Sign in Subscribe
School Safety

The Numbers Went Down. The Danger Didn't.

Broward County taught the country how to make school crime disappear from a spreadsheet. The research on what followed is in.

Ryan Petty

Ryan Petty

— 10 min read
Share
The Numbers Went Down. The Danger Didn't.
AP Photo/Cliff Owen

In 2012, Broward County schools recorded 1,056 student arrests. By 2016 the number was 392.

The state's data told the same story. Florida Department of Education figures reported by the South Florida Sun Sentinel showed that by 2018 Broward suspensions were down 27 percent from the 2012-13 school year. Arrests per 1,000 students were down 64 percent.

Superintendent Robert Runcie called that success. So did Washington. The Obama administration held Broward up as a model, and the model spread. By 2017, more than 50 of the nation's largest school districts, serving more than 6 million students, had rewritten their discipline policies. Twenty-seven states had changed their laws.

The numbers were real. What they measured was not.

A falling arrest count can mean students are behaving better. It can also mean the adults have stopped writing things down. Broward spent five years showing the country the difference.

Broward Came First

Arne Duncan hired Robert Runcie into Chicago Public Schools in 2003. Six years later Duncan became Barack Obama's Secretary of Education. In July 2011, Duncan and Attorney General Eric Holder announced a joint federal initiative aimed at "ending the school-to-prison pipeline." Two months after that, Runcie left Chicago to run Broward County Public Schools, with Duncan as a reference.

In November 2013, Runcie signed an agreement with the Broward Sheriff's Office, the State Attorney, the Public Defender, the Department of Juvenile Justice, and the NAACP. It created PROMISE. Thirteen misdemeanors, including vandalism, trespassing, and marijuana possession, would no longer go to police. Students would spend a few days at an alternative site instead.

Sheriff Scott Israel described the goal plainly: "We measure our success by the kids we keep out of jail, not by the kids we put in jail."

On January 8, 2014, the Departments of Education and Justice issued their "Dear Colleague" letter on school discipline. It told every district in America that a discipline policy could violate federal civil rights law even when it "is neutral on its face" and "is administered in an evenhanded manner," if it "has a disparate impact." Unequal numbers were now enough to bring an investigation.

Runcie understood the order of events better than anyone. "We went out early on," he told Scholastic in 2014. His staff, he said, joked that the administration had adapted Broward's framework into the national guidelines.

Washington did not invent Broward's policy. Washington found it, praised it, and put federal civil rights enforcement behind the same idea. In 2015 the White House hosted Runcie at its "Rethink School Discipline" summit. Duncan said of Broward, "I'm very, very pleased with their work." In January 2018 the Broad Center was still reporting that "PROMISE informed the White House's guidance on student disciplinary practices nationwide."

The incentive was now national. Get the numbers down.

What the Numbers Hid

PROMISE had a feature no other diversion program in Florida had. Investigators for the Marjory Stoneman Douglas Public Safety Commission put it in one line: "PROMISE is largely consistent with Florida's civil citation criteria, but the 'offense reset' every school year is unique." A student could be referred several times in a year and start the next August with a clean slate.

Runcie claimed the program kept 90 percent of students from reoffending. The Sun Sentinel explained why that statistic "can be deceiving": "A student can commit a subsequent infraction without being considered a repeat offender, as long as it's not the exact same violation, in the exact same year."

Tim Sternberg, a former assistant principal who helped run PROMISE, told the paper what that teaches a teenager. "You can develop a psyche that it is OK to commit crime because you can refresh the clock every year." He added: "They aren't tracking kids over time." My fellow commissioner Max Schachter, who lost his son at Stoneman Douglas, called the annual reset "mind-boggling."

The pressure reached the classroom. The Sun Sentinel found teachers "expressly told or subtly pressured not to send students to the administration for punishment." A former teacher, Mary Fitzgerald, described a principal who told her "it was his job to market the school." She explained how that works in practice: "You don't report theft because reporting it makes your school look dangerous." Minutes of a February 2015 faculty council meeting recorded where it led: "The message out there is that the students are untouchable."

The paper also obtained the discipline records of the student who would later attack Marjory Stoneman Douglas High School. Between May 2012 and January 2014 he was suspended for at least 67 days at Westglades Middle School. On November 25, 2013, the month PROMISE launched, he broke a bathroom faucet and was assigned three days in the program. Commission investigators found that the program tracked attendance in a database that "did not interface" with the district's main system. The records for those three days were "inconsistent and inconclusive." The forms that should have shown he completed the program were missing from his file. He was never arrested.

Precision matters here. The commission, on which I served, examined that referral and concluded that a juvenile vandalism case would have had no effect on his legal right to buy a firearm four years later. Chairman Bob Gualtieri called the PROMISE question a "red herring." On the effect of the PROMISE program on his ability to purchase a firearm, I accept that finding. However, the impact on the safety of Broward schools and the response of law enforcement in Broward County created a culture of leniency that allowed students, including the attacker to escape any consequences for their actions. This in my view, contributed to an escalating pattern of violence by the attacker that culminated on February 14, 2018.

Max Schachter said the rest: "You weren't helping him. He got progressively more violent. The system just failed him." And the system which promoted je

The faucet was never the point. The point is what a district teaches its staff when every arrest and every suspension counts against it. I said in May 2018 that these programs had created "perverse incentives for school staff, administrators & law enforcement." The record since has confirmed it.

Runcie spent months insisting there was nothing to examine. "This particular individual was never a participant in the Promise program," he said two weeks after the attack. He called the suggestion of a connection reprehensible. In May 2018, after WLRN reported otherwise, the district admitted the referral.

