The Politics of a Prison-Release Settlement

On January 29, 2022, Tyrell Brace shot into a Charlotte home and killed Elante Thompson, a computer science student with a young daughter. 

Republicans have argued that Thompson would still be alive had it not been for a February 2021 settlement then-Gov. Roy Cooper reached with the ACLU of North Carolina and other advocacy groups to lower the prison population by at least 3,500 inmates to stop the spread of COVID-19 in crowded state facilities. Brace shot Thompson after serving a 24-month prison sentence for felony and misdemeanor assault charges. 

Cooper’s critics allege that the settlement indiscriminately turned loose thousands of violent criminals like Brace, who then went on to commit other crimes. Cooper has repeatedly denied the charge, and his supporters point to the unique threat of COVID-19. More than 9,500 inmates had tested positive for the virus by the time the settlement was signed, and 47 had died. 

The settlement has become a major flashpoint as Cooper runs for an open U.S. Senate seat against Michael Whatley, former chair of the Republican National Committee and the state GOP—one of the highest-profile races in the country. Polls show Cooper ahead, though the gap appears to be narrowing. 

A joint House-Senate subcommittee on prisons that state Republican leaders formed this spring to investigate the settlement had scheduled its first hearing for 1 p.m. Monday but abruptly canceled it last Friday. Leslie Dismukes, the current secretary for the state Department of Adult Correction, and Orlando Rodriguez, the agency’s general counsel, have been called to testify. 

Central to Republicans’ argument is a May 7 New York Post analysis of the settlement that asserts nearly half of the inmates it released went on to commit new crimes, including 18 homicides. But a more careful look at the final full list of 4,455 inmates, made public this February, shows that narrative is misleading. (A judge ordered the list sealed as part of the settlement agreement and there was never a court order unsealing it; Senate Leader Phil Berger obtained the list, which made its way to Fox News before the Department of Adult Correction released it to the media.) There are no names on the list, only offender identification numbers. To find out more information about an inmate on the list, you have to plug the offender identification number into the prison system’s online information portal. 

The Assembly’s review of the full list finds it is inaccurate to say that all of these inmates were released early. Many were released on timelines unrelated to the pandemic or the settlement. Brace, for example, left prison on July 25, 2020 after serving his full minimum sentence, records show. 

He is one of 1,535 inmates on the list marked with an asterisk noting that they committed alleged violations of post-release supervision, which typically runs nine to 12 months. In cases like Brace’s, prison officials decided not to send them back to prison to meet the settlement’s goal of reducing overcrowding. In other words, these inmates weren’t released because of the settlement; they just weren’t sent back to prison. Even if Brace had returned to prison, he would have been released months before the 2022 shooting.

Our analysis found that of the roughly 3,000 inmates who were released early because of the settlement, more than 90% would have been out by the end of 2021 anyway by virtue of completing their sentences. The agreement allowed prison officials to let them out early through a variety of methods—awarding credits for things like good behavior to shave time off their sentences, releasing qualified nonviolent inmates to house arrest, and accelerating parole for eligible inmates serving life or long active sentences under a previous sentencing structure. 

The New York Post reported that the North Carolina Sentencing and Policy Advisory Commission calculated the recidivism rate based on at least 3,500 inmates. It didn’t. The commission analyzed only a sample of 1,180 inmates released between February and June 2021, finding that 48% of them were rearrested within two years. That rate was slightly higher than that of inmates who were released that same year under normal procedures, the report said. The Post also reported incorrectly that inmates like Brace were released early. 

Cooper has pushed back on efforts to pin an alleged crime wave on his leadership. “These blatant lies from Republicans have been fact-checked for months and found to be false,” his spokesperson said in a statement. 

Whatley spokesperson D.J. Griffin said in a statement that the candidate stands with Elante Thompson’s mother, Debra, who told the New York Post:  “Why would you release somebody like that? They’re already showing they’re a gangster to society. You’re going to release a menace on the street?” 

Griffin declined to answer additional questions. 

Given the intense rhetoric around the settlement, The Assembly decided to dig deeper on its origin and impact, as well as how it’s been discussed in the course of the Senate campaign. 

