Stay Informed

Get the latest updates:

Prison Policy Initiative newsletter (?)
And our other newsletters: Research Library updates (?) Prison gerrymandering campaign (?)

Email:
Name (optional):
State (optional):

Parole in Perspective: How parole decisions are made

By Leah Wang    Tweet this
October 2025
Press release

Table of Contents

Parole grant rates

Decisionmaking criteria

Decision transparency

Conclusion and recommendations

Appendices

Across the country, in meeting rooms and behind computer screens, thousands of incarcerated people are asked to explain every accomplishment, slip-up, and remorseful feeling they’ve experienced while behind bars in hopes of securing their release. And those parole applicants¹ may be denied parole for reasons that are unclear or unrelated to their preparedness to go home. In some states, 8 of every 10 of parole applicants can expect to be denied and have to wait anywhere from 1 to 15 years before their next hearing.²

It’s nearly impossible to know if parole decisions are the product of a board’s wide discretion, adherence to clear guidelines, or both, and this murkiness is part of what makes parole a dysfunctional release mechanism. Those inconsistent decisions lead to wildly different parole grant rates³ (or approval rates) from state to state.

In absolute terms, nearly every state with discretionary parole is granting release to fewer people each year, and many are holding fewer parole hearings in 2024 compared to five years ago. In order to understand these trends, we looked into how parole boards approach their decisionmaking. Few states publish detailed, disaggregated information about parole outcomes, but what few data do exist suggest there are demographic disparities and a fixation on the original crime of conviction rather than the individual’s readiness for release.

This is the second installment of our two-part report, Parole in Perspective, that references many of the Principles for Parole Reform we developed with the MacArthur Justice Center to envision what parole systems should look like. This series is intended for lawmakers, advocates, journalists, incarcerated people, and others interested in making discretionary parole a more effective pathway for release. Our analysis covers 35 states that have discretionary parole, including Louisiana (which nearly fully abolished discretionary parole in August 2024⁴) and California (which eliminated parole for most people in 1977, but still holds thousands of discretionary parole hearings each year for those eligible).

In the first part of this report, How parole boards and hearings work, we examine parole boards in these 35 states — how they’re formed, who is appointed to them, and how they carry out their work. We also look into how parole hearings work, such as whether they are public, or held in person. Our policy scan reveals that states take wildly different approaches to parole boards and hearings, most of which are unlikely to provide a fair, efficient process for those who are eligible or become eligible for release through reform. In this second part, we dive deeper into parole grant rates and the decisionmaking guidelines and realities that lead to those rates.

Discretionary parole grant rates in 35 states

We filed public records requests and pored through publicly available reports and data dashboards to find the number of parole hearings and grants in 2023 and 2024 (adding to our last update covering 2019 through 2022). For the 28 states that provided updated data through 2023 and/or 2024, some granted parole to more than three out of every four applicants, while others could barely grant one out of every four. Overall, just ten states⁵ granted parole to half or more applicants. We can’t explain every state’s grant rate or trends over time, but we know that underresourced parole boards are struggling to stay fully appointed while facing growing caseloads. This may lead to fewer hearings and rushed decisions, which are likely to be denials given the political nature of parole.

To see the full dataset, including the hearing and grant numbers underlying these rates, and notes about each state’s data, see Appendix Table 2.

Grant rates. The highest discretionary parole grant rates came from Wyoming (76%) and North Dakota (71% in 2023) — two states that have some of the lowest state prison populations and, accordingly, likely have small parole caseloads. At the other end of the spectrum, California held over 6,000 parole hearings in 2024 but only granted parole in about one in five cases, and South Carolina granted parole to an alarmingly low 4% of applicants in 2024. Twenty-one states had lower grant rates, and only seven states⁶ had higher grant rates, in 2023 or 2024 (depending on data availability) compared to 2019.

Please note that while it is tempting to compare different states’ grant rates, each state’s data are likely to be different, despite our efforts to standardize what counts as a hearing and what counts as a grant. We’ve included our sources and methodological notes in the full appendix table whenever possible to allow for further exploration of the underlying data.

Hearings. Importantly, our analysis finds that discretionary parole hearings have become more rare over time. The 25 states with aggregate hearing data had over 68,000 fewer hearings in 2024 compared to 2019.⁷ This is not the direction that parole hearings should be trending, especially amid the crisis of an aging prison population. As we mention in the first part of this report, progressive reforms tackling eligibility and sentencing have increased the number of parole-eligible people in many states. There are over 200,000 people in the United States whose parole eligibility date is in the past, according to the Council of State Governments. Why, then, are parole boards holding fewer hearings?

