Opinion: Despite undeniable obstacles, Rikers Island can still be closed

For New Yorkers seeking an end to generations of violence, dysfunction and racially disparate incarceration at Rikers Island, last year brought a glimmer of relief. In a long-running federal court case ostensibly settled nearly a decade ago in 2015, Judge Laura Taylor Swain held the city in contempt of its legal responsibility to provide safe and humane jail conditions. The judge ordered deliberations on a remedy early this year, likely to conclude with the appointment of a federal receiver to take control of the jails as early as this week.

Receivership can be a helpful step forward. According to Brennan Center expert Hernandez Stroud, independent receivers can improve the management of a correctional facility through their wide latitude to hire, fire and redeploy staff in ways that better assure everyone’s safety.

At the same time, receivership isn’t a panacea. In a commentary calling for receivership, the New York Civil Liberties Union cautioned, “While a receiver is necessary to address the dire and unsafe conditions at Rikers, the ultimate solution is to permanently close the jail, as required by a city law passed in 2019.” Notwithstanding this plea, the city is at growing risk of defaulting on its obligations.

Twin setbacks

New York City law mandates closing Rikers by Aug. 31, 2027. Yet the Adams administration has failed to schedule the opening of needed replacement jails in the Bronx, Brooklyn, and Queens until 2031, and it still has not finalized a contract for building the fourth jail slated for Manhattan.

Beyond the city’s delinquent construction timeline, Rikers can’t be closed without reducing the number of people the city’s jails must house on any given day.

Back on April 29, 2020, the city’s daily jail population had plummeted to a historic low of 3,809, amid humane efforts to release incarcerated people from the dangers of COVID-19. Since then, the population has swelled by more than 70% to just over 6,600 today.

A jail population this high can’t fit in planned replacement jails, whose citywide capacity the City Council set at 3,300 in a 2019 agreement. Although Mayor Eric Adams increased this capacity to 3,900 and will add over 300 beds in city hospitals for people with serious medical or mental health needs, these alterations remain insufficient unless the jail population shrinks substantially. Instead, the Adams administration has been warning for years that the population will surpass 7,000 and keep on growing, necessitating a “Plan B” that would upend years of planning.

In a vicious circle, each problem feeds the other. Delayed jail construction takes the pressure off public officials to responsibly lower the jail population by 2027. Conversely, projections of a too-high population create a ready-made excuse for continuing to delay construction under the rationale that even if the city erected new jails on time, Rikers still couldn’t close. The end result is inaction by public officials who might otherwise be fulfilling their part of the solution.

Misleading population projections

To cut through this knot, a good starting point is reassessing the administration’s gloomy population projection. It was based on data running only through August of 2022, making its results woefully outdated today. The formal analysis projected a jail population of nearly 7,400 as of September 2024. With this month having come and gone, it turns out the administration overestimated September’s jail population by close to 900 people.

At the Data Collaborative for Justice, an algorithm our data scientist, Liz Johnson, created based on updated trends in crime, arrests, and the jail population itself expects the population to decrease modestly over the next year. This algorithm puts the population at close to 6,500 people one year from now – significantly lower than the administration’s projection, though still far too high to close Rikers.

It is also worth emphasizing that no algorithm can magically pinpoint complex outcomes like future jail populations. Most important of all, current projections can’t anticipate the effects of needed policy reforms that have yet to be instituted. If courts, prosecutors and the mayor owned their power to launch effective jail reduction strategies starting tomorrow, the population’s future trajectory would necessarily shift downward from where it appears to be headed today. 

This is precisely why our organization’s future projections integrate new data every day and are continually updated in response to unforeseeable developments.

Cautious optimism

Let’s be clear that current trends put Rikers’ closure in jeopardy. However, the status quo offers at least three reasons to persevere.

First, we can discard the administration’s prior warning of a skyrocketing jail population destined to zoom past 7,400 this fall. Additional signs are encouraging. Both 2023 and 2024 saw drops in murders, shootings and other major crimes. For nearly two years, misdemeanor, nonviolent felony and violent felony crime trends have all leveled off. Until the pandemic struck in 2020, the longer three-decade arc of crime and jail trends since the early 1990s has been a hand-in-hand decline in both metrics.