A Unique Threat

As the spread of COVID-19 shut down much of the world in 2020, Cooper joined other governors in issuing stay-home orders. And as the medical system strained under spiking numbers of severe infections, leaders in N.C. and many other states issued mask and social-distancing mandates. Cooper gave daily updates about the state’s efforts to contain the virus. Nearly 30,000 North Carolina residents have died due to the disease in the last six years. 

Public health experts noted that, because of confined quarters and an inability to effect social distancing, incarcerated people are some of the most at-risk for transmission. At Rikers Island in New York City, for example, one infection exploded into 200 in only 12 days. 

The first infection in a North Carolina prison was confirmed on April 1, 2020, at Halifax’s Caledonia Correctional Institution (now known as Roanoke River Correctional Institution). Later that month, an outbreak at the Neuse Correctional Institution in Goldsboro affected more than 400 inmates. There was no vaccine at that point, and medical experts were still scrambling to figure out treatments. 

Sandra Marriner, who was serving a life sentence for second-degree murder at Swannanoa Correctional Center for Women in Black Mountain, was paroled in March 2021 as part of the settlement. (She was one of the plaintiffs in the lawsuit.) But before then, she lived in fear of how fast the infections were spreading. She suffered from chronic obstructive pulmonary disease and high blood pressure, putting her at greater risk of dying from the virus. Prison officials tried to prevent inmates from interacting and handed each of them three one-ply cloth masks, she said. But it wasn’t enough. She eventually became ill that fall; she believes she caught it from a caseworker. 

“I thought, ‘I’m going to be one of the ones that it takes out,’” Marriner said. Luckily, it was a mild case; she had a fever and couldn’t smell anything. She was let out of quarantine after twice testing negative. But dozens of others weren’t as lucky.

Early on, civil rights groups like the ACLU of North Carolina, Disability Rights North Carolina, and Emancipate NC began raising concerns about prison conditions. They urged leaders to take the virus more seriously and reduce the prison population, which at that point was near its 34,000 capacity. 

One-third of the state’s inmates had a disability, and almost a quarter were over 50, making them even more vulnerable, the advocates noted. Releasing some of them could help keep both other incarcerated individuals and prison staff, who go home to their families each night, safe. To not do so was a violation of their constitutional protections against cruel and unusual punishment, the advocates said.  

“Five or six years ago, we were all working with far less information than what we now have about COVID-19 and far fewer resources about how to respond to it,” Corye Dunn, director of public policy for Disability Rights North Carolina, told The Assembly. 

Those advocacy groups and several inmates filed suit against Cooper, then-Secretary of Public Safety Erik Hooks, and other officials on April 20, 2020. They wanted a judge to force the Cooper administration to implement safety protocols, such as regular testing, and reduce the overall prison population to slow the spread. Without these measures, “COVID-19 will be a death sentence for many incarcerated people,” the lawsuit warned.

On June 16, Wake Superior Court Judge Vinston Rozier Jr. granted a preliminary injunction. Prison conditions were likely unconstitutional, Rozier concluded, and state officials needed to come up with a plan to reduce overcrowding, stop inmate transfers within the prison system, and adopt other safety protocols. 

Whatley himself criticized Cooper for not doing more about COVID in correctional settings at an event the following month, telling a group of fellow Republicans that the governor “did not act to protect folks in prison, so we now have a significant spike among the prison population as well as prison guards.” 

The state GOP also issued a news release saying that Cooper only took the prison situation seriously when a judge ordered him to. 

Charlotte news station WSOC-TV reported on Whatley’s 2020 comments earlier this month, which led Whatley to weigh in at a press conference. “What we saw in many other states was they took steps to address it that involved COVID shots, they involved masks, they involved separation,” Whatley said, per The News & Observer. “Certainly, we did not call in any way for there to be a release of prisoners, let alone 4,200 of them, including the worst of the worst, with pedophiles and predators and rapists and murderers.” 

 Cooper, in turn, accused his opponent of being hypocritical. “It’s pretty clear that he’s made this COVID release the center of his political attack, but he’s been caught lying about it again and again,” he said at an event on July 13. 

