

8,205

36

1,156

12

14%

Indeterminate

Republican

Michigan

38,053

11

500

21

1%

Indeterminate

Democratic

Minnesota

9,982

33

448

23

4%

Determinate

Democratic

Mississippi

19,417

23

-

36

0%

Indeterminate

Republican

Missouri

26,044

18

-

36

0%

Indeterminate

Republican

Montana

4,723

43

4

34

0.1%

Indeterminate

Democratic

Nebraska

5,682

41

-

36

0%

Indeterminate

Republican

Nevada

12,840

30

-

36

0%

Indeterminate

Democratic

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

(based on
total size)

(based
on total
number
released)

34

Examining Prison Releases in Response to COVID 		

Jurisdiction

Appendix A

Prison
Pop. at
End of
2019

2019
Prison
Pop.
Rank

Total
number
of people
released

Release
Rank

Ratio of
number
released to
2019 prison
population

System Type

Gov Party
2020

New Hampshire

2,691

47

-

36

0%

Indeterminate

Republican

New Jersey

18,613

25

6,381

3

34%

Indeterminate

Democratic

New Mexico

6,723

38

550

20

8%

Determinate

Democratic

New York

43,500

8

3,488

5

8%

Indeterminate

Democratic

North Carolina

34,079

13

5,409

4

16%

Determinate

Democratic

North Dakota

1,794

50

240

27

13%

Indeterminate

Republican

Ohio

50,338

6

112

28

0.2%

Determinate

Republican

Oklahoma

25,679

19

464

22

2%

Indeterminate

Republican

Oregon

14,961

29

253

26

2%

Determinate

Democratic

Pennsylvania

45,702

7

165

6

0.4%

Indeterminate

Democratic

Rhode Island

2,740

46

52

32

2%

Indeterminate

Democratic

South Carolina

18,608

26

-

36

0%

Indeterminate

Republican

South Dakota

3,801

45

-

36

0%

Indeterminate

Republican

Tennessee

26,349

17

-

36

0%

Indeterminate

Republican

Texas

158,429

2

-

36

0%

Indeterminate

Republican

Utah

6,665

39

1,000

14

15%

Indeterminate

Republican

Vermont

1,608

51

255

25

16%

Indeterminate

Republican

Virginia

36,091

12

1,376

10

4%

Determinate

Democratic

Washington

19,261

24

1,016

13

5%

Determinate

Democratic

West Virginia

6,800

37

70

30

1%

Indeterminate

Republican

Wisconsin

23,956

20

1,572

9

7%

Determinate

Democratic

Wyoming

2,479

48

-

36

0%

Indeterminate

Republican

Totals

1,430,805

5.6%

34 Ind.
17 Det.

27 Rep.
24 Dem.

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

(based on
total size)

80,658

(based
on total
number
released)

35

Appendix B
Release Mechanisms

Throughout this table, DOC refers to the jurisdiction’s Department of Corrections.
State with
Releases

Triggering
Event

Mechanisms
Used

Type of
Mechanism

Modification
to Existing
Mechanism

Description of
Modification

Description of
Mechanism

Arkansas

Executive Order

Prison
overcrowding
valve

Existing

X

Suspended
eligibility
requirements

DOC, via power of EO,
suspended eligibility
requirements for early
release pursuant to
prison overcrowding
valve.

Expedited
release
planning

Existing

X

Expedited
release process

DOC sped up the
review process for
expedited release by
creating on-site review
teams and working
overtime.

California

Colorado (1)

Executive Order

Good time /
earned time
credits

Existing

X

Lifted caps;
expanded
eligibility

EO lifted caps and
expanded eligibility
criteria on awards of
earned time credits,
which moved people to
their mandatory release
dates sooner.

Colorado (2)

Executive Order

Compassionate
release

Existing

X

Expanded
eligibility

EO expanded
eligibility criteria for
Special Needs Parole,
which is the state’s
compassionate release
provision.

Colorado (3)

Executive Order

Intensive
supervision
program

Existing

X

Expanded
eligibility

EO expanded eligibility
criteria for release
to the intensive
supervision program.

Connecticut
(1)

Furlough

Existing

X

Expedited
parole process

DOC Comm'r signs
policy exception
expanding eligibility
for furlough for people
serving two years or less
and who have served
40% of sentence.

Connecticut
(2)

Parole

Existing

X

Expedited
parole process

DOC, Parole Board
and Community
Supervision agencies
collaboratively
identified people to
target for discretionary
release, prioritizing
those who were older,
had health conditions,
and were deemed to
be low risk.

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

36

Appendix B

Examining Prison Releases in Response to COVID 		

State with
Releases

Triggering
Event

Mechanisms
Used

Type of
Mechanism

Modification
to Existing
Mechanism

Description of
Modification

Description of
Mechanism

Federal BOP
(1)

Court order

Home
confinement

Existing

X

Expedited
review

Court order to expedite
the release of certain
medically vulnerable
individuals to home
confinement.

Federal BOP
(2)

Court order

Compassionate
release

Existing

X

Expedited
review

Court order to expedite
the release of certain
medically vulnerable
individuals to
compassionate release.

Compassionate
release

Existing

Home
confinement

Existing

Florida

Compassionate
release

Existing

Georgia

Commutation

Existing

Federal BOP
(3)

Federal BOP
(4)

AG memo

Statute allows
federal courts to
reduce prisoners’
sentences if they find
“extraordinary and
compelling reasons” to
do so. Courts granted
compassionate release
via this statute.
X

Expanded
eligibility/
Expedited
release process

Attorney General
issued memorandum
encouraging home
confinement release
for eligible individuals
vulnerable to COVID-19
even when electronic
monitoring was not
available.
A few people were
released through the
normal process for
conditional medical
release, which is
Florida's compassionate
release provision.

Large scale use

Prioritized people who
were serving time for
non-violent offenses
who were close to their
sentence end date.

Illinois (1)

Executive Order

Good time/
earned time
credits

Existing

X

Suspended
notice
requirements

EO suspended notice
requirements for early
release due to good
conduct credit award;
DOC created task
force to prioritize older
individuals and those
close to their release
date for early release.

Illinois (2)

Executive Order

Compassionate
release

Existing

X

Suspended
14-day time
limit; expanded
eligibility

EO suspended 14-day
time limit on and
expanded eligibility for
medical furloughs.

Home
confinement

Existing

Illinois (3)

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

DOC created a task
force to prioritize older
individuals and those
close to their release
date for early release.

37

Examining Prison Releases in Response to COVID

State with
Releases

Mechanisms
Used

Type of
Mechanism

Indiana

Sentence
reduction

Existing

Iowa

Parole

Existing

Kansas

Home
confinement

Existing

Commutation

Existing

Louisiana

Compassionate
release

Existing

Maine

Home
confinement

Existing

X

Expedited prerelease review

DOC reviewed early
release cases faster.

Kentucky

Triggering
Event

Appendix B

Executive Order

Modification
to Existing
Mechanism

Description of
Modification

Description of
Mechanism
Sentence
modifications, based on
individual applications,
were granted due to
reasons related to
COVID-19.

X

Expedited
parole process

Implemented
a double-panel
approach, in which
two three-person
panels simultaneously
reviewed individuals for
parole.
Prioritized release
to house arrest of
individuals with short
time left on sentence.

X

Large scale use

Commutation power
used multiple times
to enact large group
releases. Prioritized
people based on
medical vulnerability,
non-violent sentence,
and time left to serve.
DOC used existing
medical furlough law
to establish criteria for
release and established
a review panel to
process cases.

Maryland (1)

Executive Order

Good time /
earned time
credits

Existing

X

Expanded
eligibility

EO suspended rules
about who can receive
good time credit.

Maryland (2)

Executive Order

Parole

Existing

X

Expedited
parole process

EO orders parole
board to accelerate
release for qualifying
cases, prioritizing older
individuals.

Massachusetts

Compassionate
release

Existing

Michigan

Parole

Existing

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

Parole board increased
their use of medical
parole approvals due to
COVID-19.
X

Expedited
parole process

MDOC requested
that prosecutors
sign waivers allowing
immediate release,
which removed the 28day waiting period after
parole decisions.

38

Appendix B

Examining Prison Releases in Response to COVID 		

State with
Releases

Triggering
Event

Mechanisms
Used

Type of
Mechanism

Modification
to Existing
Mechanism

Description of
Modification

Description of
Mechanism

Minnesota (1)

Compassionate
release

Existing

X

Expanded
eligibility

Created new process
for conditional medical
release, including
having the process
be applicant driven,
coordinating with state
and local resources,
and considering
medical conditions
not previously used for
conditional medical
release.

Minnesota (2)

Sanction
reduction

Existing

Minnesota (3)

Work release

Existing

X

DOC exercised its
authority to reduce
sanctions that had
been imposed for
technical violations of
supervised release.
Expanded
eligibility

DOC broadened
eligibility criteria for
work release.

Montana (1)

Governor
Directive

Compassionate
release

Existing

X

Governor directive for
DOC and parole board
to consider medically
vulnerable people for
early release.

Montana (2)

Governor
Directive

Parole

Existing

X

Governor directive for
DOC and parole board
to consider medically
vulnerable people for
early release.

New Jersey (1)

Executive Order

Home
confinement

Existing

X

Review
committee
created

EO created a review
committee and
directed DOC to
develop lists of
medically vulnerable
people who had short
time left on sentence
for the committee's
consideration for
release to emergency
medical home
confinement.

New Jersey (2)

Executive Order

Parole

Existing

X

Expedited
parole process

EO directed parole
board to expedite
release of older,
medically vulnerable
people who had a short
time left on sentence.

Good time/
earned time
credits

Newly
Created

Wholly new
form of credit

New statute created
public health
emergency credits (i.e.,
time served credit) to
be awarded to certain
people in prison
during a public health
emergency.

New Jersey (3)

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

39

Examining Prison Releases in Response to COVID

Appendix B

State with
Releases

Triggering
Event

Mechanisms
Used

Type of
Mechanism

Modification
to Existing
Mechanism

Description of
Modification

Description of
Mechanism

New Mexico

Executive Order

Commutation

Existing

X

Large scale use

Commutation power
used to enact large
group release, instead
of granting approval to
individual applicants,
prioritizing release of
people serving time for
lower-level offenses and
who were close to their
release date.

New York (1)

Unclear

Governor instructed
DOC to release
pregnant women
serving time for nonviolent offenses.

New York (2)

Unclear

DOC prioritized early
release for older people
who were serving
time for nonviolent,
nonsexual offenses and
who were close to their
release date.

North Carolina
(1)

Good time/
earned time
credits

Existing

North Carolina
(2)

Home
confinement

Newly
created

Parole

Existing

North Carolina
(3)

A .

Lawsuit

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

X

X

Increased use
of discretionary
credits

DOC gave sentence
credit to people who
were at an increased
risk of COVID-19 who
were serving time for
nonviolent, nonsexual
offenses, which
moved people to their
mandatory release
dates sooner.

New form
of home
confinement

DOC created new
program, Extending
the Limits of Home
Confinement (ELHC),
to identify people
who could serve the
remainder of their
sentence under home
confinement.
DOC settled a lawsuit
agreeing to reduce the
prison population. As
a result, people were
released from prison
through sentence
credits, the ELHC
program, release of
people confined for
supervised release
violations, and parole.

40

Examining Prison Releases in Response to COVID 		

Appendix B

State with
Releases

Triggering
Event

Mechanisms
Used

Type of
Mechanism

Modification
to Existing
Mechanism

North Carolina
(4)

Lawsuit

Sanction
reduction

Existing

X

Parole

Existing

X

Prison
overcrowding
valve

Existing

At the
recommendation
of the governor, the
legislative oversight
committee approved
early release for a group
of people in prison
via the state's prison
overcrowding statute.

Ohio (2)

Commutation

Existing

Commutation power
used to release seven
people from prison.

Oklahoma (1)

Commutation

Existing

Oklahoma (2)

Compassionate
Release

Existing

North Dakota

Ohio (1)

A .

Governor's
declaration
of emergency
overcrowding

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

X

Description of
Modification

Description of
Mechanism
DOC settled a lawsuit
agreeing to reduce the
prison population. As
a result, people were
released from prison
through sentence
credits, the ELHC
program, release of
people confined for
supervised release
violations, and parole.

Expedited
parole process

Large scale use

Parole Board held
special session to
identify a group of
people for early parole
release, prioritizing
those with medical
conditions, with nine
months or less on their
sentence, and a reliable
place of residence.

Commutation power
used to enact large
group release, instead
of granting approval to
individual applicants,
prioritizing non-violent,
or lower-level offenses,
in order to reduce
prison population.
DOC recommended
a group of individuals
for medical parole
(i.e., compassionate
release) who were
at an increased risk
from COVID-19 and
who were not serving
sentences for sex or
violent crimes.

41

Examining Prison Releases in Response to COVID

State with
Releases

Triggering
Event

Oregon

Appendix B

Mechanisms
Used

Type of
Mechanism

Modification
to Existing
Mechanism

Description of
Modification

Description of
Mechanism

Commutation

Existing

X

Large scale use

Commutation power
used to enact large
group release, instead
of granting approval to
individual applicants,
prioritizing those
particularly vulnerable
to COVID-19, who had
served at least half of
their sentences, who
were not convicted of
committing a violent
crime against another
person, who had
housing and access
to healthcare in the
community.

Pennsylvania

Executive Order

Reprieve

Existing

X

Large scale use

EO required DOC to
establish a Reprieve
of Sentence of
Incarceration Program
to recommend
individuals for
reprieve (temporary
release from prison),
prioritizing those who
were high risk for
complications due to
coronavirus, serving a
sentence for a nonviolent offense, and
within one year of
release.

Rhode Island

Petition to
SCt from
State Public
Defender

Sentence
reduction

Existing

X

Suspended
timing
requirement in
the rules

Supreme Court order
suspended the time
limit for motions
to resentence and
authorized lower
courts to reduce the
sentences of identified
people with 90 days or
less until expiration of
sentence.

Utah

Parole

Existing

X

Expedited
parole process

Individuals who had
a confirmed parole
release date in the
near future were
released early by
removing program
completion and other
requirements.

Vermont (1)

Furlough

Existing

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

DOC prioritized release
for people who had
served their minimum
sentence or qualified
for furloughs.

42

Examining Prison Releases in Response to COVID 		

State with
Releases

Triggering
Event

Mechanisms
Used

Type of
Mechanism

Vermont (2)

Parole

Existing

Virginia

Early Release
Plan

Newly
created

Washington (1)

Commutation

Existing

Home
confinement

Washington
(3)

Appendix B

Modification
to Existing
Mechanism

Description of
Modification

Description of
Mechanism
DOC prioritized release
for people who had
served their minimum
sentence or qualified
for furloughs.

New legislation

Governor requested
legislation authorizing
DOC to create a
program for early
release, Inmate Early
Release Plan, which
allowed DOC to
discharge or place
inmates in lower level
of supervision.

X

Large scale use

Commutation power
used to enact large
group release, instead
of granting approval to
individual applicants,
prioritized people who
were serving time for
non-violent offenses
and who were close to
their release date.

Existing

X

Suspended
statutory
requirements

Due to Governor's
emergency
proclamation, DOC
modified its Rapid
Reentry program to
release individuals who
were at an increased
risk of complications
from the virus to serve
out the remainder of
their sentence on home
confinement with
electronic monitoring.

Work release

Existing

X

Suspended
statutory
requirements

Due to Governor's
emergency
proclamation, DOC
secretary issued
emergency work
release furloughs for
eligible individuals.

West Virginia
(1)

Sanction
reduction

Existing

DOC exercised its
authority to reduce
sanctions for parole
violations.

West Virginia
(2)

Work release

Existing

DOC exercised its
authority to extend
work release furloughs.

Wisconsin (1)

Good time /
earned time
credits

Existing

Certified Earned
Release program
for people serving
sentences related to
their substance abuse
disorder.

Washington
(2)

A .

Governor's
Emergency
Proclamation

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

43

Examining Prison Releases in Response to COVID

State with
Releases

Mechanisms
Used

Type of
Mechanism

Modification
to Existing
Mechanism

Description of
Modification

Description of
Mechanism

Wisconsin (2)

Parole

Existing

X

Expedited
parole process

Parole Board made
concerted effort to
grant discretionary
parole to some
individuals. Unclear if
certain groups were
prioritized.