We can’t fully explain this shift away from parole, but it’s not because there are smaller prison populations. In fact, many states are refilling their prisons after pandemic-related declines in incarcerated populations. Parole boards have just as much work as they ever did, yet they are holding fewer discretionary hearings and, in a handful of states, issuing a disproportionate number of denials. For example, Georgia, whose prison population grew over 7% from 2021 to 2023 and continues to expand, held 22% fewer parole hearings in 2024 than in 2019. Meanwhile, the number of people granted parole in Georgia is down 42%. And in Rhode Island, where the prison population has steadily grown since 2022, there were 12% fewer parole hearings in 2024 than there were five years earlier, and the number granted parole shrank by nearly one-third in that time.

Punitive reforms are partially to blame for bloated prison populations, negating progress toward decarceration by offering parole to more people. As the Sentencing Project explains in its report about delays in the parole process, state sentencing laws are increasingly harsh toward parole-eligible people, forcing them to serve more time before an initial hearing or after a denial.

Decisionmaking criteria: How parole boards evaluate applications for release

Buried in the dense text of state laws, all states with discretionary parole have language describing the fundamental task of parole boards: to determine whether someone serving a parole-eligible sentence in prison should be released.¹⁰ Parole boards publish manuals, available on their websites, that refer to and mirror those laws. However, we could not locate a central source of information that specifically describes parole decisionmaking criteria across jurisdictions, so we combed through statutes and procedural manuals for 35 state discretionary parole systems to do so. Appendix Table 3 breaks down some of the core language in each state’s parole criteria, which we discuss in further detail in the following sections.

We found that guidelines in multiple states range from overly vague and antiquated to highly specific and difficult for parole applicants to meet. It’s important to understand that the burden is on the parole applicant, not the parole board, to demonstrate readiness through these guidelines. In reality, most parole applicants do not pose a risk to the community; therefore, an effective parole system should instead start with the presumption of release, then require the parole board to justify why release is inappropriate. (See the sidebar below to learn more about states that have implemented presumptive parole in some way.)

Sure enough, every parole board discussed here should (and does) have some degree of discretion, rather than being bound to inflexible rules. But we found that parole board members tend to exercise their discretionary powers by departing from guidelines in favor of keeping people locked up. If parole boards were more heavily composed of people with relevant lived or clinical experience, they might use discretion to advocate for release when an applicant has met clear guidelines and demonstrated growth and preparation.

The release standard: Parole applicants must show that they’ll lead “a correct life”

State parole systems tend to have one core standard to which they hold parole applicants. If they meet that “release standard,” based on all the information available, then they are suitable for release on parole.¹¹ In most states (24 of 35), this standard says that the parole applicant will present “a reasonable probability” that they will not violate the law, that they will “fulfill the obligations” of a law-abiding citizen, or something very similar. Other states word this differently, but generally expect applicants to be a “good” person, leading a “correct” life.¹² It is reasonable for a parole board to be concerned with an applicant’s likelihood of violating the law. Perhaps it’s even acceptable to desire other positive traits of them. But in reality, these standards are exceedingly vague.¹³ They present a nearly impossible standard, one that no other group of people will ever have to meet, and they give boards significant leeway to issue denials.

A second sentence: Most parole boards heavily consider the original crime of conviction, despite the fact that a judge has already done so at sentencing

In all 35 states with discretionary parole, parole boards are required to consider the “seriousness” or “severity” of the crime of conviction, the applicant’s criminal history, or both, even though these have already been addressed in the process of being sentenced to prison. When judges determine sentences, they are first and foremost considering the nature of the crime and any aggravating or mitigating circumstances (factors that make the crime more or less severe). A sentence with the possibility of parole indicates that the judge (or state legislature) believes the timing of the individual’s release should depend on their acceptance of responsibility, their successful completion of required programming, their demonstrated plan for going home, and other factors. These are the criteria parole boards are better suited to consider; the judge has already considered the criminal case itself.¹⁴

Worded just slightly differently, guidelines for six parole boards¹⁵ include consideration of whether releasing someone may “depreciate” or “diminish” the seriousness of the crime, or undermine respect for the law. Similarly, in South Dakota, the parole board must consider the “effect of release on administration of justice.” These provisions are related to the optics of releasing certain people, rather than their actual preparedness; high-profile or chronic parole denials often cite these reasons. Larger reforms to broaden parole eligibility also run up against rhetoric claiming that parole “elevate[s] and benefit[s] perpetrators of crime” while showing disregard for victims.