Emancipate NC attorney Elizabeth Simpson said state officials, including Cooper, fought their virus-mitigation efforts every step of the way: “It was an extremely adversarial litigation.” 

Prison officials also said they were already trying to ease prison overcrowding before the suit, accelerating parole for some people serving life sentences under a previous sentencing structure, sending qualified inmates to house arrest, and awarding discretionary credits to shave time off inmates who they felt posed little risk to public safety. 

North Carolina was not alone in this effort. In October 2020, a California appellate court ordered the release or transfer of 1,700 inmates out of San Quentin State Prison. A month later, New Jersey Gov. Phil Murphy signed an executive order to release more than 2,000.

But they weren’t working fast enough for advocates, or for the judge. In his July 2020 ruling, Rozier wrote that he was “extremely concerned by the apparent indifference” of state officials. He said officials had failed to provide him sufficient information about what they were doing to combat the virus and that they had not complied with his orders to stop transfers of inmates between facilities. He also said state officials had failed to make other changes to prevent the spread. 

Over the next several months, the advocacy groups and state officials haggled over whether the state was complying with the court’s orders. The case appeared to be headed to trial. 

In December 2020, Rozier appointed Thomas Maher, then the executive director of Duke’s Wilson Center for Science and Justice, to ensure that state officials followed his orders.

Two months later, the parties decided to settle. Rather than reducing the prison population by 18,000 inmates, as the advocacy groups initially sought, the settlement agreement required the state to lower it by at least 3,500 within six months. 

Putting the Agreement into Practice

The Cooper administration used three primary criteria to move people out of prisons. One was to identify nonviolent offenders who met at least one of the prison system’s mandatory criteria, such as being 65 or older, and let them serve the remainder of their sentences under house arrest. 

A second method was awarding credits to reduce inmates’ time so they could get out as close to their minimum sentence release date as possible. Inmates could earn credits through good behavior or by participating in certain programs. In many cases, the inmates who were awarded these credits were just a few months shy of hitting their minimum sentences. In the sample of 1,180 inmates that the sentencing commission reviewed, 61% were released through this criteria. 

Officials also accelerated parole for inmates who received life or longer active sentences for crimes committed between 1981 and 1994, when state legislators eliminated parole. The settlement allowed state officials to parole these inmates if they were medically vulnerable, had participated in a structured reentry program, and were expected to pose a low risk to public safety. 

State prison officials also looked at people who had already completed their sentences but had been flagged for violating post-release supervision, in order to determine who should be sent back to prison and who could continue in supervision. 

Typically, supervised release lasts for nine to 12 months. If someone violates the terms of their release, they can be sent back to prison for the remainder of that period. If the violation was a new crime, they may be resentenced separately. From the available records, it’s difficult to tell exactly how long any of these people might have remained back in prison–but those who were allowed to remain on post-release supervision due to the settlement appear to be largely people with technical violations, like missing a check-in.

These cases accounted for 34% of the names on the settlement agreement list.  

The Sentencing and Policy Advisory Commission’s analysis of an 1,180-inmate sample from the full list concluded that 89% had been let out fewer than three months before their projected release date. 

Nearly half had been convicted of nonviolent, low-level felonies, which typically come with a  sentence of less than a year. Another 37% were convicted of mid-level felonies—offenses such as child abuse, assault with a deadly weapon and habitual impaired driving. The remaining 16% had been convicted of serious felonies such as second-degree murder, rape, armed robbery, and large-scale drug trafficking but were approaching release dates set well in advance of the agreement. 

The difference in recidivism rates was clustered in the latter subgroup. Fifty-one percent were rearrested within two years, compared with 31% of those in that same category released under normal circumstances. 

But that does not mean they were rearrested for violent crimes, nor does it reflect whether they were ultimately convicted on those charges. The New York Post reported that 18 people committed murder after being released. But of those, at least seven of them were already out of prison before the settlement was even signed.