Wisconsin (3)

Sanction
reduction

Existing

A .

Triggering
Event

Appendix B

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

DOC used their existing
powers to release
people serving time for
technical violations.

44

Examining Prison Releases in Response to COVID

Appendix C
Combinations of Release Criteria by Jurisdiction and Release Group
Jurisdiction*

Type of Crime
Nonviolent /
Low-level
offenses

Nonsexual
offenses

No
crimes
against a
person

Low-risk
to public
safety

Arkansas

Circle

Circle

California (1)

Circle

Circle

California (2)

Circle

Circle

Circle

California (3)

Circle

Circle

Circle

California (4)

Circle

Circle

Circle

California (5)

Circle

Colorado (1)
Colorado (2)
Colorado (3)

Circle

Lowrisk of
recidivism

Conduct
in prison

Technical
violations of
supervision

Medically
vulnerable

Fulfilled a
certain %/
amount
of
sentence

% or
amount
served
req'dc

Already
eligible

Short time
left on
sentence

Amount
of time
left on
sentence

Circle

60 days

Circle

180 days
1 year

Circle

Circle

<29

Circle

1 year

Circle

Circle

Circle

65

Circle

Circle
Circle

180 days

Circle

Release
date
before
August
2021 (
up to 16
mos)

Circle

Circle

Circle

Circle

12 mos

Circle

Housing
plan

Circle

Circle
Circle

Connecticut (3)

Circle

ROBINA INSTITUTE

Accrual of
sufficient
good time
credits

Circle

Circle

OF CRIMINAL LAW ANO CRIMINAL JUSTICE

Age

Reentry

30

Connecticut (2)

A .

Agebased

Time served

Circle

Circle

Connecticut (1)

COVID risk to person in
prison

Risk to Public Safety

Circle

50
Circle

40%

45

Examining Prison Releases in Response to COVID

Jurisdiction*

Type of Crime
Nonviolent /
Low-level
offenses

Nonsexual
offenses

No
crimes
against a
person

Appendix C
COVID risk to person in
prison

Risk to Public Safety
Low-risk
to public
safety

Lowrisk of
recidivism

Conduct
in prison

Federal BOP (1)

Technical
violations of
supervision

Medically
vulnerable

Agebased

Age

Time served
Accrual of
sufficient
good time
credits

% or
amount
served
req'dc

Already
eligible

Circle
Circle

Circle

Circle

Florida

Amount
of time
left on
sentence

Circle

18
months

Housing
plan

Circle

25%

Circle

Circle

Circle

Circle

Georgia

Circle

Illinois (1)

Circle

180 days

Circle

9 months

Circle

12
months

Circle

Illinois (2)
Illinois (3)

Circle

Illinois (4)

Circle

Iowa (1)

55

Circle

25%

Circle

Circle

Iowa (2)

Circle

Kansas

Circle

Kentucky (1)

Circle

Circle

Kentucky (2)

Circle

Circle

Louisiana

Circle

Circle

A .

Short time
left on
sentence

Circle

Federal BOP (2)
Federal BOP (3)

Fulfilled a
certain %/
amount
of
sentence

Reentry

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

Circle

Circle

Circle

< 5 years

Circle

Circle

<6
months

Circle

Circle

6 months

Circle

46

Examining Prison Releases in Response to COVID

Jurisdiction*

Type of Crime
Nonviolent /
Low-level
offenses

Nonsexual
offenses

Maine

Maryland (1)

Circle

Circle

Maryland (2)

Circle

Circle

Appendix C
COVID risk to person in
prison

Risk to Public Safety

No
crimes
against a
person

Low-risk
to public
safety

Circle

Circle

Lowrisk of
recidivism

Conduct
in prison

Technical
violations of
supervision

Accrual of
sufficient
good time
credits

Fulfilled a
certain %/
amount
of
sentence

% or
amount
served
req'dc

Already
eligible

50%

Circle

60

Circle

60

Reentry
Short time
left on
sentence

Amount
of time
left on
sentence

Circle

1 year (18
months
"considered")

Circle

4 months

Circle

Housing
plan

Circle

Circle

Circle

Minnesota (1)

Circle

Circle

Circle

Minnesota (2)

Circle

Circle
Circle

Circle

Montana (1)

Circle

Montana (2)

Circle

Montana (3)

Circle

Circle

Circle

Circle

New Jersey (2)

Circle

New Jersey (3)

Circle

New Jersey (4)

Circle

Circle

New Jersey (5)

Circle

Circle

New Jersey (6)

Circle

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

50%

Circle

90 days

Circle

Not
specified

Circle

90 days

Circle

New Jersey (1)

A .

Age

Circle

Circle

Minnesota (3)

Agebased

Circle

Massachusetts

Michigan

Medically
vulnerable

Time served

65

Circle
Circle

Circle
Circle
Circle

60

Circle

Circle
Circle
Circle

Circle

1 year

Circle

47

Examining Prison Releases in Response to COVID

Jurisdiction*

Type of Crime
Nonviolent /
Low-level
offenses

New Mexico

Nonsexual
offenses

Circle

New York (1)

Circle

New York (2)

Circle

Circle

North Carolina
(1)
North Carolina
(2)
North Carolina
(3)

Circle

Circle

North Carolina
(4)

Appendix C
COVID risk to person in
prison

Risk to Public Safety

No
crimes
against a
person

Low-risk
to public
safety

Lowrisk of
recidivism

Conduct
in prison

Technical
violations of
supervision

Medically
vulnerable

Age

Accrual of
sufficient
good time
credits

Fulfilled a
certain %/
amount
of
sentence

% or
amount
served
req'dc

Circle

Circle
Circle

55

Already
eligible

Reentry
Short time
left on
sentence

Amount
of time
left on
sentence

Circle

30 days

Circle

6 months

Circle

90 days

Circle

Circle

Circle

Circle

Circle

Circle

65

Circle

Circle

Circle

Circle

50(F)

Circle

12
months

Circle

Circle

Circle

65

Circle

12
months

Circle

9 months

Circle

90 days

North Dakota

Circle

Ohio (1)

Circle

Ohio (2)

Circle

Ohio (3)

Circle

Oklahoma (1)

Circle

Oklahoma (2)

Circle

Housing
plan

Circle

Circle
Circle

Circle

Oregon (1)

Circle

Circle

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

Circle

Circle

50%
Circle

Oregon (3)
Oregon (4)

60

Circle

Oregon (2)

A .

Agebased

Time served

2 months

Circle
Circle

Circle

Circle
Circle

Circle

6 months

Circle

48

Examining Prison Releases in Response to COVID

Jurisdiction*

Type of Crime
Nonviolent /
Low-level
offenses

Nonsexual
offenses

Oregon (5)

Appendix C
COVID risk to person in
prison

Risk to Public Safety

No
crimes
against a
person

Low-risk
to public
safety

Lowrisk of
recidivism

Circle

Conduct
in prison

Technical
violations of
supervision

Circle

Pennsylvania
(1)

Circle

Circle

Circle

Pennsylvania
(2)

Circle

Circle

Circle

Rhode Island

Circle

Medically
vulnerable

Agebased

Age

Time served
Accrual of
sufficient
good time
credits

Fulfilled a
certain %/
amount
of
sentence

% or
amount
served
req'dc

Already
eligible

Circle
Circle

NA

Circle

Vermont

Circle

Washington

Circle

Circle

Circle

Circle

Circle

West Virginia

Circle

Wisconsin (1)

Circle

Wisconsin (2)

Circle

Wisconsin (3)

- -------Totals

39

19

9

15

5

5

Amount
of time
left on
sentence

3

29

Housing
plan

Circle

Circle

Circle

Short time
left on
sentence

Circle

Utah

Virginia

Reentry

Circle

9-12
months

Circle

9-12
months

Circle

90 days

Circle

Circle

1 year

Circle

Circle

180 days

Circle

Not
specified

Circle

12
months

Circle
16

6

6

6

36

18

*Jurisdiction does not include Indiana or Washington, D.C. The information we gathered for Indiana did not contain enough detail to describe release groups in this way. Those convicted in Washington, D.C., are housed
in federal prisons, so their releases are included in the federal release groups. Jurisdictions that are listed more than once had distinct release groups utilizing different criteria.

49

Appendix D.
Compassionate Release and Geriatric Parole Provisions
by Jurisdiction
This appendix sets forth the results of our statutory survey of compassionate release and geriatric parole provisions.
There are three tables. The first details jurisdictions that have compassionate (medical) release provisions. The
second table details states that have provisions requiring the combination of age plus a medical condition. The
third table details jurisdictions that have provisions allowing for release based on age, often combined with time
served. Prior to the first and third tables, we briefly describe the major themes in eligibility and exclusion factors
for each type of release. In some jurisdictions compassionate release and geriatric parole are included in the same
statutory provision, but for purposes of our analysis, we broke them out separately to better analyze the distinct
requirements for each release mechanism.

COMPASSIONATE RELEASE PROVISIONS
∙ Eligibility: Of the 51 jurisdictions that have compassionate release, all maintain eligibility requirements
that categorize different types of medical conditions that would qualify an individual for consideration.
» Terminal Illness. A majority of jurisdictions allow consideration for compassionate release if an
individual has a terminal illness. However, the life expectancy required for an illness to qualify as
terminal varies greatly. Kansas has the strictest statute, requiring that death is likely within 30 days,80
while Arkansas and South Carolina have the longest time frame, requiring that death is likely to
result within 2 years.81 The most common time frame is 12 months.
» Permanent or Severe Incapacitation. A second category of eligibility is for people who have a
permanent or severe incapacitation. These conditions may not be terminal, but are serious enough
to render the person unable to care for themselves or severely limit their functioning, or require a
degree of medical care that is not possible in the prison facility. This category includes people who
would benefit from hospice or palliative care.
» Cognitive or Mental Disorders. A third area, but less common, are provisions allowing compassionate
release for cognitive or mental disability.
∙ Exclusion Groups: Most jurisdictions also have clear exclusion groups for compassionate release. Most
frequently, people who are sentenced to death, life without parole, or life sentences are ineligible to
apply for compassionate release. These are generally people who have been convicted of very serious
offenses such as murder or particularly violent sex crimes. A few jurisdictions also exclude people who
have been convicted of less serious offenses. Of the 51 compassionate release statutes, about 80% also
explicitly take into account an individual’s risk to public safety either at the initial eligibility phase or
during the review and hearing phase (once the initial eligibility factors have been met).

80. Kan. Stat. § 22-3729 (2022).
81. Ark. Code Ann. § 12-29-404 (2022); S.C. Code Ann. § 24-21-715 (2022).

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

50

Examining Prison Releases in Response to COVID 		

Appendix D

Table 5. Compassionate Release Provisions by Jurisdiction
* Indicates statute was enacted in 2021 or 2022 and was not in place at the start of the pandemic.

Eligibility Groups
Jurisdiction

Alabama
Ala. Code § 15-22-41
to -43
Ala. Code § 14-14-1 to -7

Alaska

Terminal
Illness?

Permanent or
Severe Physical
Incapacitation?

Other
Conditions?

"Incurable
condition"
Likely to result
in death
within 12
months

Render individual
unable to
perform daily life
functions without
assistance

--

--

Medical disability
that requires care
and supervision
better addressed
outside the prison

Arizona

If expected
to die
imminently
(3-6 months)

Ariz. Rev. Stat. Ann. § 411604.11 (B)

Arkansas
Ark. Code Ann. § 1229-404
Ark. Code Ann § 16-93708

California
Cal. Penal Code § 3550
Cal. Code Regs. tit. 15, §
3359.1 to .6
Cal. Penal Code § 1170(e)
Cal. Code Regs. tit 15, §§
3076, 3076.3 to .5

Colorado
Colo. Rev. Stat § 17-22.5403.5
Colo. Rev. Stat § 17-1-102
Colo. Rev. Stat § 17-2-201

Connecticut
Conn. Gen. Stat. §§ 54131a to

Yes

Cognitive
disability that
requires care
and supervision
better
addressed
outside the
prison

Sexual assault or
sexual abuse

Yes

--

Temporary
removal for
provision
of medical
treatment not
available in
prison

None

Not Explicitly

N / Not
explicitly
stated

"Incurable
condition"
Likely to result
in death
within two
years

Permanent
and Irreversible
incapacitation
requiring
immediate and
long-term care

Individual would
benefit from
hospice care

Sentenced
to death or
life without
possibility of
parole, sexual
offense

Yes

Y

"Incurable
condition"

Results in
permanent
inability to
perform daily
life activities and
need for 24-hour
care

--

Sentenced
to death or
life without
possibility of
parole or firstdegree murder
of a peace
officer

Yes

Irreversible
physical illness,
condition, or
disease that
requires costly
medical care of
treatment

Behavioral or
mental health
disorders that
require care and
costly treatment

Sentenced to
life without
parole, felony
crime of
violence

Yes

Debilitation,
incapacitation,
or infirmity as
a result of a
non-terminal
condition,
disease, or
syndrome

Requiring
continuous
palliative or endof-life care

Likely to result
in death
within 12
months

Anticipative
life
expectancy
of twelve
months or less

Six months or
less to live

-131g; -131k

A .

If yes, when?
Capital murder
or sex offenses

Alaska Stat. § 33.16.085

Ariz. Rev. Stat. Ann. §
31-403(D)

Exclusion
Group(s)

Any
consideration
for risk to public
Revocable?
safety?

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

Y

Eligibility Stage

Y

Eligibility Stage

Review/Hearing
Stage

Y

Eligibility Stage

Y

Review/Hearing
Stage

(Doesn't apply
to terminal
illness)

Capital felonies
or murder
with special
circumstances

Yes

Y

Eligibility Stage

51

Examining Prison Releases in Response to COVID

Appendix D

Eligibility Groups
Jurisdiction

Delaware

Terminal
Illness?

Permanent or
Severe Physical
Incapacitation?

Other
Conditions?

--

Serious medical
illness or infirmity

--

Incurable,
progressive illness
or debilitating
injury

--

No recovery
and death is
imminent

Permanent
and irreversible
incapacitation

--

"Entirely
incapacitated"

--

Del. Code Ann. tit. 11, §
4217(a) to (c)

Exclusion
Group(s)

Any
consideration
for risk to public
Revocable?
safety?
If yes, when?

None

Yes

Y

Eligibility Stage

Del Code Ann. tit. 11, §
4346(e)

Federal BOP
18 U.S.C. § 3582(c)(1)(A)

Disease or
condition with
end-of-life
trajectory

None

Yes
Review/Hearing
Stage

N / Not
explicitly
stated

Death within
eighteen
months

Florida
Fla. Stat. § 947.149
Fla. Admin. Code Ann. r.
33-601.603(7)(b)

Georgia
Ga. Code Ann. § 42-943(b)

Hawaii
Haw. Code. R. §§ 23700-26

Idaho
Idaho Code § 20-1006

Illinois
730 Ill. Comp. Stat.
5/3-3-14*

Indiana
220 Ind. Admin. Code
1.1-4-1.5

Sentenced to
death

Yes

People with
disabilities

None

Yes

Y

Eligibility Stage

Y

Eligibility Stage

Likely to result
in death
within 12
months

Progressive
and incurable
condition for
which death
is imminent

Impedes
capacities so
much that care
outside of prison
would be more
appropriate

Cognitive
impairment
such that
prison is not
rehabilitative

None

Not Explicitly

N / Not
explicitly
stated

Irreversibly
terminally ill

Permanently
and irreversibly
physically
incapacitated

--

Sentenced to
death

Yes

Y

Irreversible
and incurable

Severe,
permanent
medical condition
that prevents
individuals from
functioning
independently

Condition
that will result
in mental
incapacity
within six
months

None

Yes

--

Condition that
would be more
effectively
treated in
another facility

None

Likely to cause
death within
18 months

Terminal
medical
condition

Eligibility Stage

Review/Hearing
Stage

Yes
Eligibility Stage

N / Not
explicitly
stated

N / Not
explicitly
stated

Iowa (no provisions)

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

52

Examining Prison Releases in Response to COVID 		

Appendix D

Eligibility Groups
Jurisdiction

Kansas
Kan. Stat. Ann § 22-3728
to -3729

Kentucky
Ky. Rev. Stat. Ann §
439.3405

Louisiana
La. Stat. Ann. § 15:574.20

Maine
Me. Stat. tit. 34-A, § 3036A(10)*

Maryland

Terminal
Illness?