In that vein, Alabama and Idaho statutes refer to the intent of parole as ensuring prison space is devoted to, or focused on, the “most serious offenders “or the “highest-risk” people. This language might at least suggest that there is an incentive to move most people out of overcrowded prisons and into the community, but Alabama’s parole grant rates in particular are far too low for this optimistic interpretation. And Idaho’s parole system earned a solid “F” in our 2019 report for having policies that make parole hearings and preparation more difficult, keeping 2,500 people behind bars past their eligibility date, so this statutory language rings hollow.

Parole boards have tunnel vision when it comes to conduct and accomplishments behind bars

About half of states’ parole statutes make reference to an individual’s conduct, their participation in programs, education, treatment, and other accomplishments while incarcerated. Certainly, these criteria constitute some of the forward-looking and objective factors we wish were more central in parole decisionmaking. But they are sometimes misapplied in ways that lead to people being unfairly denied release.

Programming and treatment. To be sure, parole boards do often acknowledge participation in and completion of programming, earning educational certificates or degrees, and consistent work histories. Yet the bar is set unfairly high for these criteria: Board members may see an incompletion, withdrawal, or other “gap” in an applicant’s resume, and start to ask questions, even if those changes occurred because of a transfer, sickness or injury, or sincere schedule conflict.

We have also seen parole boards grant parole on the condition that they complete a particular program or treatment regimen, even though they may have a long waitlist or may not even be available where an applicant is incarcerated. Mandated programming, particularly with treatment known to be effective inside or outside of prisons, is no reason to keep someone behind bars when they’re otherwise ready to be released. This gridlock arises because of siloed systems: The parole board decides the criteria for someone’s release, while the prison system runs these programs and decides who is eligible to take them. Lawmakers need to be aware that these two parts of the criminal legal system are working against each other. If parole policies allowed for participation in these programs to occur in the community, people granted parole could actually go home and receive treatment.¹⁶

Misconduct records. Similarly, parole boards apply uneven logic to conduct in prison, which is measured by disciplinary reports: Board members can home in on just one instance of alleged misconduct (even if an applicant only has one “ticket,” in total) in order to discredit an otherwise exemplary or unproblematic incarceration.¹⁷ This double standard cannot be overstated: Consistent good behavior is not good enough to earn parole, but even minor misbehavior can serve as a reason to keep someone behind bars. For example, the California parole commissioners who recently denied parole to the Menendez brothers cited prison rule violations — namely, using cell phones to communicate with their wives — as a major reason for the denials. The lengthy disciplinary policies found in every prison system can punish eventual parole applicants in multiple ways, both prolonging incarceration and increasing the likelihood of being denied parole.¹⁸

Risk assessments: Most parole boards look at calculated scores, but their final impact on decisions is mixed

Most parole boards we analyzed used a risk assessment tool or score calculator at some point in their deliberation process. In some states, the entirety of their parole guidelines could be boiled down to one worksheet, producing one score, which the board could lean on in their decision. But as we’ve learned by now, deviating from guidelines or a risk assessment score is expected and even common.

Risk assessment tools are algorithms that produce a numerical score describing someone’s supposed risk. They’re intended to give the impression of fairness and objectivity in decisionmaking — whether in pretrial detention, prison classification, parole, or supervision — by utilizing existing criminal legal system data. However, those data reflect years of racial bias by various actors in the system.¹⁹ Factors that a person can’t change over time (like arrest history and the crime of conviction) tend to weigh too heavily in risk calculations, while the final score doesn’t capture how a parole applicant has grown, changed, and prepared for release.

Alabama and West Virginia are just two states whose statistical parole data show that the guidelines-recommended grant rate was far higher than their actual parole grant rate every single month.

Risk assessments are somewhat redundant in parole decisionmaking: Many of the same factors that a parole board are supposed to consider separately are incorporated into these tools. Discretion can come in handy in overriding an otherwise unfavorable score, if a parole applicant can explain their circumstances and meet the release standard in their state. And fortunately, several states have laws or policies directing the parole system to be transparent in how often final decisions agree with their actuarial guidelines. Yet unfortunately, the data suggest that states often flagrantly disregard those guidelines (as shown in the charts above).

Not every parole system was transparent about the finer details of its use of risk assessments, but in the following footnote, we’ve compiled a list of states and the risk assessment tools that parole boards use or consult at some point.²⁰ It’s worth investigating how risk scores are generated and how parole board members interpret what the scores mean.