Second, we already know how to reduce the use of jail while maintaining public safety. Two reports released in 2021 identified numerous reforms backed by research. A special issue released last March in the journal Vital City drew attention to many solid proposals: expanding supervised release as an alternative to pretrial detention; investing more funds in targeted housing and mental health services for justice-involved people; and adopting proven practices for tackling the case processing delays that unacceptably prolong people’s stays at Rikers. My own article identified a mix of strategies that could shrink the jail population well below 4,000 – even under a realistic assumption of far-from-perfect execution.

Some of the most widely recommended reforms would likely increase public safety. Research tells us that even when the current charge involves violence, releasing New Yorkers before trial if they are facing a first-time arrest reduces recidivism. Because housing instability has been found to increase recidivism, it is more beneficial to link homeless people to supportive housing than to warehouse them in jail only to return them to the streets once their case ends. Women and people well into adulthood (often pegged at ages 55 and up) pose a low recidivism risk, alongside high vulnerability to trauma and other adverse health effects of incarceration – yet the women’s jail population has been increasing more than men’s at the same time that women face an extensively reported and ongoing risk of sexual abuse at Rikers.

Despite documented policy levers for smartly reducing incarceration, it can’t be stressed enough that there is no viable means of pulling them without the judiciary. A single data point illustrates this reality: 85% of today’s jail population consists of people held before trial – due to worsening case processing delays and because the city’s judges too often set bail that people cannot afford. A second data point – continually updated on our organization’s new Rikers dashboard – shows the grave inequity the status quo is producing: Today’s jail incarceration rate is 11 times higher for Black than white people.

The good news is that these facts lead directly to our third reason for optimism: Since assuming office nearly a year and a half ago, New York state Chief Judge Rowan Wilson has taken a series of positive steps. 

Last summer, Wilson launched new judicial trainings on the state’s bail reform law. He publicly endorsed the Treatment Not Jail Act – state legislation that would expand early access toward the outset of a case to the city’s demonstrably effective mental health courts. Chief Administrative Judge Joseph Zayas announced a case processing reform consisting of scheduling orders and case conferences in between court dates. Case conferences where prosecutors and defense attorneys talk to each other directly outside the courtroom were one of the key practices in a 2019 Brooklyn pilot that slashed the length of time people languished in pretrial detention.

Obstacles to address

But it’s not enough to paint promising possibilities and ignore inevitable roadblocks.

It will require great diligence for the court system’s praiseworthy leadership team to change everyday practice in the city’s courtrooms. Court leaders will have to respect individual judges’ independence, while guiding them away from long entrenched but ineffective case processing techniques and guiding them toward more care in reserving bail for the low proportion of cases that pose a genuine flight risk – the sole legal criterion that can justify bail in New York.

To back up the court system’s case processing reform, I’ve previously written in favor of state legislation setting strict pretrial detention time limits of six months (with limited exceptions and a longer time limit for homicides). A less stringent failsafe could be an automatic hearing before a specially assigned judge in each borough whenever someone reaches the six-month mark in detention. The purpose would be to interrogate what must happen and who must do it to deliver on the individual’s constitutional right to a speedy trial and, in the meantime, to consider release options involving community supervision and individualized services.

Crucially, closing Rikers hinges on the Adams administration owning its accountability for able stewardship. An urgently needed first step would be immediately ordering all relevant city agencies to engage in scrupulous study of how to shorten the status quo’s late timeline for borough-based jail construction.

To incentivize real change, it would be tactical malpractice to compromise over the August 2027 closure deadline before key officials take significant productive steps within their purview.

Implementation issues

Even with the best of intentions, we can’t ignore that changing business-as-usual is plain hard. Positive results will hinge on fruitful collaboration among city officials, judges and attorneys. 

Efforts must take seriously the boring nuts and bolts of implementation by drawing on solid “what works” research, memorializing operational plans in writing, training and retraining judges and prosecutors to avoid slippage into old routines and self-critically monitoring results and refining policies yet again. Implementation science makes for bad soundbite material, but it’s important.

While no one can guarantee the future, our accumulated knowledge suggests the city’s too-high jail population and further obstacles to closing Rikers are solvable. They will, however, require city and state leaders to summon both a will to act and a requisite aptitude for proceeding assiduously.