Permanent or
Severe Physical
Incapacitation?

Other
Conditions?

Likely to cause
death within
30 days

Significant
incapacitation to
the extent that
person does not
have the capacity
to cause physical
harm

--

Likely to result
in death
within 1 year

Massachusetts

Michigan
Mich. Comp. Laws §
791.235(10), (11)
Mich. Comp. Laws §
791.244

Minnesota
Minn. Stat. §244.05

Mississippi
Miss. Code Ann. § 47-7-4

A .

If yes, when?
Sentenced to
death or life
without parole,
sentenced for
an "off-grid"
offense

Yes

Severe lung, heart, -or neuromuscular
disease; limited
mobility; or
dependency on
life support

None

Yes

Likely to result
in death with
1 year *

Unable to
perform activities
of daily living;
confinement to a
bed or chair

--

First- or seconddegree murder
or sentenced to
death

Yes

Terminal
condition for
which care
outside prison
is appropriate

Severely
incapacitating
condition for
which care
outside prison is
appropriate

--

Anyone with
greater than
minimumsecurity
classification

Yes

--

Chronically
debilitated or
incapacitated

Mental health
conditions
included as well

Sentence to
any convictions
without
possibility of
parole.

Yes

Likely to cause
death with 18
months

Physical or
cognitive
incapacitation
that appears
irreversible

--

None

Yes

Terminal
condition
that is serious
and complex,
cannot
function
without
personal
assistance

Non-terminal
condition
that is serious
and complex,
cannot function
without personal
assistance

Mental disorder
which results in
impaired ability
to do daily
activities

Sentenced to
life without
parole or firstdegree criminal
sexual offense

Yes

Likely to result
in death
within 12
months

Grave illness
or condition
requiring
extended medical
management

--

None

Yes

Significant
and
permanent
physical
medical
condition that
is terminal in
nature

Significant and
permanent
physical medical
condition
that is totally
incapacitating
and/or terminal in
nature

--

Md. Code Ann., Corr.
Servs. § 7-309

Mass. Gen. Laws ch. 127,
§ 119A

Exclusion
Group(s)

Any
consideration
for risk to public
Revocable?
safety?

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

Y

Review/Hearing
Stage

Eligibility Stage

N / Not
explicitly
stated

Y

Review/Hearing
Stage

Y

Review/Hearing
Stage

Y

Eligibility Stage

Y

Eligibility Stage

Y

Eligibility Stage

Y

Eligibility Stage

Sex crimes

Not Explicitly

Y

53

Appendix D

Examining Prison Releases in Response to COVID

Eligibility Groups
Jurisdiction

Missouri
Mo. Rev. Stat. § 217.250

Montana
Mont. Code Ann. § 4623-210

Nebraska
Neb. Rev. Stat. § 831,110.02

Nevada
Nev. Rev. Stat. § 209.3925

New Hampshire
N.H. Rev. Stat. Ann. §
651-A:10-a

New Jersey
N.J. Rev. Stat. § 30:4123.51e*

New Mexico
N.M. Stat. Ann. § 31-21-25.1

Terminal
Illness?

Permanent or
Severe Physical
Incapacitation?

Other
Conditions?

Likely to result
in death with
6 months

--

Confinement
is greatly
endangering or
shortening the
person's life

Sentenced to
life without
parole, those
who have yet to
serve minimum
term

Not Explicitly

Y

Likely to cause
death within
6 months

Medical condition
that requires
extensive medical
attention

--

Sentenced to
death or life
imprisonment
without the
possibility of
release

Yes

Y

Terminally ill
because of
an existing
medical
condition

Permanently
incapacitated
because of an
existing medical
condition

--

Sentenced to
death or life
imprisonment
without the
possibility of
release

Yes

Expected to
die within 18
months

Physically
incapacitated or
in ill health

--

Sentenced to
death or life
imprisonment
without the
possibility of
release

Yes

Terminal
illness
requiring
excessive
care and
treatment
costs

Debilitating,
incapacitating, or
incurable medical
condition or
syndrome
requiring
excessive care
and treatment
costs

--

Sentenced to
death or life
imprisonment
without the
possibility of
release

Yes

Expected to
die within 6
months

Permanent
physical
incapacity that
results in inability
to perform daily
life activities and
creates a need for
24-hour care

--

None

Yes

"Incurable
condition"

Permanent
and irreversible
physical
incapacitation

--

Likely to result
in death
within six
months

A .

Exclusion
Group(s)

Any
consideration
for risk to public
Revocable?
safety?

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

If yes, when?

Eligibility Stage

Y

Review/Hearing
Stage

Y

Eligibility Stage

Y

Review/Hearing
Stage

Y

Review/Hearing
Stage

First-degree
murder

Yes

Y

Review/Hearing
Stage

54

Appendix D

Examining Prison Releases in Response to COVID 		

Eligibility Groups
Jurisdiction

New York
N.Y. Exec. Law § 259-r
to -s

North Carolina
N.C. Gen. Stat. § 15A-1369
to 1369.5
N.C. Gen. Stat. § 148-4

North Dakota
N.D. Cent. Code § 1259-08

Ohio
Ohio Rev. Code Ann. §
2929.20(A)(5), (N) to (S)
Ohio Rev. Code Ann. §
2967.05
Ohio Admin. Code
5120:1-1-40

Oklahoma

Exclusion
Group(s)

Any
consideration
for risk to public
Revocable?
safety?

Terminal
Illness?

Permanent or
Severe Physical
Incapacitation?

Other
Conditions?

Results in
person being
severely
debilitated or
incapacitated

Nonterminal
condition,
disease, or
syndrome that
has resulted
in physically
debilitation and/
or incapacitation

Cognitive
debilitation and
incapacitation

First-degree
murder
(including
attempt and
conspiracy)

Yes

Incurable
condition

Permanent
and irreversible
physical
incapacitation

--

Capital felonies;
Class A, B1, or
B2 felonies,
registration
offenses

Yes

Death is
"likely and
imminent"

Immediate risk to
individual's health
that requires
complex or
intensive medical
care

--

None

Not Explicitly

Y

Incurable
condition

Severe and
permanent
medical or
disability
that impairs
independent
functioning

Cognitive
disability

Sentenced to
death or life
sentence, sexual
offense, murder,
Chapter 2941
offenses

Not Explicitly

Y

--

- Dying or near
death (likely
within six
months)

Life sentence
without parole

Yes

Y

Sentenced to
life without
possibility
of release or
parole, violent
offense, sexual
offense

Yes

Likely to cause
death within
six months

Likely to cause
death within
12 months

--

57 Okla. Stat. § 332.18

In imminent
danger of death
(because of
either terminal
or non-terminal
medical issue)
with death
likely within six
months

If yes, when?
Y

Eligibility Stage

Y

Eligibility Stage

Eligibility Stage

- Medically frail
or medically
vulnerable

Oregon

--

Or. Rev. Stat. § 144.122
Or. Rev. Stat. § 144.126

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

Severe medical
condition

--

Review/Hearing
Stage

N / Not
Explicitly
Stated

55

Appendix D

Examining Prison Releases in Response to COVID

Eligibility Groups
Jurisdiction

Pennsylvania
42 Pa. Cons. Stat. Ann.
§ 9777

Rhode Island
R.I. Gen. Laws § 13-8.1-1
to 1-4

Terminal
Illness?

Permanent or
Severe Physical
Incapacitation?

Other
Conditions?

Terminally
ill, not
ambulatory,
and likely
to die in the
"near future"

Seriously ill and
not expected to
live for more than
one year

--

Diagnosis
likely to lead
to profound
decline

Incapacitation
such that the
individual
cannot function
independently
and/or requires
extensive medical
treatment

Cognitive
disability
(including
"cognitive insult"
or other mental
conditions)

Life sentence
without parole

Not Explicitly

Y

Permanent
and irreversible
physical
incapacitation
that requires
immediate
and long-term
residential care

--

Sentenced to
death or life
without the
possibility of
parole

Yes

Y

Having a
terminal
illness

Seriously ill and
not likely to
recover

Individuals
whose medical
needs are better
addressed
from a private
residence or
facility (has
additional
eligibility
criteria)

Sentenced
to death or
medically
indigent

Yes

Expected to
die within one
year

No longer able
to provide selfcare because
of physical
deterioration

No longer able
to provide selfcare because of
psychological
deterioration

Sentenced to
death

Yes

Incurable
condition

Likely to continue
indefinitely
and limit daily
functioning

Intellectual
disabilities,
mental illness,
need for long
term care

Sentenced to
death or life
without parole,
active ICE
detainees

Yes

Likely to result
in death
within 18
months

South Carolina
S.C. Code Ann. § 2421-715
S.C. Code Ann. § 2421-610
S.C. Code Ann. § 24-3-210

South Dakota
S.D Codified Laws § 2415A-55 to -68

Tennessee
Tenn. Code Ann. § 41-21227 (i)(1) to (i)(5)

Texas
Tex. Gov't Code Ann. §
508.146
37 Tex. Admin. Code §
143.34

A .

Exclusion
Group(s)

Any
consideration
for risk to public
Revocable?
safety?

"Incurable
condition"
Likely to result
in death
within two
years

Likely to result
in death
within 6
months

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

If yes, when?
None

Yes

Y

Review/Hearing
Stage

Eligibility Stage

Y

Review/Hearing
Stage

Y

Eligibility Stage

Eligibility Stage

N / Not
Explicitly
Stated

56

Examining Prison Releases in Response to COVID 		

Appendix D

Eligibility Groups
Jurisdiction

Utah

Terminal
Illness?

Permanent or
Severe Physical
Incapacitation?

Other
Conditions?

--

Medical infirmity,
disease, or
disability

Mental health
disease or
disability;
medical
condition
requiring
palliative or
nursing home
care

None

Incurable,
progressive illness
or debilitating
injury

--

None

--

--

First- or seconddegree murder,
crime involving
a youth, sexual
offense, treason,
robbery and
carjacking
offenses

Not Explicitly

Y

Individual
is unable to
engage in daily
activities without
assistance and
the care or
treatment is very
costly

--

Sentenced to
life without
parole

Yes

Y

Utah Admin. Code R671314-1(1)

Vermont
Vt. Stat. Ann. tit. 28, §
502a(d)
Vt. Stat. Ann. tit. 28, § 808

Virginia
Va. Code Ann. § 53.1-40.01

"Incurable
disease"
Likely to result
in death
within 18
months

"Chronic or
progressive
medical
condition"

--

Wash. Rev. Code §
9.94A.728(1)(c),(d)

Washington, D.C.
D.C. Code § 24-403.04
D.C. Code § 24-461 to
-468

West Virginia
W. Va. Code § 5-1-16

A .

Yes
Eligibility Stage

Yes

Physical or
-medical condition
that is not
terminal

First-degree
murder or
certain crimes
committed
while armed

Yes

No recovery
expected

"Extremely
serious medical
condition"

None

Yes

Likely to result
in death
within 120
days

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

N / Not
Explicitly
Stated

Y

Eligibility Stage

Incurable
illness or
medical
condition

28 C.F.R. §§ 2.77, 2.78

W. Va. Const. art. 7, § 11

If yes, when?

Eligibility Stage

Likely to result
in death
within 12
months

Washington

Exclusion
Group(s)

Any
consideration
for risk to public
Revocable?
safety?

--

Y

Review/Hearing
Stage

Eligibility Stage

N / Not
Explicitly
Stated

57

Appendix D

Examining Prison Releases in Response to COVID

Eligibility Groups
Jurisdiction

Wisconsin

Terminal
Illness?

Permanent or
Severe Physical
Incapacitation?

Other
Conditions?

--

Extraordinary
health condition
that may
require medical
treatment not
available without
prison

--

Wis. Stat. § 304.06

Wyoming
Wyo. Stat. Ann. § 7-13-424

A .

Terminal
illness likely to
cause death
within 12
months

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

Physical
-incapacitation
that makes
independent
functioning
impossible and/
or requiring
treatment outside
prison

Exclusion
Group(s)

Any
consideration
for risk to public
Revocable?
safety?
If yes, when?

Class A or B
felonies

Yes

Sentenced to
death or life
imprisonment
without parole

Yes

Y

Review/Hearing
Stage

Y

Review/Hearing
Stage

58

Examining Prison Releases in Response to COVID 		

Appendix D

GERIATRIC PAROLE PROVISIONS
∙ Eligibility: Geriatric parole allows individuals to apply for release based on age alone or a combination
of age and time served or age, time served, and medical conditions. The majority of jurisdictions—14—
have provisions that require age to be considered in combination with a specific period of time served,
9 jurisdictions have provisions that require the presence of a certain medical condition in combination
with an age/time served requirement, and just 3 jurisdictions allow for consideration of release based
solely on age. Four jurisdictions have both types of geriatric provisions.
» Age: Age requirements vary across jurisdictions. For example, California, alone, allows consideration
for individuals as young as 50 years old82 while four states require individuals to be at least 7083 before
they are eligible. The average age requirement is 63, while the most common age requirement is
60. A handful of states also forgo bright line requirements altogether, requiring “advanced age” or
“elderly” status in order to qualify, but not providing any specificity as to what that means.84 As noted
above, age is rarely an eligibility factor on its own,85 and a person must usually also meet time-served
or medical requirements to be considered for release.
» Time Served: Another point of distinction for geriatric parole is the amount of time that an individual
must have served prior to being eligible for release. Of the 15 jurisdictions that articulate time served
requirements, the lowest is 5 years86 and the highest is 30 years.87 Seven of the fifteen jurisdictions
require that individuals of a certain age must have served at least 10 years of their sentence, which is
the most common length for these provisions.
∙ Exclusions: The exclusionary groups for geriatric parole differed depending on whether age was paired
with medical issues or time served. When combined with medical issues, the same exclusions as in
compassionate release applied, excluding from eligibility those who had been convicted of capital
offenses, or who were sentenced to death or life without parole. When combined with time served,
the exclusionary groups were much broader. Often people convicted of crimes of violence or sexual
offenses were also excluded from consideration. Mississippi had an especially broad provision, excluding
habitual offenders, people convicted of crimes of violence, sex crimes, and even drug trafficking.88

82. Cal. Penal Code § 3055 (2022).
83. S.C. Stat. § 24-21-715(A)(2) (2022); S.D. Codified Laws § 24-15A-55 (2022); Tenn. Code. Ann §§ 40-35-501(x)(1)(A) (2022); 18 U.S.C. §
3582(c)(1) (2018).
84. See, e.g., Tex. Gov’t Code Ann. § 508.146 (2021) (elderly); Utah Admin. Code R. 671-314-1 (2022) (advancing age).
85. Age can be considered on its own in Georgia, Texas, and Utah. Ga. Const. art. IV, par. II(e); Tex. Gov’t Code Ann. § 508.146 (2021)
(elderly); Utah Admin. Code R. 671-314-1 (2022) (advancing age).
86. Va. Stat. § 53.1-40.01 (2022) (providing that people 65 and older must have served five years of their sentence in order to be
considered for release).
87. S.D. Stat. § 24-15A-55 (2022) (providing that a person who is 70 years old and has served at least 30 years of their sentence may
be considered for release).
88. Miss. Code Ann. § 47-7-3(1)(h)(iii) (2022).

