Fix broken windows, both the concept and on the subway

We must stop conflating broken windows policing with stop-and-frisk and zero tolerance


It’s no secret that in the 1970s through the 1990s, the New York City subway was seen as the poster child for urban decay.

AP Photo

This essay is reprinted with permission from the Violence Reduction Project

In September 2017, the Manhattan District Attorney’s office began to decline prosecution of fare evasion. The other New York City district attorneys followed suit. They also stopped prosecuting other rule violations and petty crimes in the subway. Having policed some of the largest transportation hubs in the nation and having spent five years of my career policing the subways, I knew that once disorder crept into a transportation system, the system will become less safe, causing riders to flee for other alternatives.

Unfortunately, I was right.

In 2018, crime in the subway increased. Whether from fear or a desire to avoid beggars, the homeless and the mentally ill, more riders shifted to other transportation modes, and ridesharing apps made this choice rather simple. Between 2017 and 2018, subway ridership decreased by about 3%. Even during the COVID-19 pandemic, when ridership is down 30%, crime is still up, and this despite the system being shut down for the first time in history, from 1 am to 5 am every night.

In addition to declining ridership, fare evasion costs the Transit Authority more than $200 million a year. This puts maintenance and capital projects on hold. This perfect storm of decline started with the decision to tolerate disorder and not prosecute low-level offenses.

The theory of broken windows, introduced by James Wilson and George Kelling in a 1982 Atlantic Monthly article, was never popular among a certain kind of reformer because, at its core, Wilson and Kelling believed in the positive possibility of policing, that good police could actually maintain order and prevent more serious crime. This conflicted with the “root causes” model, still accepted as faith in most of academia, that refuses to believe that crime reduction can come from anything except reductions in poverty, racism, unemployment and other social ills.

The power of Broken Windows theory

The fact that broken windows policing did reduce crime in New York City and elsewhere in the 1990s did little to mollify critics. And the cause of broken windows wasn’t helped when it morphed into stat-based Zero Tolerance policing in the 2000s in New York City.

Used correctly, broken windows is a powerful tool in a beat cop’s tool belt. And Kelling and Wilson envisioned it being used alongside another tool in that belt, police officer discretion. In 2016 I sat on a panel at Princeton University with George Kelling. During the question and answer session, Kelling asked the audience members who were critical of his theory if they had actually read the Atlantic Monthly article. Very few hands were raised.

Somewhere along the line, the theory of broken windows policing – the idea that enforcing low-level crimes such as fare evasion, graffiti and public drinking can prevent more serious crimes – was written off as harmful to low-income communities despite the fact that these communities most benefited from more public order and vastly less crime.

It’s no secret that in the 1970s through the 1990s, the New York City subway was seen as the poster child for urban decay. One just needs to watch Coming to America or The Warriors to see what Hollywood thought of the New York City subway. Crime was rampant, cars were covered with graffiti and filled with filth, and people literally put their lives on the line using it. In 1979, the system was averaging 250 felony crimes a week (about 13,000 that year) when Curtis Sliwa started the Guardian Angels to combat widespread violence on the system.

In 1984, Bernhard Goetz, who after being robbed three years earlier took to carrying a gun for protection to ride the subway, shot four Black teens who attempted to rob him. The majority of the public and New Yorkers initially supported Goetz.

In the mid-1980s, the New York City Transit Authority received funds and began a serious effort to clean up the subway.

The “Clean Car Program” began in 1984. Over a five-year period, graffiti, something most people had accepted as an intractable problem, was eliminated from subway cars. In 1991, the new Chief of Transit Police, Bill Bratton, was the first to openly put Broken Windows to a real-world test. After a series of focus groups to gauge what riders considered to be the “broken windows” of the subway, Bratton announced a crackdown on fare evasion. Bratton believed that by cracking down on fare evasion and other petty crimes, and making a big public spectacle of its enforcement, felony crime would abate. And he was right.

Previously turnstile jumpers were simply counted by station agents. More than 170,000 jumpers were counted daily (though the real figure was much higher). One out of every seven people arrested for fare evasion had an open warrant for a prior crime, and one in 20 was carrying some sort of weapon.

By fixing broken windows and enforcing low-level crimes and quality-of-life issues on the subway, felonies began a steep decline, from over 15,000 felonies in 1990 to about 5,000 in 1996.

While statisticians like pointing out that correlation does not mean causation, the cop in me tells me that sometimes it does. The lawyer in me knows that circumstantial evidence supports Broken Windows, and the academic in me knows academic studies support this.

One key to fixing broken windows is to stop conflating broken windows with stop-and-frisk and zero tolerance. Those who most benefit from broken windows – those who need and want good policing – are those going to work who can’t afford to call an Uber.

The views and opinions expressed here are the author’s own and do not reflect those of any organization or employer.

Click here to read more essays from the Violence Reduction Project

Ari L. Maas is a police officer with over 17 years of experience in two departments in two states. He is a licensed attorney and has previously commanded the late-night train patrol, which has one of the largest daily deployments of transit cops in the country.