Then came the statewide grand jury. Its second interim report said it flatly: "School Districts are intentionally not reporting SESIR incidents." SESIR is the state's school crime and safety reporting system. The jurors explained the motive: "the people ultimately in charge of the school districts directly benefit from maintaining – if not improving – an impression of safety and order in their districts." The final report, in August 2022, named Broward among five districts with findings against them. It described inaccurate incident reporting "whether as an inadvertent consequence of misguided policies or as a direct result of intentional malfeasance."

An impression of safety and order.

What the Research Found

Broward is one district. The research covers many more.

In December 2018, the RAND Corporation published the first randomized controlled trial of restorative practices in a large urban district, Pittsburgh. Restorative justice is sold as the way to close the "school-to-prison pipeline." RAND found that suspensions fell. The number went down there too. RAND also found this: "Arrest rates among PERC schools did not decrease." And this: "Academic outcomes did not improve in PERC schools, and actually worsened for grades 6–8." PERC was the name of the Pittsburgh program. The math losses were concentrated among black students and in predominantly black schools. Lead author Catherine Augustine told EdSource her team was "disappointed by the findings at the middle school level."

A second RAND randomized trial, in Maine middle schools, found no difference in school climate between schools that tried restorative justice and schools that did not.

In Philadelphia, researchers Matthew Steinberg and Johanna Lacoe studied a ban on suspensions for low-level misconduct. Most schools did not comply. In the poorest and lowest-performing schools, the ones that could not make the ban work, students who had never been suspended fell further behind. Racial disparity in suspensions, the stated reason for the reform, grew. The Fordham Institute, which published the study, drew the lesson: discipline reform "is best initiated at the school level rather than the district level, where the law of unintended consequences is more apt to prevail."

In New York City, Max Eden, then at the Manhattan Institute, compared student and teacher surveys before and after Mayor Bill de Blasio "required principals to seek central office approval to suspend a student who committed repeated infractions." The results: "More than half of schools saw a deterioration in mutual respect compared to a fifth saw an improvement." On fighting, gang activity, and drug use, "three times as many schools saw a deterioration as saw an improvement." The worst shifts came in secondary schools where more than 90 percent of students were minorities. Eden is careful to call his study descriptive. The pattern it describes is not subtle.

The students these policies were written to protect paid the most for them.

Defenders of restorative justice have one strong study, and they will cite it. In Chicago, researchers Anjali Adukia, Benjamin Feigenberg, and Fatemeh Momeni found that high schools trained in restorative practices cut out-of-school suspensions by 18 percent and student arrests by 19 percent, with no damage to classmates' test scores. That finding is real and deserves to be taken seriously.

Look at how Chicago did it. The district selected 73 high schools, in part for their "interest in participating." A federal grant paid for coaches who worked alongside staff several times a week. That is a program.

What Broward built, and what Washington exported, was a target.

Ten Years to Walk Away

Runcie kept his job for more than three years after the attack. He left in 2021, after the statewide grand jury indicted him for perjury over testimony about a technology contract. The school board sent him off with a severance package worth roughly $740,000. A judge dismissed the charge in 2023. An appeals court reinstated it in 2024. The state then dropped it on the day jury selection was to begin. He was never convicted of anything.

The grand jury was less forgiving of the board that employed him. It found that "the School Board of Broward County has—through deceit, malfeasance, misfeasance, neglect of duty and incompetence—mismanaged" a bond program sold to voters on school safety. In August 2022, Governor Ron DeSantis suspended four board members on its recommendation.

Washington reversed first. The Dear Colleague letter was rescinded in December 2018, days after the Federal Commission on School Safety reported that "schools ignored or covered up—rather than disciplined—student misconduct in order to avoid any purported racial disparity in discipline numbers that might catch the eye of the federal government." President Trump's Executive Order 14280, signed in April 2025, ordered new federal guidance on school discipline.

Broward moved last. After 2018 the district stopped resetting PROMISE referrals each year, capped them at three, and began reporting them to the Department of Juvenile Justice. In October 2023 the school board abolished the program. A student's fate after a misdemeanor, CBS Miami reported, "will be decided by law enforcement officials."

Ten years. That is how long it took a district to walk away from a policy whose flaws were on the public record by the summer of 2018.

Measure What Matters

My daughter Alaina was one of the seventeen killed at Marjory Stoneman Douglas. I spent the years that followed on the commission that investigated why. The lesson of Broward is larger than one program. Institutions do what they are measured on.

Tell a principal that suspensions are the metric, and suspensions will fall. Tell a deputy that arrests are the metric, and arrests will fall. Neither number tells you whether a single child is safer.

Florida changed the measure. The Marjory Stoneman Douglas High School Public Safety Act required behavioral threat assessment teams with law enforcement at the table. The commission recommended that schools report diversion data to the Department of Juvenile Justice, so a student's record follows him when he changes schools. After the grand jury reported, the state tightened its incident reporting rule to take discretion away from districts, and the Office of Safe Schools moved to see district incident data in real time.

None of this means abandoning second chances. A middle schooler who vandalizes a bathroom does not need a criminal record. He needs an adult who writes it down, a system that still remembers it next year, and a team that notices when the tenth incident looks different from the first. Diversion with a record is mercy. Diversion without one is neglect.

Find out how many incidents your district reported to the state last year, and who audits that number. Find out whether a student's discipline history follows him from one school year to the next. Find out whether the threat assessment team is told when a student is diverted from arrest.

A number that falls because the adults stopped counting is not progress. It is a warning nobody wrote down.

Sources

The Broward record

Washington

The research

The aftermath

Read more