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

59

Appendix D

Examining Prison Releases in Response to COVID

Table 6. Combined Compassionate/Geriatric Release Provisions by Jurisdiction
Eligibility Groups
Jurisdiction

Alabama

Age

Time Served

Life threatening
illness or chronic
debilitating
disease that leaves
the individual
unable to function
independently

Capital murder
or sex offenses

Posing a low risk
to community is
part of eligibility
determination

Y

55

--

Suffers from a
chronic infirmity,
illness, condition,
disease, or behavioral
or mental health
disorder that causes
"serious impairment"
that limits daily
functioning

Sentenced to
life without
parole, felony
crime of
violence

Yes

Y

Debilitation,
incapacitation, or
infirmity as a result
of the individual's
advanced age

Capital felonies
or murder
with special
circumstances

Yes

Colo. Rev. Stat § 17-22.5403.5
Colo. Rev. Stat § 17-1-102
Colo. Rev. Stat § 17-2-201

Connecticut

“Advanced”

--

There is a need for
long-term nursing
home care or
confinement would
endanger or shorten
life

Sentenced to
life without
parole, those
who have
yet to serve
minimum
term

Not Explicitly

Y

65

--

Suffering from a
chronic infirmity,
illness, or disease
related to aging

First-degree
murder

Yes

Y

Suffering from a
chronic infirmity,
illness, or disease
related to aging that
has progressed such
that the person is
incapacitated

Capital
felonies;
Class A, B1, or
B2 felonies,
registration
offenses

Yes

Permanently
incapacitated such
that they don't have
independent mobility

Sentenced
to life
imprisonment
without parole

Yes

Suffers from
functional
impairment, infirmity,
or illness

Serving life
without parole

Yes

N.M. Stat. Ann. § 31-21-25.1

North Carolina

65

--

N.C. Gen. Stat. § 15A-1369
to 1369.5
N.C. Gen. Stat. § 148-4

Oregon

"Elderly"

--

Or. Rev. Stat. § 144.122
Or. Rev. Stat. § 144.126

Rhode Island

65

R.I. Gen. Laws § 13-8.1-1
to 1-4

A .

(Doesn't apply
to terminal
illness)

--

Mo. Rev. Stat. § 217.250

New Mexico

Review/Hearing
Stage

"Advanced
age"

-131g; -131k

Missouri

If yes, when?

--

Ala. Code § 14-14-1 to -7

Conn. Gen. Stat. §§ 54131a to

Exclusion
Group(s)

60

Ala. Code § 15-22-41
to -43

Colorado

Medical
Requirements?

Any
consideration
for risk to public
Revocable?
safety?

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

--

Y

Eligibility Stage

Review/Hearing
Stage

Y

Eligibility Stage

Review/Hearing
Stage

N / Not
Explicitly
Stated

Y

Review/Hearing
Stage

60

Appendix D

Examining Prison Releases in Response to COVID 		

Eligibility Groups
Jurisdiction

South Carolina

Age

70

Time Served

--

Medical
Requirements?

Exclusion
Group(s)

Any
consideration
for risk to public
Revocable?
safety?
If yes, when?

Suffering from
chronic infirmity,
illness, or disease
related to aging

Sentenced
to death or
life without
possibility of
parole

Yes

At least 10
years

Care needs that at
least double the
average annual
medical cost

Sentenced
to death,
medically
indigent, Class
1 or 2 felony

Yes

15 years
of 75% of
sentence
(lower of the
two)

Condition related to
the aging process
that causes acute
vulnerability to
COVID-19

None

Yes

--

Chronic infirmity,
illness, or disease
related to aging

--

Incapacitation
because of age that
diminishes one's
ability to provide for
themselves

S.C. Code Ann. § 2421-715
S.C. Code Ann. § 2421-610

Y

Eligibility Stage

S.C. Code Ann. § 24-3-210

South Dakota

65

S.D Codified Laws § 2415A-55 to -68

Washington, D.C.

60

D.C. Code § 24-403.04
D.C. Code § 24-461 to
-468
28 C.F.R. §§ 2.77, 2.78

65

Wyoming

--

Wyo. Stat. Ann. § 7-13-424

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

Y

Review/Hearing
Stage

Y

Review/Hearing
Stage

First-degree
murder or
certain crimes
committed
while armed

Sentenced to
death or life
imprisonment
without parole

Yes

Y

Review/Hearing
Stage

61

Examining Prison Releases in Response to COVID

Appendix D

Table 7. Geriatric Parole Provisions by Jurisdiction
Eligibility Groups
Jurisdiction

Age

Time Served

Exclusion Group(s)

Any consideration
for risk to public
safety?

Revocable?

If yes, when?
California

50

Cal. Penal Code § 3055

Colorado

64

At least 20 years
of continuous
incarceration on the
current sentence

Sentenced to
death or life
without possibility
of parole or firstdegree murder of a
peace officer

Yes

At least 20 years

Violent offense or
sexual offense

Yes

None

Yes

Colo. Rev. Stat § 17-22.5-403.5
Colo. Rev. Stat § 17-1-102

Review/Hearing
Stage

N / Not
explicitly
stated

Y

Review/Hearing
Stage

Colo. Rev. Stat § 17-2-201

Federal BOP

70

At least 30 years

18 U.S.C § 3582(c)(1)(A)

Louisiana

Eligibility Stage

60

At least 10 years

La. Stat. Ann. § 15:574.4

Violent offenses or
sexual offenses

Yes
Eligibility Stage

N / Not
Explicitly
Stated
N / Not Stated
Explicitly

Those who
don't meet
programming
requirements

Maryland

60

At least 15 years

Individuals
registered as sex
offenders

Not Explicitly

N / Not Stated
Explicitly

60

At least 10 years

"Habitual
offenders," crimes
that prohibit
parole release, sex
crimes, trafficking
of controlled
substances

Not Explicitly

N / Not Stated
Explicitly

65

At least a majority of
the maximum term

Violent crimes,
Yes
crimes against
Eligibility Stage
children, sexual
offenses, vehicular
homicides,
"habitual criminals,"
sentenced to death
or life without the
possibility of parole

N / Not Stated
Explicitly

60

At least 10 years or
one-third of total term
(shorter)

Violent crimes and
sexual offenses

N / Not
Explicitly
Stated

Md. Code Ann., Crim. Law §
14-101 (f)

Mississippi
Miss. Code Ann. § 47-7-3

Nevada
Nev. Rev. Stat. § 213.12155

Oklahoma
57 Okla. Stat § 332.21

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

Yes
Eligibility Stage

62

Examining Prison Releases in Response to COVID 		

Appendix D

Eligibility Groups
Jurisdiction

Age

Time Served

Exclusion Group(s)

Any consideration
for risk to public
safety?

Revocable?

If yes, when?
South Dakota

70

At least 30 years

S.D Codified Laws § 24-15A-55
to -68

Texas

65

--

Tex. Gov’t Code Ann. § 508.146

Utah

"Advanced
age"

--

Utah Adm. Code R671-314-1

Virginia

60

At least 10 years

65

At least 5 years

Washington

"Advanced
age"

60

Washington, D.C.

Yes

Sentenced to
death or life
without parole,
active ICE
detainees

Yes

None

Yes

Eligibility Stage

Y

--

None

Not explicitly

N

At least 20 years

None

Yes

N / Not
Explicitly
Stated

Wisconsin

60

At least 10 years

Wis. Stat. § 302.113 (9g)

65

At least 5 years

ROBINA INSTITUTE

N / Not
Explicitly
Stated

Not explicitly

Review/Hearing
Stage

OF CRIMINAL LAW ANO CRIMINAL JUSTICE

N / Not
Explicitly
Stated

Class 1 Felonies

D.C. Code § 24-403.04

A .

Y

Review/Hearing
Stage

Eligibility Stage

Va. Code Ann. § 53.1-40.02

Wash. Rev. Code § 9.94A.728
(1) (d)

Sentenced to
death or medically
indigent

Class A or B
Felonies

Yes

Y

Review/Hearing
Stage

63

Appendix E.
Narrative Descriptions of COVID Prison Releases by
Jurisdiction
In this section, we include a brief narrative of the prison releases made in each jurisdiction in response
to the COVID pandemic. Fourteen states are excluded because they did not make releases: Alabama,
Alaska, Arizona, Idaho, Mississippi, Missouri, Nebraska, Nevada, New Hampshire, South Carolina, South
Dakota, Tennessee, Texas, and Wyoming.

Arkansas
Citing the Arkansas Emergency Services Act of 1973 (Ark. Code Ann. §§ 12-75-101, et seq.), Arkansas
governor Asa Hutchinson issued an executive order (EO 20-06) giving state agencies the authority to
publicly identify and suspend for thirty days:
provisions of any regulatory statute, agency order or rule that in any way prevents, hinders, or
delays the agency’s ability to render maximum assistance to the citizens of this state while they
are adhering to guidelines to prohibit the spread of disease or seeking assistance from the state
to obtain benefits or services related to health, education, employment, or any service rendered
by the state in response to or to mitigate the spread of COVID-19.89
Governor Hutchinson later extended the deadline of Executive Order 20-06 via another executive order
(EO 20-16). The Arkansas Department of Corrections (ADC) used these executive orders to suspend two
eligibility requirements for people in prison being considered for early release due to a prison overcrowding
state of emergency (these requirements can be found in A.C.A. § 12-28-604). Specifically, ADC suspended
the requirements (found in Section (b)(1)) that eligible people in prison have been incarcerated for six
months and be Class I or Class II. After those requirements were suspended, 300 people were released
from prison on May 12, 2020. Then, between May 12 and July 3, 2020, 430 more people were released. All
those released from prison had been incarcerated for nonviolent and non-sexual offenses.90

California
On March 31, 2020, the California Department of Corrections and Rehabilitation (CDCR) announced its
plan to expedite the transition to parole for certain people in prison. All who were released from prison
in this expedited process had 60 days or less left to serve on their sentences and were not serving time
for a violent crime as defined by law, a sex offense, or domestic violence. In order to expedite the process
89. Ark. Code Ann. § 12-75-101 (West 2022); Ark. Code Ann. §12-28-604 (West 2020); Ark. Exec. Order No. 20-06 (Mar. 17, 2020),
https://governor.arkansas.gov/images/uploads/executiveOrders/EO_20-06._.pdf
90. Ark. Code Ann. § 12-75-101 (West 2022); Ark. Code Ann. §12-28-604 (West 2020); Ark. Exec. Order No. 20-06 (Mar. 17, 2020),
https://governor.arkansas.gov/images/uploads/executiveOrders/EO_20-06._.pdf; Ark. Exec. Order No. 20-16 (Apr. 15, 2020),
https://governor.arkansas.gov/images/uploads/executiveOrders/EO_20-16._.pdf; John Moritz, Arkansas prison system below
capacity for first time since ‘07, Ark. Democrat Gazette (July 3, 2020), https://www.arkansasonline.com/news/2020/jul/03/
states-prison-system-below-capacity/?news-arkansas; Public Health Rule Suspension Notice, Arkansas Department of
Corrections, (Apr. 22, 2020), https://doc.arkansas.gov/wp-content/uploads/2020/09/Public_Health_Rule_Suspension_-_EO2016_20-06_Notice.pdf; Ninette Sosa, ADC: 1,243 inmates considered for early release due to COVID, KNWA Fox24 (Apr. 29, 2020),
https://www.nwahomepage.com/lifestyle/health/coronavirus/adc-1243-inmates-considered-for-early-release-due-to-covid/;
Ninette Sosa, Approx. 800 inmates approved for early release; COVID-19 precaution, KNWA Fox24 (May 14, 2020), https://
www.nwahomepage.com/lifestyle/health/coronavirus/approx-800-inmates-approved-for-early-release-covid-19-precaution/.

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

64

Examining Prison Releases in Response to COVID 		

Appendix E

of release, CDCR created on-site strike teams and allowed staff to work overtime in order to accomplish
all the tasks necessary to implement the releases. Between CDCR’s announcement on March 31 and the
end of the day on April 12, 2020, 3,418 people were released from prison.91
On July 10, 2020, CDCR announced a new set of cohort releases. These releases were divided into four
cohorts: the 180-day cohort; the 365-day, under 30 cohort; the 365-day, over 30 cohort; and the high-risk
medical cohort. The 180-day cohort consisted of people who had 180 days or less left to serve on their
sentences, were not serving time for violent crimes, were not required to register as sex offenders, and
did not have assessment scores that indicated high risks for violence. This cohort was screened and
released on a rolling basis. Both of the 365-day cohorts consisted of people that were in correctional
institutions with large populations of high-risk patients, had 365 days or less to serve on their sentences,
were not serving time for violent crimes, were not required to register as sex offenders, and did not have
assessment scores that indicated high risks for violence. Members of the 365-day, over 30 cohort were
immediately eligible for release. Members of the 365-day, under 30 cohort were reviewed on a caseby-case basis for release. To be eligible for the high-risk medical cohort, individuals had to be deemed
high risk for COVID-19 complications by the California Correctional Health Care Services and have an
assessment indicating a low risk for violence; they also could not be serving a sentence of life without
parole, be condemned to death, or be a high-risk sex offender. Between July 1, 2020, and November
25, 2020, CDCR released 7,596 people from prison through the cohort system, the majority of which
belonged to the 180-day cohort.92

Colorado
On March 25, 2020, Colorado Governor Jared Polis issued an executive order (D 2020 016) suspending
regulatory statutes concerning criminal justice. These statutes included caps and criteria on awards of
earned time credits, criteria for release to Special Needs Parole, and the requirement of successfully
completing a regimented discipline program before the Department of Corrections has the authority
to establish and directly operate an intensive supervision program. Then on April 23, 2020, Governor
Polis issued a second executive order (D 2020 043) amending and extending D 2020 016; this order
suspended the standards and criteria for intensive supervision programs and the notice requirement for
placement of a non-paroled offender in an intensive supervision program. As of the end of April 2020,
almost 150 people had been released from prison through this executive order. There was an additional
reduction in Colorado’s incarcerated population of approximately 3,350 as of that April, likely caused by a

CDCR Announces Plan to Further Protect Staff and Inmates from the Spread of COVID‑19 in State Prisons, Cal. Dep’t of
Corr. and Rehabilitation (Mar. 31, 2020), https://www.cdcr.ca.gov/news/2020/03/31/cdcr-announces-plan-to-further-protectstaff-and-inmates-from-the-spread-of-covid-19-in-state-prisons/; Decl. Gipson Supp. Defs.’ Opp’n Pl.’s Emergency Mot. Re
Prevention & Management COVID-19, No. 01-cv-01351-JST (N.D. Cal. Apr. 13, 2020), https://storage.courtlistener.com/recap/gov.
uscourts.cand.76/gov.uscourts.cand.76.3275.0.pdf; Frequently Asked Questions for Plan on Expedited Release and Increased
Physical Space within State Prisons, Cal. Dep’t of Corr. and Rehabilitation, https://www.cdcr.ca.gov/covid19/frequently-askedquestions-for-plan-on-expedited-release-and-increased-physical-space-within-state-prisons/ (last visited Mar. 27, 2022).
92. Cal. Dep’t of Corr. and Rehabilitation, https://www.cdcr.ca.gov/covid19/covid-19-response-efforts/#R (last visited Mar. 27, 2022);
CDCR Announces Additional Actions to Reduce Population and Maximize Space Systemwide to Address COVID-19,
Cal. Dep’t of Corr. and Rehabilitation (July 10, 2020), https://www.cdcr.ca.gov/news/2020/07/10/cdcr-announces-additionalactions-to-reduce-population-and-maximize-space-systemwide-to-address-covid-19/; CDCR Div. of Corr. Pol’y Rsch. and
Internal Oversight Off. of Rsch, Release Occurrences from CDCR's In Custody Population Released Early Due to COVID 19
Between July 01, 2020 and November 25, 2020 Version 24, CalMatters (Nov. 26, 2020), https://calmatters.org/wp-content/
uploads/2020/12/Expedited-release-demographics-to-Nov.-25.pdf; Robert Lewis, California’s Post-Prison Chaos: Thousands
Released Early, Including Many With Coronavirus, CapRadio (Aug. 12, 2020), https://www.capradio.org/articles/2020/08/12/
californias-post-prison-chaos-thousands-released-early-including-many-with-coronavirus/; Byrhonda Lyons, High-Risk
Inmates Aren’t Prioritized in State’s Early Releases, CapRadio (Dec. 13, 2020), https://www.capradio.org/articles/2020/12/13/
high-risk-inmates-arent-prioritized-in-states-early-releases/.
91.