Less than half of states with discretionary parole publish hearing decisions, and far fewer actually explain their reasoning

Although we are not advocating for the broad release of personal information, we were curious about how many states publish individual parole decisions, particularly if they contained the parole board’s reasoning in issuing a grant or denial, and if denials were accompanied by specific and actionable steps for a better chance next time.

Table 3. Fifteen states in our analysis publish individual parole decisions on the internet, though two states require searching decisions by name. Way fewer states actually publish the reasons for denial, though they may provide these in writing directly to a parole applicant.

State	Publishes individual parole decisions	Publishes reasons for denying parole
Alabama	Yes	No
Alaska	No	No
Arkansas	Yes (must search by name)	No
California	Yes	No
Colorado	No	No
Connecticut	No	No
Georgia	No	No
Hawaii	No	No
Idaho	Yes	No
Iowa	No	No
Kentucky	No	No
Louisiana	Yes	No
Maryland	No	No
Massachusetts	Yes (life sentences only)	Yes
Michigan	No	No
Mississippi	No	No
Missouri	No	No
Montana	Yes	Yes
Nebraska	No	No
Nevada	Yes	No
New Hampshire	No	No
New Jersey	No	No
New York	Yes	No
North Dakota	Yes	No
Oklahoma	Yes (must search by name)	No
Pennsylvania	No	No
Rhode Island	Yes (meeting minutes)	Yes
South Carolina	Yes	No
South Dakota	Yes (meeting minutes and “all action” reports)	No
Tennessee	No	No
Texas	No	No
Utah	Yes	Yes (very limited)
Vermont	No	No
West Virginia	No	No
Wyoming	No	No

 

Of the 15 states that we found publish individual parole outcomes online, only four publish more detail on the board’s rationale. In Utah, this additional information is very limited if included at all, while more information is provided in Massachusetts, Montana, and Rhode Island. These documents provide a window into how the parole board came to its decision, whether it adhered to any guidelines, and whether they provide recommendations to applicants. For example, Montana’s monthly disposition reports show countless individuals denied parole for reasons such as, “Nature and/or Severity of Offense(s); Parole at this time would diminish seriousness of the offense.” It is disheartening to find parole boards clinging to the original crime of conviction as their reason for denial, even if statutes explicitly allow for it. Another boilerplate reason we found in Montana’s dispositions, “Strong Objection from Criminal Justice Agencies,” is not one of their statutory factors.

A parole outcome in Montana from earlier in 2025. The parole board leaves some suggestions for what programming to enter, but some of the reasons for denial are unlikely to change in two years.

Putting it all together: What should discretionary parole look like?

As a writer for The Marshall Project wrote ten years ago, “no one really knows how to get parole.” Indeed, the way most parole statutes are written, parole can be denied for nearly any reason at all. Parole statutes pay lip service to important dynamic factors, like rehabilitation, preparation, public support, and education, but too often issue denials for static factors, like the original crime, or overly subjective factors, such as the optics of releasing someone who has been convicted of such a crime. Such denials send a harmful message: The parole board neither recognizes nor rewards transformation. The current state of parole, though different from state to state, is clearly a bottleneck to release, contrary to its original intent. By our calculations, discretionary parole is a broken promise for over 110,000 people annually,²¹ with fewer people receiving hearings and far fewer receiving parole over the past five years.

Some advocates and experts have suggested presumptive parole as a promising “fast lane” to release for many applicants, which would allow parole boards to devote more time to complex hearings. The presumption of release is a fairer framework for parole in general, though it’s not widespread (see the sidebar for more information). In addition to this paradigm shift centering success and transformation, most parole systems can substantially improve their functionality and equity by tapping into reforms we’ve discussed in these companion reports, including:

• expanding parole boards;
• diversifying who is appointed to parole boards;
• ensuring hearings are accessible;
• clarifying guidelines and omitting static factors; and
• increasing data collection and decision transparency.

Of course, these recommendations only touch on some aspects of the immense universe of parole, but they are some of the most pertinent to parole applicants and their supporters who are getting ready for hearings.

Discretionary parole is not beyond repair. We hope to make lawmakers and the public increasingly aware of parole’s potential as well as its shortcomings, and identify ways their states can implement the fundamental principles of parole. Using these reports as a guide, we encourage readers to look deeper into the statutes, policies, and data underlying their parole board’s operation, and identify levers for change.²³

• Home
• About
• Publications
• Issues
• Data
• Blog
• Trump Tracker
• Contact
• Careers
• Donate

Contact us online or write us: Instagram Twitter Facebook LinkedIn

Donate