Calvin Locklear appears to be the only one who killed someone during the time period he would have still been imprisoned were it not for the settlement. He was released on May 21 and shot two people, killing a 33-year-old man and injuring a 17-year-old girl, a month later. His projected release date was August 12, 2021. 

The Fight Over Facts

The actual data on recidivism is a challenge to unwind, and politicians have used that complication to their advantage. High-profile homicides, like the fatal stabbing of 23-year-old Ukrainian refugee Iryna Zarutska on a Charlotte light rail train last August, have also provided ample opportunity to capitalize on misinformation. 

At a July 9 event, Whatley held up a mugshot of DeCarlos Brown, the man charged with killing Zarutska. Gruesome surveillance videos showing the assault captured national headlines, and state Republican leaders used her death to push for a namesake tough-on-crime bill. Gov. Josh Stein signed it into law late last year. Brown has thus far been deemed mentally unfit to stand trial.

Whatley claimed that Brown was released early as part of Cooper’s settlement deal. But as The Assembly previously reported, Brown actually got out of prison on September 20, 2020, after serving a 73-month sentence for armed robbery. 

In 2021, his sister accused him of assault, a potential violation of his 12-month post-release supervision. But after his sister declined to press charges, state prison officials decided not to revoke the supervision. That supervision ended in September 2021, four years before he killed Zarutska. Even if he’d been sent back to prison for the violation, he would have been out within a year—and three years before the crime.

Whatley also pulled up a mugshot of Lucas Scronce, who served a year and six months on drug and stolen property charges before he was released on July 12, 2021, two months before his projected release date. In February 2024, he was accused of killing his 16-month-old daughter with a drug overdose, to which he pleaded guilty to second-degree murder and is now serving up to 19 years in prison. As tragic as that case may be, it still took place nearly three years after the standard release date for his previous crime. 

Monday’s now-canceled hearing was meant to be a high point of a months-long campaign to blame Cooper for the settlement and to paint him as soft on crime. But this isn’t the first time the agreement has come under scrutiny. 

In 2021, it was hotly debated at the March and September meetings of the North Carolina Sentencing and Policy Advisory Commission, which makes recommendations to the General Assembly about sentencing policies and laws. Republicans also condemned the agreement during a March 2021 hearing of the Joint Appropriations Committee on Justice and Public Safety. 

“Did y’all ever consider that data of what this is going to mean for the people that are living out here, releasing 3,500 criminals?” then-Rep. Allen McNeill, a Moore County Republican who left office in 2023, asked correction officials Tim Moose and Todd Ishee.

GOP leaders in the General Assembly formed the new subcommittee in April to investigate the prisoner release and have framed it as a chance to finally hold Cooper accountable. 

House Speaker Destin Hall called the settlement a “miscarriage of justice” but has said little about what the committee’s investigation actually entails. 

“While the settlement itself was public in 2021, the individuals on Roy Cooper’s early release list were kept secret until earlier this year, when Gov Ops obtained and released it,” a Hall spokesperson said in a statement. “Now that the identities of those on the settlement list are finally accessible, it is evident that the list included dozens convicted of murder, rape, and other violent crimes. North Carolinians deserve to know how and why Roy Cooper put their safety at risk this way.”

Berger spokesperson Lauren Horsch has said Cooper “could have fought the early releases in court but refused to,” and should be held accountable for “keeping the public in the dark about the violent criminals that were released, including those that had lifetime sentences.” She did not respond to questions about the hearing or about whether Berger had tried to obtain more information about the settlement before this year. 

A Cooper campaign spokesperson dismissed the hearing as a political stunt and accused lawmakers of trying to “boost Roy’s opponent only 80 days before early voting starts.” Cooper has not been called to testify. 

Aaron Littman, a professor of law at UCLA who has studied the agreement for the COVID Behind Bars Data Project, said he’s puzzled by the uproar five years later. 

“Was it perfect? No. Was it everything that advocates were asking for? No,” he said. “But unlike in some other states, it reflected a serious attempt to grapple with a problem in state prisons without allowing it to get totally out of control. I think portraying it as some extravagant and unconsidered release of people is totally inaccurate.” 

The hearing has not yet been rescheduled.