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

65

Examining Prison Releases in Response to COVID

Appendix E

moratorium on transfers from county jails, regular attrition and paroles being granted. As of September
2020, 160 more people (for a total of 310) were released thanks to these executive orders.93

Connecticut
On March 24, 2020, the State of Connecticut Department of Correction (DOC) issued a press release
announcing its response to the COVID-19 pandemic. Part of that response involved having the Community
Release Unit, Parole and Community Services Division, and The Board of Pardons and Paroles work
collaboratively to use discretionary release for people in prison who were considered low risk, had stable
home plans, were 60 years old or older, and had health conditions which made them more susceptible
to catching the virus. In practice, the DOC prioritized releasing people over 50 years old with medical
conditions making them susceptible to COVID-19. In April, the DOC Commissioner signed an exception
for medical furlough allowing people to be released that had completed at least 40% of a maximum
two-year prison sentence. On May 6, 2020, DOC issued a press release announcing that since March 1,
2020, the prison population had dropped by 1,609 people. Then on June 2, 2020, DOC issued a similar
press release announcing that the prison population had dropped by 2,000 people since March 1, 2020.
However, this estimate, which was wholly attributed to COVID releases, failed to take into account the
ordinary release rate of the parole board. To develop a more accurate estimate, we compared releases
in 2019 to releases in 2020. In March, April, and May, there were 1,270 discretionary releases. During that
same period in 2020, there were 1,670 discretionary releases. Therefore, we estimate that there were an
additional 357 releases in 2020 in response to the COVID-19 pandemic.94

Federal BOP
At the federal level, COVID releases were triggered through multiple mechanisms, including at the
direction of the Attorney General, court orders, and compassionate release provisions. On March 26,
2020, United States Attorney General William Barr (AG Barr) issued a memorandum directing the BOP
to prioritize the use of home confinement as a tool for combatting the COVID-19 pandemic and the
dangers it posed to vulnerable individuals. The memorandum specified that the BOP should prioritize
people in prison who were vulnerable to COVID-19 as defined by the Centers for Disease Control and
Prevention, those in lower security facilities, those with good behavior while in prison, those scoring low
on the PATTERN risk assessment, those with verifiable re-entry plans, and those with less serious offenses.
On April 3, 2020, AG Barr issued a second memorandum which instructed the BOP to prioritize the

93. Colo. Exec. Order No. D 2020 016 (Mar. 25, 2020), https://www.colorado.gov/governor/sites/default/files/inline-files/D%20
2020%20016%20Suspending%20Certain%20Regulatory%20Statutes%20Concerning%20Criminal%20Justice_0.pdf;
Colo. Exec. Order No. D 2020 043 (Apr. 23, 2020), https://www.colorado.gov/governor/sites/default/files/inline-files/D%20
2020%20043%20Extending%20016.pdf; Ali Budner, Hundreds Of Colorado Inmates Have Been Released Early Because Of
Coronavirus. This Is One Man’s Story, Colo. Pub. Radio (Sept. 14, 2020) https://www.cpr.org/2020/09/14/colorado-coronavirusupdate-prison-inmate-early-releases/; LJ Dawson, ACLU pressures Polis to release more inmates at risk for COVID-19, Colo.
Springs Indy (Sept. 9, 2020), https://www.csindy.com/news/aclu-pressures-polis-to-release-more-inmates-at-risk-for-covid-19/
article_dcbe5b2a-f228-11ea-acb6-573f7c4885c9.html; Allison Sherry, Colorado Corrections Agrees To Release Some Inmates
Early To Reduce Prison Populations, Colo. Pub. Radio (Apr. 23, 2020) https://www.cpr.org/2020/04/23/colorado-correctionsagrees-to-release-some-inmates-early-to-reduce-prison-populations/?eType=EmailBlastContent&eId=07a90610-fdf3-4e1b994a-712a02c5065c.
94. DOC COVID-19 Release Facts, Conn. State Dep’t of Corr. (May 5, 2020), https://portal.ct.gov/-/media/DOC/Pdf/Coronavirus-3-20/
RELEASE-FACT-SHEET-covid-19-050520.pdf?la=en; Rich Kirby, CT Prison Population Down 16 Percent: Report, Patch
(June 3, 2020), https://patch.com/connecticut/across-ct/ct-prison-population-down-16-percent-report; Press Release,
Conn. State Dep’t of Corr., CT Prison Population Down 2,000 since March 1st (June 2, 2020), https://portal.ct.gov/-/media/
DOC/Pdf/Coronavirus-3-20/Press-Release-re-Populations-Drops-by-2000--060220.pdf; Press Release, Conn. State Dep’t of
Corr., The Department of Correction continues essential operations while managing reentry planning and routine offender
releases without disruption (March 24, 2020), https://portal.ct.gov/-/media/DOC/Pdf/Coronavirus-3-20/Press-Release-DOCreentry-032420.pdf; Press Release, Conn. State Dep’t of Corr., Prison Population Drops by 1,609 people since March 1st (May 6,
2020), https://portal.ct.gov/-/media/DOC/Pdf/Coronavirus-3-20/Press-Release-DOC-Release-Data-050620.pdf.

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

66

Examining Prison Releases in Response to COVID 		

Appendix E

facilities most impacted by COVID-19 in reviewing people for release from prison to home confinement.
The memorandum also expanded the eligibility for home confinement: rather than only considering
the those who met the ordinary criteria for home confinement, AG Barr directed the BOP to consider
all people in prison considered at risk for COVID-19 and to transfer them even if electronic monitoring
was not available, as long as such a move was consistent with maintaining public safety. Between AG
Barr’s first memorandum (March 26, 2020) and January 2, 2022, the BOP released 36,367 people to home
confinement, including those individuals mentioned in the following paragraph who were ordered to be
released by the courts.95
On April 13, 2020, the ACLU of Ohio and the Ohio Justice and Policy Center brought a class action habeas
petition seeking modification of the sentences of medically vulnerable people in prison in Elkton Federal
Correctional Institution (a low security federal prison). On May 19, 2020, a federal judge of the United
States District Court for the Northern District of Ohio ordered the Bureau of Prisons to expedite the
release of 837 of those medically vulnerable people through home confinement and compassionate
release.96 In September 2020, the Bureau of Prisons (BOP) settled a case with a class of people in federal
prison; a provision of the Settlement Agreement required that the BOP "endeavor to release individuals
approved for home confinement to home confinement within 14 days of the approval decision" unless
public safety or the individual's home situation counseled against such release. On December 11, 2020,
Judge Michael Shea of the District Court for the District of Connecticut found that federal administrators
in Danbury Federal Correctional Institution were not acting quickly to release eligible individuals.
Seventeen people who had previously been cleared for release to home confinement had been waiting
substantially longer than 14 days to be released, according to Judge Shea. Thus, Judge Shea ordered that
those 17 people be released to home confinement by 5 p.m. on December 12, 2020.97
Under the First Step Act, signed into law in 2018, federal courts may reduce a person’s prison sentence if
they find “extraordinary and compelling reasons” to do so. Some courts have found that the pandemic
provided such reason and released vulnerable people to protect them from the virus. As of June 11, 2021,
3,221 people in federal prison had been granted compassionate release by judges since the beginning
of the pandemic.98

95. William Barr, Memorandum for Director of Bureau Prisons: Increasing Use of Home Confinement at Institutions Most Affected by
COVID-19, Off. of the Att’y Gen. (Apr. 3, 2020), https://www.bop.gov/coronavirus/docs/bop_memo_home_confinement_april3.pdf;
William Barr, Memorandum for Director of Bureau Prisons: Prioritization of Home Confinement as Appropriate in Response
to COVID-19 Pandemic, Off. of the Att’y Gen. (Mar. 26, 2020), https://www.bop.gov/coronavirus/docs/bop_memo_home_
confinement.pdf; Frequently Asked Questions regarding potential inmate home confinement in response to the COVID-19
pandemic, Fed. Bureau of Prisons, https://www.bop.gov/coronavirus/faq.jsp (last visited Mar. 31, 2020).
96. A Federal Judge Issues Order To Enforce Compliance, Requiring Elkton Prison Officials To Expedite Transfer & Release
Of Medically Vulnerable Subclass Through Home Confinement And Compassionate Release, ACLU Ohio (May 19, 2020),
https://www.acluohio.org/en/press-releases/federal-judge-issues-order-enforce-compliance-requiring-elkton-prison-officials;
Order on Mot. to Enforce. No. 4:20-cv-00794-JG (N.D. Ohio May 19, 2020), https://www.acluohio.org/sites/default/files/Orderon-motion-to-enforce.pdf.
97. Whitted v. Easter, No. 3:20-cv-00569 (MPS), 2020 U.S. Dist. LEXIS 232843 (D. Conn. Dec. 11, 2020), https://plus.lexis.com/api/
permalink/40a753f8-11da-440f-9be5-aa4b302b72c9/?context=1530671; Edmund Mahony, Judge says Danbury prison is slow
to enforce COVID-19 agreement and orders immediate release of 17 inmates, Hartford Courant (Dec. 11, 2020), https://www.
courant.com/coronavirus/hc-news-coronavirus-judge-releases-inmates-20201211-20201211-utrrx4536zekrltjuv3bhfkgsi-story.html.
98. Keri Blakinger & Joseph Neff, 31,000 Prisoners Sought Compassionate Release During COVID-19. The Bureau of Prisons
Approved 36, The Marshall Project (June 11, 2021), https://www.themarshallproject.org/2021/06/11/31-000-prisoners-soughtcompassionate-release-during-covid-19-the-bureau-of-prisons-approved-36; Greg Newburn, COVID-19’s toll on Florida
prisons highlights the need for compassionate release, Fla. Pol. (Sept. 2, 2020), https://floridapolitics.com/archives/364267greg-newburn-covid-19s-toll-on-florida-prisons-highlights-the-need-for-compassionate-release/.

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

67

Examining Prison Releases in Response to COVID

Appendix E

Florida
As of May 8, 2020, the Florida prison system’s only COVID-19-related releases were three people who had
been released from prison on conditional medical release. Florida’s conditional medical release program
gives those who are permanently incapacitated or terminally ill a chance to be released. Governor Ron
DeSantis repeatedly made clear that he did not view releasing people from prison as a viable option for
addressing the pandemic and Florida abolished parole in 1983. In August of 2020, advocates were still
asking for some sort of meaningful response to the spread of COVID-19 in Florida prisons.99

Georgia
On March 31, 2020, the Georgia State Board of Pardons and Paroles issued a news release announcing
that it had begun reviewing cases for clemency release in response to the COVID-19 pandemic. The
Board stated that it was considering people who were serving time for nonviolent offenses and were
within 180 days of completing their prison sentences. According to one report, during April and May of
2020 the Board released 2,550 people from prison, including granting early releases for 918 people who
were within 180 days of the end of their sentences. This is more than a 150% increase in releases from
the roughly 850 people the Board releases in an average month. After May 2020 the Board returned to
its average release rate.100

Illinois
On March 23, 2020, Illinois Governor J. B. Pritzker issued an executive order (EO 2020-11). Among other
COVID-19 measures, EO 2020-11 temporarily suspended the requirement that the Department of
Corrections (DOC) provide no less than 14 days’ notice to the State Attorney if a person in prison receives
an earlier release date because of sentence credits earned for good conduct. Thanks to EO 2020-11, by
March 27, 2020, at least six people were granted good conduct time and released from the Decatur
Correctional Center’s special wing that houses mothers and their newborn babies.101
According to the Illinois Northern District Court, 644 people were released from the Illinois DOC’s
custody between March 2, 2020, and April 10, 2020, through various efforts. This likely includes the six
previously mentioned mothers with newborns but also includes those released through various other
methods. For example, Governor Pritzker reviewed and granted at least two commutation petitions
during that time. The DOC also created a population management task force to prioritize the review
of individuals for possible release. Every day between March 2 and April 10, 2020, the DOC identified
and reviewed people in prison within nine months of their release date to determine whether they
were eligible for early release. The DOC also continued to place people on electronic monitoring or
99. Florida COVID-19 FAQ, Families Against Mandatory Minimums (2020), https://famm.org/wp-content/uploads/Florida-COVIDFAQ.pdf; Samantha Gross, Florida prisons boss can’t release inmates amid COVID-19. But can he furlough them?, Mia.
Herald (May 8, 2020) https://www.miamiherald.com/news/special-reports/florida-prisons/article242573521.html#storylink=cpy;
Grace Toohey, Advocates call for steps to release some Florida prisoners as COVID-19 spreads, Orlando Sentinel (Aug. 14,
2020) https://www.orlandosentinel.com/coronavirus/os-ne-coronavirus-lawmakers-concern-prison-cases-deaths-20200814ctn7bdmkwfcljpnpu3pghpg22a-story.html.
100. Stanley Dunlap, COVID Races Through Some Georgia Prisons, Sickening Staff and Inmates, Ga. Pub. Broad. (Sept. 29,
2020), https://www.gpb.org/news/2020/09/29/covid-races-through-some-georgia-prisons-sickening-staff-and-inmates; Press
Release, State Bd. of Pardons and Paroles, Board Considering Releases to Address COVID-19 in Georgia Prisons (March 31,
2020),
https://pap.georgia.gov/press-releases/2020-03-31/board-considering-releases-address-covid-19-georgia-prisons;
Joshua Sharpe, Georgia to release some inmates due to COVID-19 fears, The Atlanta Journal-Constitution (Mar. 31, 2020),
https://www.ajc.com/news/local/breaking-georgia-release-some-inmates-due-covid-fears/np6zhBrlP1oe2jOkUmWVoL/.
101. Ill. Exec. Order No. 2020-11 (Mar. 23, 2020), https://www2.illinois.gov/sites/coronavirus/Resources/Pages/ExecutiveOrder2020-11.
aspx; Annie Sweeney, Facing growing coronavirus threat, Illinois prison officials release moms jailed with their babies: ‘Oh
my goodness, there was no words’, Chi. Trib. (Mar. 27, 2020), https://www.chicagotribune.com/coronavirus/ct-coronaviruswoman-babies-released-prison-20200327-t6rfew4m6jbuxmw4lrw5v47dfi-story.html.

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

68

Examining Prison Releases in Response to COVID 		

Appendix E

home detention, concentrating on those 55 years or older who had served at least a fourth of their
sentence and were within 12 months of release. About 450 people were released from prison through
various forms of sentence credit, restoration of credit, and electronic detention. Finally, Governor Pritzker
issued an executive order (EO 2020-21) on April 6, 2020, which suspended the 14-day limit for medical
furloughs and allowed furloughs for medical purposes at the discretion of the Director of the DOC. Sixtyfive furloughs were granted between March 2 and April 10, 2020.102

Indiana
According to reports, between March and June of 2020, Indiana’s prison population dropped by 1,015
(from 26,891 to 25,876). This does not necessarily reflect an intentional effort to decrease the prison
population, though: fewer people were entering Indiana prisons because of the way the pandemic
disrupted the court system, while people in prison were still being released as their sentences ended.
In fact, Governor Eric Holcomb took no action to reduce the prison population, stating, “I do not believe
in releasing those low-level offenders.” Instead, the governor left it to local courts to decide whether
anybody should be let out early, encouraging county officials to take steps to reduce their county jail
populations (and there are reports that this did happen, to some extent). However, 27 people did receive
COVID-19-related sentence modifications from March through May 2020, and it is likely that at least
some of those modifications resulted in those individuals being released early.103

Iowa
As of April 22, 2020, the Iowa Board of Parole had approved 572 people for early release from the state’s
prisons in order to reduce overcrowding during the COVID-19 pandemic. While the Board did not list
any certain targets for release, the director of the Iowa Department of Corrections stated that the Board
has the authority to release those who would likely succeed in a community setting (e.g., have a plan for
safe housing, low risk of reoffense, etc.). At some point shortly thereafter, the Board also implemented
a double-panel approach, in which two three-person panels simultaneously reviewed individuals for
parole. They also reconsidered those who had previously been recommended for, but denied, parole.
If additional people were released from prison due to COVID-19, the exact number does not appear to
have been tracked. However, 4,724 people were paroled in fiscal year 2020, compared to 4,527 in fiscal
year 2019, so we estimate that at least 197 more people were paroled due to COVID.104

Kansas
In early April 2020, Kansas Governor Laura Kelly announced that officials were reviewing a list of people
in prison with short amounts of time left on their sentences and “viable plans” for reentry. More than 500
cases were reviewed through this process, but the Department of Corrections only released 6 people
from prison to serve the rest of their sentences on house arrest. After that, on May 1, 2020, Governor Kelly
announced that the releases would stop. She cited an outbreak in one correctional facility (none of the
102. Ill. Exec. Order No. 2020-21 (Apr. 6, 2020), https://www2.illinois.gov/Pages/Executive-Orders/ExecutiveOrder2020-21.aspx;
Money v. Pritzker, 453 F. Supp. 3d 1103 (N.D. Ill. 2020).
103. Jake Harper, Indiana Left It To County Courts To Release Prisoners During The COVID Crisis. Most Of Them Haven’t, WFPL
News (July 18, 2020) https://wfpl.org/indiana-left-it-to-county-courts-to-release-prisoners-during-the-covid-crisis-mostof-them-havent/; Prison health needs greater priority in Indiana, The Republic (July 17, 2020), http://www.therepublic.
com/2020/07/17/prison_health_needs_greater_priority_in_indiana/.
104. O. Kay Henderson, Hundreds of Iowa prisoners getting early parole due to COVID-19 concerns, KMAland (Apr. 22, 2020),
https:/www.kmaland.com/news/hundreds-of-iowa-prisoners-getting-early-parole-due-to-covid-19-concerns/article_
e508b620-83fd-11ea-b6d6-738156a1cd88.html?eType=EmailBlastContent&eId=07a90610-fdf3-4e1b-994a-712a02c5065c; Erin
Jordan, Iowa parole board releases fewer offenders in fiscal 2020, despite COVID-19, The Gazette (Feb. 8, 2021), https://www.
thegazette.com/crime-courts/iowa-parole-board-releases-fewer-offenders-in-fiscal-2020-despite-covid-19/.

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

69

Examining Prison Releases in Response to COVID

Appendix E

six released individuals had been incarcerated in that specific facility) as the reason--officials did not
want to risk spreading the virus by releasing people infected with COVID-19 into the community. There
are no reports of additional releases thereafter.105

Kentucky
On April 2, 2020, Kentucky Governor Andy Beshear announced an executive order (EO 2020-267), which
commuted the sentences of 186 people in jails and prison, all of whom were identified as being medically
vulnerable to COVID-19 and serving time for non-violent, non-sexual offenses. The order does not specify
how many were released from prisons versus jails; however, since 10% of those released in later executive
orders were released from prison, we estimate that a similar percentage, or 19 people, may have been
released from prison as a result of this executive order. The next executive order was issued on April 10,
2020, when Governor Beshear commuted the sentences of 697 people, of which 33 people were serving
time in prison facilities. Next, Governor Beshear issued an executive order (EO 2020-293) on April 24,
2020, which commuted the sentences of 352 people, 15 of whom were in prison. The order targeted
people in jails and prisons who were vulnerable to COVID-19 because of age or medical conditions who
were serving sentences for non-violent, non-sexual offenses and had 5 or fewer years remaining on their
sentences. Finally, on August 25, 2020, Governor Beshear issued another executive order (EO 2020-699),
which commuted an additional 646 sentences, of which 123 were people serving time in prison. The
order targeted people who were serving sentences for non-violent, non-sexual offenses and were either
medically vulnerable or were nearing the end of their sentences. Each executive order required people
who were released to have a verified housing plan.106

Louisiana
In April of 2020, the Louisiana Department of Corrections (DOC) set up a review panel to consider medical
furlough for people in prison because of the COVID-19 pandemic. The panel consisted of six members:
representatives from the DOC, Louisiana Division of Probation and Parole, state Board of Pardons and
Parole, Louisiana Sheriff’s Association, and Louisiana District Attorneys Association, as well as a victim’s
advocate appointed by the governor. An individual needed five out of these six members’ votes in order
to be released. In order to even be considered for release, the person needed to have an underlying
health condition, be serving time for a non-violent and non-sexual offense, have a release date within six
months, and have proof of a post-release housing plan. Only 68 people were released from prison by the
panel before it was suspended in June.107

105. John Hanna & Heather Hollingsworth, Coronavirus Outbreak Prompts Kansas to Stop Prison Releases, U.S. News (May 1,
2020), https://www.usnews.com/news/best-states/kansas/articles/2020-05-01/coronavirus-outbreak-prompts-kansas-to-stopprison-releases; Alice Speri, Prison Officials in Kansas Ignored the Pandemic. Then People Started Dying, The Intercept (July
2, 2020), https://theintercept.com/2020/07/02/coronavirus-kansas-prison-lansing-correctional/.
106. Ky. Exec. Order. No. 2020-267 (Apr. 2, 2020), https://governor.ky.gov/attachments/20200402_Executive-Order_2020-267_
Conditional-Commutation-of-Sentence.pdf; Ky. Exec. Order. No. 2020-293 (Apr. 24, 2020), https://governor.ky.gov/
attachments/20200424_Executive-Order_2020-293_Conditional-Commutation.pdf; Ky. Exec. Order. No. 2020-278 (May
15,
2020),
https://russellvilleky.org/index.php/covid-19-information/governor-executive-orders/588-governor-executiveorder-2020-278; Ky. Exec. Order. No. 2020-699 (Aug. 25, 2020), https://governor.ky.gov/attachments/20200825_ExecutiveOrder_2020-699_Commutations.pdf; Kentucky's Response to COVID-19, Kentucky.gov (Oct. 20, 2020), https://governor.ky.gov/
Documents/20201020_COVID-19_page-archive.pdf; Brian Planalp, Nearly 1,000 Kentucky prison sentences to be commuted,
Beshear says, FOX19 (Apr. 2, 2020), https://www.fox19.com/2020/04/02/watch-live-gov-beshear-provides-update-covidkentucky/; Chris Williams, Lawmakers question DOC officials on Governor Beshear’s COVID-19 commutations, WHAS11
(Dec. 3, 2020), https://www.whas11.com/article/news/kentucky/kentucky-governor-prisoners-released-covid-19-lawmakersface-off-corrections-officials/417-84c107ff-50e3-452d-b050-31a7d3b5db05.
107. Julie O’Donoghue, Louisiana Prisons Need To Do More COVID-19 Releases, Advocates Say, La. Illuminator (Sept. 3, 2020),
https://lailluminator.com/2020/09/03/louisiana-prisons-need-to-do-more-covid-19-releases-advocates-say/; Lea Skene, Review
panel to consider medical release for some Louisiana state prison inmates due to coronavirus, The Advoc. (Apr. 14, 2020),
https://www.theadvocate.com/baton_rouge/news/coronavirus/article_62e9f822-7e79-11ea-bfb3-933881495eb6.html.

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

70

Examining Prison Releases in Response to COVID 		

Appendix E

Maine
Because of the COVID-19 pandemic, the Maine Department of Corrections (DOC) began reviewing
candidates for release from prison to home confinement more quickly than usual by increasing the
number of employees who were reviewing cases. More people began finishing their sentences that way,
but the criteria for home confinement were not expanded in response to the pandemic. In order to be
eligible for home confinement in Maine, individuals had to be classified as minimum security, be within
18 months of their release date, and have finished at least half of their sentence. During the pandemic,
the DOC actually began using an even stricter set of criteria than usual in order to prioritize cases. A
person needed to have a plan for their housing and well-being; must not have committed a crime
against a person; and must have been within a year of their release date (though those with a release
date within 18 months were still considered). Other factors that affected the decision-making process
were the person’s medical history and past compliance with probation conditions. As of April 21, 2020,
the DOC had reported releasing more than 60 people from prison to home confinement; as of May 3,
2020, that number had risen to 73. Finally, as of May 27, 2020, 95 people had been released from prison
through the program; 5 had returned to prison for violating rules and 20 people were awaiting release
through the program.108

Maryland
In response to the COVID-19 pandemic, Maryland Governor Lawrence J. Hogan, Jr. issued two virtually
identical executive orders (7 months apart) to facilitate the release of people from prison. Each order
gave the Maryland Division of Correction the power to accelerate the releases of people within four
months of their release date via diminution credits; prompted immediate consideration of release to
home detention for some; and accelerated consideration of parole for those 60 years old and older
(not convicted of a violent crime and with good behavior and a home plan). No one convicted of sexual
offenses were eligible for release through any of these three paths. The first order, issued on April 18th,
2020, was prompted by demands from a coalition of advocates led by the ACLU. The order was expected
to result in the release about 700 people who were within four months of their release date and 100 who
might be eligible for expedited parole. The second order, issued on November 17th, 2020, was expected
to make 1200 more people eligible for early release. There are no reports of how many of those people
were actually released as a result of these orders.109

108. Callie Ferguson, He pleaded to leave Maine prison early amid pandemic. The answer was no, Bangor Daily News (June
8, 2020), https://bangordailynews.com/2020/06/08/mainefocus/an-inmate-pleads-to-leave-early-but-maine-prisons-arentletting-out-as-many-as-some-would-like/; Megan Gray, Maine prisons pressured to release more inmates, and information,
during pandemic, Portland Press Herald (May 3, 2020), https://www.pressherald.com/2020/05/03/maine-prisons-pressuredto-release-more-inmates-and-more-information-during-pandemic/?rel=related; Susan Sharon, Maine Released Dozens
Of Prisoners To Prevent COVID-19 Spread. But Advocates Say More Should Be Done, Me. Public (April 21, 2020), https://
www.mainepublic.org/courts-and-crime/2020-04-21/maine-released-dozens-of-prisoners-to-prevent-covid-19-spread-butadvocates-say-more-should-be-done.
109. Md. Exec. Order No. 20-04-18-01 (Apr. 18, 2020), https://governor.maryland.gov/wp-content/uploads/2020/04/PrisonerRelease-4.18.20.pdf; Md. Exec. Order No. 20-11-17-03 (Nov. 17, 2020), https://governor.maryland.gov/wp-content/
uploads/2020/11/Prisoner-Release-RENEWAL-11.17.20.pdf; Luke Broadwater, With coronavirus spreading, Maryland
Gov. Hogan signs order for expedited release of hundreds of prisoners, The Balt. Sun (Apr. 19, 2020), https://www.
baltimoresun.com/coronavirus/bs-md-pol-hogan-prisoners-20200419-7mzvooaoxfbyngowb2xdeucrme-story.html?utm_
source=The+Marshall+Project+Newsletter&utm_campaign=f1efaf562f-EMAIL_CAMPAIGN_2020_04_20_11_35&utm_
medium=email&utm_term=0_5e02cdad9d-f1efaf562f-166145513; Danielle Gaines, Hogan Issues Order to Guide Speedier
Inmate Releases During COVID-19 Outbreak, Md. Matters (Apr. 19, 2020), https://www.marylandmatters.org/2020/04/19/
hogan-issues-order-to-guide-speedier-inmate-releases-during-covid-19-outbreak/; Ovetta Wiggins, Md. governor signs
executive order to allow early release of prisoners to slow the spread of coronavirus, The Wash. Post (Nov. 18, 2020), https://
www.washingtonpost.com/local/md-politics/maryland-prisoner-release-coronavirus-/2020/11/18/136cc1f8-29dc-11eb-9b14ad872157ebc9_story.html.

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

71

Examining Prison Releases in Response to COVID

Appendix E

Massachusetts
Towards the beginning of the COVID-19 pandemic, three Massachusetts organizations (the Committee
for Public Counsel Services, the Massachusetts Association of Criminal Defense Lawyers, and the ACLU
of Massachusetts) initiated a lawsuit, seeking the release of various people in prison across the state. On
April 3, 2020, the Supreme Judicial Court issued its opinion, which created a presumption of eligibility
for release for certain individuals (people held on bail for certain offenses and people being held for
a probation violation hearing) and required all correctional facilities and prisons to report, among
other things, the number of people released daily. The ACLU of Massachusetts tracked these reports
and recorded, from March 27, 2020, through August 15, 2021, that Massachusetts released 36 people,
approved 47 for medical parole, and released another 1,055 on parole. Several releases were those held
for technical violations of parole pursuant to the earlier court order. Otherwise, it is unclear whether
these releases were directly related to COVID-19. The United States Attorney’s Office also has a page on
which it tracks COVID-19 related releases of people in prison in the District of Massachusetts. According
to that page, 54 people (who had already been sentenced) either were resentenced to time served or
had their motion to release granted as of August 30, 2021 (62 more motions were pending).110

Michigan
Between March 20 and June 5, 2020, Michigan’s prison population decreased by 1,958. About half of
that decrease, it is estimated, was caused by a decline in intake from county jails and courts; the other
half is attributed to fewer people on parole returning to prison for technical violations and accelerated
parole reviews. Although the Michigan Department of Corrections (MDOC) affirmed it had no authority
to release people before their earliest release date, MDOC did state that the parole board was working to
expedite paroles for eligible people in prison. For example, MDOC was requesting that prosecutors sign
waivers allowing immediate release, which removed the 28-day waiting period after parole decisions.
Nonviolent people who were 60 years or older and had health issues were prioritized for parole
consideration. According to the MDOC legislative liaison, Kyle Kaminski, from March 2020 through May
2020 an additional 500 paroles were approved, as compared to the same period during 2019.111

Minnesota
The Minnesota Department of Corrections (DOC) implemented three programs aimed at reducing the
prison population because of the COVID-19 pandemic. The first program, which ran from April 16 through
August 24, 2020, was the COVID-19 Conditional Medical Release Program. In order to be granted release
under this program, a person’s application was first reviewed to determine medical eligibility using health
risk assessments developed based on guidance from the Center for Disease Control and Prevention and
the Minnesota Department of Health. Then a public safety assessment was conducted; and finally, the
DOC Commissioner made the final decision in all applications that made it through the first two steps.
110. Deborah Becker, More Than 600 Massachusetts Prisoners Released Amid Pandemic, WBUR (Apr. 22, 2020), https://www.
wbur.org/news/2020/04/22/mass-prisoners-released-coronavirus; Comm. for Pub. Counsel Services v. C.J. of Tr. Ct., 142 N.E.3d
525 (Mass. 2020); COVID-19 Related Prisoner Releases, The U.S. Att’y Off. Dist. of Mass., https://www.justice.gov/usao-ma/covid19-related-prisoner-releases (last visited Mar. 29, 2022); CPCS v. Chief Justice of the Trial Court, Comm. for Pub. Couns. Services
(Apr. 3, 2020), https://www.publiccounsel.net/cpcs-v-chief-justice-of-the-trial-court/; Tracking COVID-19 in Massachusetts
Prison and Jail, ACLU Mass., https://data.aclum.org/sjc-12926-tracker/ (last visited Mar. 29, 2022).
111. Paul Egan, Michigan prison population sees record drop during coronavirus pandemic, Detroit Free Press (June 9, 2020)
https://www.freep.com/story/news/local/michigan/2020/06/09/michigan-prison-population-drops-coronavirus/5326185002/;
Heather Walker, Coronavirus prompts prisons to parole inmates more quickly, WOOD (Apr. 14, 2020), https://www.woodtv.
com/health/coronavirus/coronavirus-prompts-prisons-to-parole-some-early/; 2021 Coronavirus (COVID-19) Response Q&A,
Answer 24, Mich. Dep’t of Corr., https://www.michigan.gov/corrections/0,4551,7-119-9741_12798-521973--,00.html (last visited
Mar. 29, 2022).

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

72

Examining Prison Releases in Response to COVID 		

Appendix E

Though 161 people were approved for conditional medical release, 156 were released because five were
disciplined subsequent to approval. Some other reasons listed for denying applications were community
safety, no residence, medical condition deemed not serious enough, and having a life sentence, which
made the person automatically ineligible for the program.
MN DOC also developed a Sanction Reduction Program during the pandemic. This involved a review
of individuals who had been returned to prison due to technical violations to determine who could be
released early. There were 28 people released through this program. Finally, DOC expanded its work
release program during the pandemic. This was done by broadening the eligibility criteria: individuals
further away from their supervised release date, with higher recidivism risk scores, and with previously
excluded offenses were made eligible for work release because of the pandemic. No start or end date
was listed for this program, but as of September 7, 2021, 264 individuals were released through the
expanded work release program.112

Montana
On April 1, 2020, Montana Governor Steve Bullock issued a directive regarding COVID-19. In relevant part,
the directive tasked the Montana Department of Corrections with providing assistance to the Board of
Pardons and Paroles to consider early release for certain individuals, “but only so long as they do not pose
a public safety risk and can have their medical and supervision needs adequately met in the community.”
The Board considered those aged 65 or older, those with medical conditions putting them at a high risk
during the pandemic and those who were otherwise medically frail, those who were pregnant, and
those who were nearing their release date. According to reports, as of November 2020, only three people
had been granted parole based on those conditions; one additional person was granted medical parole
with COVID-19 listed as a factor in the Board’s consideration.113

New Jersey
On April 10, 2020, New Jersey Governor Philip Murphy signed Executive Order No. 124, requiring the parole
board and department of corrections (DOC) to implement an expedited process to consider certain people
in prison for parole and furlough. Targets for this process included those older than 60, those with high-risk
medical conditions, those denied parole within the previous year, and those with short amounts of time
left on their sentences. Anyone convicted of murder, sexual assault, or other serious crimes was not eligible
for the expedited process. As of late October 2020, more than 1,200 people had been released from prison
under Executive Order 124, according to Governor Murphy. At least 300 of those 1,200 people were approved
by the DOC commissioner for furlough (as of August 5, 2020), though there were no exact numbers detailing
how many were granted parole and how many were granted furlough.114
112. COVID-19 Updates, Minn. Dep’t of Corr. (last visited Mar. 29, 2020), https://mn.gov/doc/about/covid-19-updates/; Mgmt. Analysis
and Dev., Research summary: Prison population management (Dec. 21, 2020), https://mn.gov/obfc/assets/Appendix%20
A%20Ombuds%20for%20Corrections%20COVID%20Report_tcm1157-470275.pdf.
113. Steve Bullock, Directive implementing Executive Orders 2-2020 and 3-2020 related to state correctional and statecontracted correctional facilities, Off. of the Governor: State of Mont. (Apr. 1, 2020), https://bloximages.newyork1.vip.townnews.
com/montanarightnow.com/content/tncms/assets/v3/editorial/4/f3/4f3aedda-7452-11ea-8967-bf80c092f212/5e84f12a200ce.
pdf.pdf; Mara Silvers, Breaking out in prison: COVID-19 gaining traction in Montana correctional facilities, Mont. Free Press
(Nov. 8, 2020), https://www.bozemandailychronicle.com/coronavirus/breaking-out-in-prison-covid-19-gaining-traction-inmontana-correctional-facilities/article_2dffc0de-1c92-5fcc-ab3a-d934783c76aa.html.
114. N.J. Exec. Order No. 124 (Apr. 10, 2020), http://d31hzlhk6di2h5.cloudfront.net/20200410/c0/64/ce/2c/0ef068b5d2c6459546c33a46/
EO-124.pdf; Joe Atmonavage, First wave of 50 inmates approved for release from N.J. prisons under Murphy’s order, NJ.com
(Apr. 27, 2020), https://www.nj.com/coronavirus/2020/04/first-wave-of-50-inmates-approved-for-release-from-nj-prisonsunder-murphys-order.html; Daniel Israel, Curbing the spread of COVID-19 in state prisons, Hudson Rep. (Oct. 20, 2020),
https://hudsonreporter.com/2020/10/20/curbing-the-spread-of-covid-19-in-state-prisons/; Murphy Administration blew it on
prison release. But it can still save lives, NJ.com (Aug. 5, 2020), https://www.nj.com/opinion/2020/08/murphy-administrationblew-it-on-prison-release-but-it-can-still-save-lives-editorial.html.

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

73

Examining Prison Releases in Response to COVID

Appendix E

Then, in September of 2020, the New Jersey state legislature passed the Public Health Emergency Credit
Bill (S-2519). The bill awards up to eight months of public health emergency credits to people in prison
who are within a year of their maximum parole eligibility dates (excluding those convicted of murder
or aggravated sexual assault, and repetitive, compulsive sex offenders). The law’s effects continue on a
rolling basis while the state is under a public health emergency declaration. As of March 31, 2021, around
3,675 individuals had been released from New Jersey prisons because of S-2519. Approximately 2,258
of those releases occurred on November 4, 2020 (the day the law took effect) and after that about 300
people were released each month. In the summer of 2021, Governor Murphy ended the state’s public
health emergency and thus closed the window for earning credits through S-2519. The last person to be
released early was on October 4, 2021. Between November 4, 2020, and October 4, 2021, 5,181 individuals
were released.115

New Mexico
New Mexico Governor Michelle Lujan Grisham signed an executive order on April 6, 2020, designed to
release people in prison to prevent the spread of COVID-19. The executive order essentially commutes
individuals’ sentences, placing them on parole instead. In order to qualify for commutation, a person in
prison had to meet strict criteria: be scheduled to be released in the next 30 days; have a parole plan;
not be serving time for driving while under the influence, domestic abuse, or assault on a peace officer;
not be a sex offender (even if they are not currently serving time for a sex offense); and not be serving an
enhanced sentence for use of a firearm. The first group of releases, the day after the order was signed,
consisted of 10-12 people. As of September 30, 2021, that number had risen to over 550.116

New York
On April 30, 2020, New York Governor Andrew Cuomo announced that the New York Department of
Corrections and Community Supervision (DOCCS) would be releasing some pregnant women from
prison due to concerns about the COVID-19 pandemic. The pregnant women were all serving time
for non-violent offenses and had under six months remaining on their sentences. Eight women were
released on parole starting on May 6, 2020.117

115. Suzette Parmley, Has 'COVID Time' Legislation Worked, and What Does It Mean for NJ's Criminal Justice Reform Future?,
Law.com: N.J. L. J. (May 11, 2021), https://www.law.com/njlawjournal/2021/05/11/has-covid-time-legislation-worked-and-whatdoes-it-mean-for-njs-criminal-justice-reform-future/; Lauren del Valle & Leah Asmelash, New Jersey releases more than 2,200
eligible inmates under nation's first public health crisis sentencing law, CNN (Nov. 4, 2020), https://www.cnn.com/2020/11/04/
us/new-jersey-prisoners-covid-trnd/index.html; Karen Yi, NJ Cut Its Prison Population By 40% During 11 Months Of the
Pandemic, Gothamist (Nov. 22, 2021), https://gothamist.com/news/njs-cut-its-prison-population-by-40-during-11-months-ofthe-pandemic.
116. Associated Press, County Jails Contend With High-Risk Environment for COVID-19, U.S. News & World Rep. (Sept. 30,
2021),
https://www.usnews.com/news/best-states/new-mexico/articles/2021-09-30/county-jails-contend-with-high-riskenvironment-for-covid-19; Elise Kaplan, Gov. orders early release of some inmates, Albuquerque J. (Apr. 6, 2020), https://
www.abqjournal.com/1440938/gov-orders-early-release-of-some-inmates-from-prison.html; Rachel Knapp, More than 500
inmates and counting released due to pandemic concerns, KRQE News 13 (July 7, 2021), https://www.krqe.com/health/
coronavirus-new-mexico/more-than-500-inmates-and-counting-released-due-to-pandemic-concerns/; Jeff Proctor, Massive
COVID-19 outbreak at a southern NM prison hits just one type of inmates — sex offenders. That’s by design., N.M. In Depth
(June 27, 2020), https://nmindepth.com/2020/06/27/massive-covid-19-outbreak-at-a-southern-nm-prison-hits-just-one-typeof-inmates-sex-offenders-thats-by-design/.
117. Justin Bey, 8 pregnant women to be released from New York prison over virus fears, CBS News (May 6, 2020), https://
www.cbsnews.com/news/coronavirus-new-york-prisons-pregnant-women-freed-covid-19/; Angelina Chapin, I Was 7 Months
Pregnant in Prison. Then COVID-19 Hit, The Cut (May 27, 2020), https://www.thecut.com/2020/05/i-was-pregnant-in-a-newyork-prison-then-covid-19-hit.html; Nick Reisman, New York Moves To Release Pregnant Inmates, Spectrum News 1 (May 1,
2020), https://spectrumlocalnews.com/nys/central-ny/ny-state-of-politics/2020/04/30/new-york-pregnant-inmate-release.

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

74

Examining Prison Releases in Response to COVID 		

Appendix E

Also in April of 2020, DOCCS announced that it would begin releasing other people in prison in an effort
to combat the pandemic. DOCCS generated lists of potentially eligible people who were serving time
for nonviolent, nonsexual offenses and were within 90 days of their release date. Most of them were
over 55 years old. The lists were reviewed, and DOCCS conducted investigations of each eligible person
to ensure they had an adequate housing plan and, if needed, access to treatment programs. More than
3,480 people had been released through this process as of December 10, 2020.118

North Carolina
At the beginning of the COVID-19 pandemic, the North Carolina Department of Public Safety (DPS)
began increasing its use of discretionary sentence credits in order to release people sooner, in an effort
to prevent the spread of COVID-19. This increased use of sentence credits applied to those serving time
for nonviolent, nonsexual offenses who were at an increased risk from COVID-19 (such as people with
underlying conditions and pregnant people). Between March 1, 2020, and May 3, 2020, 485 people were
released via this method. On July 14, 2020, it was reported that a total of 600 people had been released,
likely through sentence credits.119
On April 13, 2020, DPS announced that it was beginning an initiative to decrease the prison population
which would come to be known as Extending the Limits of Confinement (ELC). ELC is not considered
an early release but rather allows people in prison to serve the remainder of their sentence at home
under strict conditions. In order to be eligible for ELC, a person must not have been convicted of a
violent crime against a person and must fall into one of the following categories: 1) pregnant offenders,
2) offenders age 65 and older with underlying health conditions, 3) female offenders age 50 and older
with health conditions and a release date in 2020, 4) offenders age 65 and older with a release date
in 2020, 5) offenders on home leave or work release with a release date in 2020. These requirements
were periodically modified throughout the pandemic, allowing for release dates further into the future,
among other changes. As of May 3, 2020, 182 people had returned to the community through ELC; on
May 7, 2020, that number had increased to 192; and, finally, as of the middle of June 2020 359 people
had been “released” through ELC.120
On April 8, 2020, five North Carolina organizations (the ACLU of North Carolina, Disability Rights North
Carolina, Emancipate NC, Forward Justice, and the National Juvenile Justice Network) filed a lawsuit
seeking to ensure that North Carolina public officials took further action to stop the spread of COVID-19,
particularly in prisons. The case, NC NAACP v. Cooper (Rights of Incarcerated People), was settled on
February 25, 2021, bringing about the release of 4,450 people in state custody between February 15, 2021,

118. Jonathan Bandler, Coronavirus: Prison inmates sue New York for early release over COVID-19 fears, Lohud (Apr. 17, 2020),
https://www.lohud.com/story/news/2020/04/17/inmates-westchester-prisons-sue-new-york-early-release-over-covid-19fears/5151361002/; Chelsia Marcius, ‘A lack of compassion’: Lawyers say New York prisons are dragging their feet releasing
eligible inmates amid COVID concerns, New York Daily News (Dec. 10, 2020), https://www.nydailynews.com/coronavirus/nycovid-new-york-prisons-early-release-20201210-swdkwiioezadbdmjzlw4rnt5ua-story.html.
119. Samantha Kummerer, 'Feels a little hopeless:' North Carolina families frustrated with prison system as COVID-19 pandemic
continues, ABC 11: WTVD-TV (July 14, 2020), https://abc11.com/north-carolina-prison-nc-covid-cases-19-covid-19/6317355/; Gary
Robertson, Associated Press, Virus Directives Mean Hundreds More NC Prisoners Go Home, U.S. News & World Rep. (May 7, 2021),
https://www.usnews.com/news/best-states/virginia/articles/2020-05-06/outer-banks-counties-announce-date-for-reopening-tovisitors; Mackenzie Stasko, Hundreds of inmates released early from NC prisons over COVID-19 risk, CBS 17 (May 4, 2020), https://
www.cbs17.com/news/north-carolina-news/hundreds-of-inmates-released-early-from-nc-prisons-over-covid-19-risk/.
120. Adult Correction Actions on COVID-19, N.C. Dep’t of Pub. Safety, https://www.ncdps.gov/our-organization/adult-correction/
adult-correction-actions-covid-19#may--20 (last visited Apr. 3, 2022); Pamela Walker, Pandemic Prompts Department of
Public Safety to Transition Some Offenders to Supervision in the Community, N.C. Dep’t of Pub. Safety (Apr. 13, 2020), https://
www.ncdps.gov/news/press-releases/2020/04/13/pandemic-prompts-department-public-safety-transition-some-offenders.

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

75

Examining Prison Releases in Response to COVID

Appendix E

and August 21, 2021. The Division of Adult Correction and Juvenile Justice released people through three
main mechanisms: ELC, discretionary sentence credits, and post release supervision and parole actions.121

North Dakota
In March 2020, the North Dakota Parole Board held a special meeting to grant early parole to 120 people
as part of its COVID-19 mitigation efforts. The Parole Board considered a person’s medical conditions, the
amount of time left on their sentence (the Board was looking for those with nine months or less left), and
whether they had a reliable place of residence. In April 2020 the Parole Board heard 141 cases and granted
parole for 120 of those people. As of May 8, 2020, the Parole Board director said that the Board had met
facility goals concerning COVID-19 and would not be holding another special meeting for extra requests.122

Ohio
On April 7, 2020, Ohio Governor Mike DeWine released a COVID-19 update which, in relevant part,
announced that the state’s overcrowding emergency statute would be used so that certain people in
prison scheduled to be released within 90 days could be considered for early release. A number of
factors disqualified people from consideration: being convicted of serious charges (such as sex offenses,
homicide-related offenses, kidnapping, etc.); having been denied judicial release in the past; having a
prior incarceration in Ohio; having convictions from another state; having active warrants or detainers;
and having had a serious prison rule violation in the last five years. Governor DeWine recommended
141 qualifying individuals to the Ohio Correctional Institution Inspection Committee (a joint-legislative
committee that provides external oversight of Ohio's prisons), which approved 105 for early release. On
April 15, 2020, Governor DeWine officially approved the release of those 105 individuals. Then, on April
17, 2020, Governor DeWine also commuted the sentences of 7 people because of their age and / or preexisting conditions that could be exacerbated by COVID-19.123

Oklahoma
On April 10, 2020, Oklahoma Governor Kevin Stitt approved commutations for 452 people in an effort
to reduce the state’s prison population because of the COVID-19 pandemic. Though his criteria for the
commutations were not specified, most people were incarcerated for either drug possession crimes or
property crimes. Just under a quarter of the people who received commutations (111) were released on April

121. Adjusted Reentry Dates for Offenders, N.C. Dep’t of Pub. Safety (2021), https://www.ncdps.gov/adjusted-reentry-datesoffenders; NC NAACP V. Cooper (Rights of Incarcerated People), ACLU of N.C. (2021), https://www.acluofnorthcarolina.org/
en/cases/nc-naacp-v-cooper-rights-incarcerated-people; Prison Population Reduction Efforts, N.C. Dep’t of Pub. Safety
(2021), https://www.ncdps.gov/our-organization/adult-correction/prisons/prison-population-reduction-efforts#post-releasesupervision-and-parole-commission-actions.
122. Julie Martin, More than 100 DOCR inmates granted early parole due to COVID-19, KFYR (May 8, 2020), https://www.kfyrtv.
com/content/news/More-than-100-DOCR-inmates-granted-early-parole-due-to-COVID-19-570318211.html.
123. Corr. Inst. Inspection Comm., https://www.ciic.state.oh.us/about (last visited Jan. 21, 2022); COVID-19 Update: Liquor Sales,
Office of Small Business Relief, Ohio Prisons, SNAP Payments, Mike DeWine - Governor of Ohio (Apr. 7, 2020), https://
governor.ohio.gov/wps/portal/gov/governor/media/news-and-media/liquor-sales-office-of-small-business-relief-ohio-prisonssnap-payments; Nick Swartsell, DeWine Authorizes Release of 105 Inmates as Coronavirus Cases in Ohio Prisons Swell
into the Hundreds, CityBeat (Apr. 16, 2020), https://www.citybeat.com/news/dewine-authorizes-release-of-105-inmates-ascoronavirus-cases-in-ohio-prisons-swell-into-the-hundreds-12169663; 2 high-profile prisoners have sentences commuted
by DeWine amid coronavirus crisis, WTOL 11 (Apr. 17, 2020), https://www.wtol.com/article/news/health/coronavirus/ohioprisoners-sentences-commuted/512-3786f7da-1d88-4d6f-b57c-6e18ce89e6c6.

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

76

Examining Prison Releases in Response to COVID 		

Appendix E

16, 2020. The remaining people were not released from prison immediately, due to other charges that were
not commuted, but were set to be released early (at varying times) thanks to the commutations.124
The Oklahoma Department of Corrections (DOC) identified 126 people in prison whose medical conditions
put them at a higher risk from COVID-19. After eliminating those serving time for violent crimes, sex
crimes, and other (unspecified) categories, the director of the DOC sent a letter to the Oklahoma Pardon
and Parole Board on May 1, 2020, recommending 14 people for medical parole, as he is authorized to do
by Oklahoma statute. By the time the Pardon and Parole Board considered this list, one of those 14 had
already received parole and another waived his right to be considered for parole due to the proximity of
his release date. On May 13, 2020, the Pardon and Parole Board approved the remaining 12 people for
medical parole.125

Oregon
On June 25, 2020, Oregon Governor Kate Brown announced that she would be commuting the sentences
of 57 people in prison. She had received a list from the Oregon Department of Corrections (DOC) of 61
people for consideration. The 57 people Governor Brown authorized for commutations were particularly
vulnerable to COVID-19, had served at least half of their sentences, were not convicted of committing a
violent crime against another person, had housing, and had access to healthcare. Most of those released
were white men convicted of drug and property offenses.126
On September 28, 2020, Governor Brown announced that she had commuted 66 more sentences
through a similar process; on September 21 the DOC sent her a list of 69 people to consider. All the
people Governor Brown selected had a plan for housing and health care; 10 were medically vulnerable
to COVID-19 and the other 56 were within two months of their release dates. On December 15, 2020,
Governor Brown announced that she was set to commute the sentences of approximately 130 people to
be released December 17, 2020. All those receiving commutations were not serving person crimes, had
housing plans for after their release, and had a record of good conduct for 12 months; some had less than
six months left on their sentences and others were at a higher risk for COVID-19.127

124. Kayla Branch, Error from Stitt's office leads to inaccurate count of prisoners to be released, The Oklahoman (Apr. 16, 2020),
https://www.oklahoman.com/article/5660266/error-from-stitts-office-leads-to-inaccurate-count-of-prisoners-to-be-released;
Shardaa Gray, Oklahoma inmates released from prison across the state because of COVID-19, FOX25 (Apr. 16, 2020), https://
okcfox.com/news/local/oklahoma-inmates-released-from-prison-across-the-state-because-of-covid-19; Hicham Raache, Gov.
Stitt approves hundreds of prison commutations to mitigate coronavirus spread, KFOR (Apr. 10, 2020), https://kfor.com/
news/coronavirus/gov-stitt-approves-hundreds-of-prison-commutations-to-mitigate-coronavirus-spread/?eType=EmailBlastC
ontent&eId=07a90610-fdf3-4e1b-994a-712a02c5065c.
125. 2001 OK. HB 2924, https://plus.lexis.com/document/?pdmfid=1530671&crid=963d660e-8557-4a0d-9685-d2a96306b197&pd
docfullpath=%2Fshared%2Fdocument%2Fstatutes-legislation%2Furn%3AcontentItem%3A4J1B-8GH0-0033-41Y8-0000000&pdcontentcomponentid=125155&pdworkfolderlocatorid=NOT_SAVED_IN_WORKFOLDER&prid=798894bf-b319-45878f64-108c9d0e480d&ecomp=-t4hk&earg=sr4; Chris Polansky, Board Recommends Special Medical Parole For 12 State
Inmates, Public Radio Tulsa (May 13, 2020), https://www.publicradiotulsa.org/local-regional/2020-05-13/board-recommendsspecial-medical-parole-for-12-state-inmates#stream/0.
126. Shane Kavanaugh, Oregon Gov. Kate Brown approves early release of 57 inmates vulnerable to coronavirus, The Oregonian
(June 25, 2020), https://www.oregonlive.com/coronavirus/2020/06/gov-kate-brown-approves-early-release-of-57-inmatesvulnerable-to-coronavirus.html; Conrad Wilson, Oregon Governor Commutes Sentences Of 57 Inmates Vulnerable To
COVID-19, OPB (June 25, 2020), https://www.opb.org/news/article/oregon-governor-commutes-57-prison-sentences-covid-19/.
127. Jayati Ramakrishnan, Governor commutes sentences for 66 more inmates, including 10 considered vulnerable to COVID,
The Oregonian (Sept. 29, 2020, https://www.oregonlive.com/pacific-northwest-news/2020/09/governor-commutes-66-moreinmates-sentences-including-10-considered-vulnerable-to-covid.html; Conrad Wilson, Oregon prisons to release more
inmates as COVID-19 outbreaks continue, East Oregonian (Dec. 15, 2020), https://www.eastoregonian.com/coronavirus/
oregon-prisons-to-release-more-inmates-as-covid-19-outbreaks-continue/article_7f423606-3f12-11eb-b1ee-572e37fe342a.html.

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

77

Examining Prison Releases in Response to COVID

Appendix E

Pennsylvania
On April 10, 2020, Pennsylvania Governor Tom Wolf issued an executive order in response to the COVID-19
pandemic requiring the Pennsylvania Department of Corrections (DOC) to establish a Reprieve of Sentence
of Incarceration Program through which the DOC would recommend qualified individuals to Governor Wolf
for consideration for issuance of a conditional reprieve. To qualify, a person had to be vulnerable to COVID-19
(either because of their age, medical conditions, or pregnancy); either eligible for release within the next
twelve months or within nine months of their minimum eligibility release date; and not convicted of certain
enumerated crimes (largely violent and sex-related crimes). Between the issuance of the order and February
10, 2021, Governor Wolf granted 165 reprieves. On June 22, 2020, Governor Wolf announced that, in addition
to the reprieve program, a number of measures had been taken to reduce the prison population. These
measures included furloughing paroled individuals from centers to home plans, maximizing parole releases,
reviewing parole detainers, expediting the release process for anyone with a pending approved home plan,
and reviewing and releasing people who are beyond their minimum sentences. However, we could find
no documentation of the number of people affected by these measures. Some news articles reported that
the prison population decreased by 3,312 people; however, this was likely due primarily to a decrease in
admissions rather than an increase in releases.128

Rhode Island
Following a request by Rhode Island prosecutors, prison authorities, and public defenders, on April 3,
2020, the Rhode Island Supreme Court issued an order establishing a process whereby superior courts
and district courts would enter orders providing for the “immediate release” of 52 people from prison.
The orders reduced their sentences, making the individuals eligible for release. All 52 people had less
than 91 days left on their sentences and were serving time for nonviolent offenses.129

Utah
At the beginning of March 2020, the Utah Board of Pardons and Parole (BOPP) began collaborating with
the Utah Department of Corrections (UDC) to release some individuals early because of the COVID-19
pandemic. All the people released through this effort had already had a hearing and been granted a
release; on average they were released a little more than two months before their originally scheduled
release date. In July 2020 the BOPP reported that, from the beginning of March 2020 through the end of
June 2020, 730 individuals were released with COVID-19 related considerations. As of November 7, 2020,
press reported that more than 1,000 people were released from prison early because of the pandemic.130

128. Pennsylvania's Prison Population Reduced by Nearly 3,500 Since March 1, Erie News Now (June 22, 2020), https://www.
erienewsnow.com/story/42276002/pennsylvanias-prison-population-reduced-by-nearly-3500-since-march-1; J.D. Prose, PA
legislators to Wolf: Give more reprieves to inmates vulnerable to COVID-19. His response., The Times (Feb. 10, 2021), https://
www.timesonline.com/story/news/2021/02/10/pennsylvania-inmates-release-covid-19/4455476001/; Tom Wolf, Order of the
Governor of the Commonwealth of Pennsylvania Regarding Individuals Incarcerated in State Correctional Institutions,
Commonwealth of Penn. Off. of the Governor (Apr. 10, 2020), https://famm.org/wp-content/uploads/2020.4.10-TWW-SCIreprieve-release-order-COVID-19.pdf.
129. In re Req. for Prison Census Control In Resp. to COVID-19 (Apr. 3, 2020), https://www.courts.ri.gov/PDF/In%20re%20Request%20
for%20Prison%20Census%20Control%20(Order).pdf; Mark Reynolds, R.I. Supreme Court OKs release of 52 inmates,
Providence J. (Apr. 4, 2020), https://perma.cc/3JUV-NFLY.
130. Annie Knox, Some Utah inmates get early homecoming, others fear the worst as virus spreads in prison, Deseret News (Nov.
7, 2020), https://www.deseret.com/utah/2020/11/7/21546012/coronavirus-parole-inmates-get-early-homecoming-others-fearthe-worst-as-virus-spreads-in-prison; Utah’s Prison Population Trend in Context of COVID-19, State of Utah Bd. of Pardons
and Paroles (July 30, 2020), https://bop.utah.gov/index.php/home-top-public-menu/2-uncategorised/167-bopp-coronavirusinformation-2.

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

78

Examining Prison Releases in Response to COVID 		

Appendix E

Vermont
Due to concerns regarding the COVID-19 pandemic, Jim Baker, the commissioner for the Vermont
Department of Corrections (DOC), made an effort to decrease the Vermont prison population during
the early stage of the pandemic. Baker started by “looking at which inmates can be let out on furlough
and who can be released on probation.” This resulted in almost 100 people being released from prison
between March 19, 2020, and March 26, 2020, which brought the DOC’s total number of releases since
late February to over 200. It was then reported on April 30, 2020, that 255 people had been released by
the DOC since March 13, 2020.131

Virginia
On April 22, 2020, the Virginia General Assembly approved a proposed budget amendment from Governor
Ralph Northam which gave the director of the Virginia Department of Corrections (DOC) authority to
release people from prison early during the COVID-19 pandemic. The early release plan developed by the
DOC had several requirements. People had to have a viable housing plan, be at a low to medium risk of
recidivism, have less than one year left to serve, have their good time earning at a certain level, and have
no active detainer. If they were convicted of a class 1 felony or a sexually violent offense, they were not
considered. All final decisions about early releases were made by the Director of Corrections, who also
considered the individuals’ medical conditions. People were released periodically throughout the DOC’s
period of authority: 62 were released during the week after the amendment was approved, and 606 had
been released as of October 15, 2020. When the DOC’s authority to release individuals early terminated
at midnight on July 1, 2021, a total of 1,376 people had been released early.132

Washington
On April 15, 2020, Washington Governor Jay Inslee issued both a commutation order and an emergency
proclamation; both were in response to the COVID-19 pandemic and aimed at reducing the prison population.
The commutation order and emergency proclamation combined to result in the release of 1,016 people
from prisons. In order to be eligible for release, a person must not have had a violent or sexual offense and
had to have a release date within 180 days of the Governor’s orders. The releases were via commutation of
sentences (422 people), work release furlough (66 people), or a “rapid reentry” process (528 people) by which
the individuals served the remainder of their sentences at home with electronic monitoring.133
131. Sheldon Burnell, Courts reviewing calls for more prisoners to be released from Vermont prisons due to COVID-19, Community
News Service at UVM (Apr. 30, 2020), https://www.communitynews.net/home/courts-reviewing-calls-for-more-prisoners-tobe-released-from-vermont-prisons-due-to-covid-19; Anna Merriman, ‘It’s very difficult to control’: Many Vermont inmates
released so that those who remain can be spread out, Valley News (Mar. 26, 2020), https://www.vnews.com/Vermont-NHprisons-working-to-reduce-population-to-prevent-virus-spread-33512589.
132. COVID-19 Response: Inmate Early Release Plan, Va. Dep’t of Corr. (2020), https://vadoc.virginia.gov/media/1506/vadoc-covid19early-release-plan.pdf; Frank Green, 62 inmates released early - so far - in COVID-19 relief for prisons, Richmond Times-Dispatch
(Apr. 29, 2020), https://perma.cc/N2X6-LWCC; Katherine Hafner, Virginia has been releasing hundreds of prisoners during the
pandemic. Critics argue it’s not enough, Virginian-Pilot (Oct. 15, 2020), https://www.pilotonline.com/government/virginia/vp-nwinmate-early-release-numbers-20201015-ksstatr6n5fy3ipj2x7uqamwtm-story.html; Press Release, Va. Dep’t of Corr., PandemicRelated Early Release of State Inmates Coming to an End as Authority Expires (June 16, 2021), https://vadoc.virginia.gov/newspress-releases/2021/pandemic-related-early-release-of-state-inmates-coming-to-an-end-as-authority-expires/; Va. Dep’t of Corr.,
COVID-19/Coronavirus Updates, https://vadoc.virginia.gov/news-press-releases/2021/covid-19-updates/ (last visited Mar. 27, 2022).
133. Emergency Commutation in Response to COVID-19, Jay Inslee (Apr. 15, 2020), https://www.governor.wa.gov/sites/default/files/
COVID-19%20-%20Commutation%20Order%204.15.20%20%28tmp%29.pdf; Wash. Proclamation No. 20-50, (Apr. 15, 2020),
https://www.governor.wa.gov/sites/default/files/proclamations/20-50%20-%20COVID-19%20Reducing%20Prison%20Population.
pdf; COVID-19 Frequently Asked Questions (FAQ), Wash. State Dep’t of Corr., https://www.doc.wa.gov/corrections/covid-19/faq.
htm#reentry (last visited Mar. 27, 2022); COVID-19 Incarcerated Population Reduction Efforts | Commutations, Wash. State Dep’t
of Corr. (2020), https://www.doc.wa.gov/corrections/covid-19/docs/reduction-efforts-commutation.pdf; COVID-19 Incarcerated
Population Reduction Efforts | Rapid Reentry, Wash. State Dep’t of Corr. (2020), https://www.doc.wa.gov/corrections/covid-19/docs/
reduction-efforts-rapid-reentry.pdf; COVID-19 Incarcerated Population Reduction Efforts | Work Release Furloughs, Wash. State
Dep’t of Corr. (2020), https://www.doc.wa.gov/corrections/covid-19/docs/reduction-efforts-work-release-furlough.pdf.

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

79

Examining Prison Releases in Response to COVID

Appendix E

West Virginia
At the beginning of the COVID-19 pandemic, the West Virginia Division of Corrections and Rehabilitation
began to consider how it could decrease prison population density in the state in order to mitigate the
spread of the COVID-19 virus. After weighing release options within its authority under state law, the
Division had released about 70 people who were serving short terms for parole-related sanctions as of
April 1, 2020.134

Wisconsin
At the beginning of the COVID-19 pandemic, Wisconsin released some people from prison through a
variety of mechanisms in order to help address the COVID-19 crisis. Between March 1, 2020, and May
8, 2020, 53 people were granted release by the Wisconsin Parole Commission. Between March 2, 2020,
and May 4, 2020, 1,447 individuals who had been detained because they violated terms of their parole,
probation, or extended supervision were released by the Wisconsin Department of Corrections (DOC).
There were 65 individuals who had been in a Wisconsin prison for treatment as part of the Alternative
to Revocation program were released on April 2, 2020. Finally, 7 people were released through Certified
Earned Release, a program for people who were within 12 months of their release and who were
sentenced between 2009 and 2011. As of May 8, 2020, a total of 1,572 people had been released from
Wisconsin prisons.135

134. Leslie Ruben, W.Va. taking steps to reduce inmate population amid COVID-19 pandemic, WCHS/WVAH (Apr. 1, 2020), https://
wchstv.com/news/coronavirus/wva-taking-steps-to-reduce-inmate-population-amid-covid-19-pandemic.
135. Gina Barton & Natalie Brophy, Coronavirus can spread quickly through a prison -- so what can Wisconsin do to keep inmates,
guards and the public safe?, Milwaukee Journal Sentinel (Apr. 2, 2020), https://www.jsonline.com/story/news/2020/04/02/
coronavirus-wisconsin-inmates-risk-virus-spreads-prisons/5097246002/; COVID-19 (Coronavirus): FAQs, State of Wis. Dep’t
of Corr., https://doc.wi.gov/Pages/COVID19(Coronavirus)/FAQS/COVID19FAQs.aspx (last visited Mar. 30, 2022); Emily Hamer,
Wisconsin DOC has released nearly 1,600 inmates so far to combat COVID-19 spread, Wis. State J. (May 8, 2020), https://
madison.com/wsj/news/local/crime-and-courts/wisconsin-doc-has-released-nearly-1-600-inmates-so-far-to-combat-covid-19spread/article_03537daa-e1ec-5fe8-ac68-f5cf38ce8be5.html (https://perma.cc/58WK-99UW.

A .

ROBINA INSTITUTE
OF CRIMINAL LAW ANO CRIMINAL JUSTICE

80