Science Talk

The City That Became Safe: What New York Teaches about Urban Crime and Its Control

U.C. Berkeley School of Law professor Franklin Zimring talks about his article, "How New York Beat Crime," in the August issue of Scientific American

U.C. Berkeley School of Law professor Franklin Zimring talks about his article, "How New York Beat Crime," in the August issue of Scientific American.

Podcast Transcription

Steve:          Welcome to the Scientific American podcast, Science Talk, posted on August 9th, 2011. I'm Steve Mirsky. This week on the podcast:

Zimring:          When everybody's crime was going down, New York's crime was going down more. When everybody else's crime stopped going down, New York continued to have declines.

Steve:          That's Franklin Zimring. He's a professor at the University of California, Berkeley School of Law and he studies the use of empirical research to inform legal policy; and he's the author of an article in the August issue of Scientific American titled, "How New York Beat Crime", about the reasons for the huge drops in crime over the last couple of decades in America’s largest city. He was in New York recently and visited the Scientific American offices, where he I and his editor, Davide Castelvecchi, sat down for a chat about his article.

Steve:          So, what does New York teach about urban crime and its control?

Zimring:          Well, in the first instance, what happens is that the reason New York is singled out is that it just found its way into the Guinness Book of World Records; that this, the largest city in the United States, has also had the biggest drop in violent and safety crime that anybody has ever reliably documented in any city.

Steve:          Violent and safety crime being murder?

Zimring:          Robbery, burglary, auto theft. And the statistics, if we were to do just the size of the decline, the homicide rate from its 1990 peak, went down 82 percent. The robbery rate went down 84 percent. The rape rate is down by more than three quarters, and we're not sure by how much more than three quarters; burglary 85 percent. And if you take the 1990 auto theft rate in the city of New York, the 2009 auto theft rate is 7 percent of what it was in 1990.

Steve:          So it's down 93 percent.

Zimring:          It's down and these are absolutely unprecedented.

Steve:          Now, some people were concerned because the police who would get the credit for any rates in that drop are also the ones who compiled the stats, but you have evidence that the stats are real from external sources.

Zimring:          Absolutely. First of all, that's a wonderful concern to have because the people who are keeping score, the umpires, are also the coach of one of the little league teams here. But the point is that homicide down 82 percent also is a medical examiner statistic that you can check against, and it works beautifully. Robbery, you can check, against robbery killings and it turns out that robbery killings have declined more than the general robbery rate. And auto theft is a situation where you have car insurance claims for auto theft and they’ve gone down by the same magnitude at the same times. You also have for a New York Metropolitan area, which isn't just a city, but which is mostly the city, victim surveys that show that both the robbery and burglary victimizations that people are talking about in those surveys went down just as much and at the same times as the official statistics. So in this dimension, all of the problems with police reporting are still there, but they haven't gotten in the way and the official story is the true story.

Steve:          Now its important to recognize that there was a nationwide decrease in the first decade of the last 20 years that we're looking at, but that was much smaller than the total decrease that New York saw.

Zimring:          Yes, but it is a very important complication. Between 1991 and the year 2000, crime dropped everywhere in the United States, and it dropped rather mysteriously in the United States. There were a lot of different theories of causes but no good science that shows anything principally that would be a persuasive explanation. The general size of that decline was a little tiny bit under 40 percent. But that then becomes a very important tailwind for the New York crime decline. Because that means that about half of everything that happened in New York City was happening everywhere, and that you can't very well say the New York police weren't happening everywhere; so that half of the crime decline probably had nothing to do with New York's specific changes, which were changes in policing citywide and changes in population for only one of the four big boroughs, for Manhattan. But that tailwind is part of the Guinness Book of World Record stories. Having said that…

Steve:          When you say Guinness Book, do you mean literally Guinness Book of Records…?

Zimring:          I mean I'm ready to submit it for the next edition.

Steve:          Oh, okay. I thought it was a recognized Guinness record.

Zimring:          No, no. It's a metaphorical academic.

Steve:          But clearly it's really, I mean, I've lived in New York City all my life and I remember the '70s as being a scary time here, and still into late '80s and 1990, we were very leery about late night subway rides, things like that, and now it feels like you are in a whole different city.

Zimring:          It's a different city. Now, what we know about the size of the New York decline, comparing it to other places, it's two rules of thumb. New York's decline lasted twice as long; the declined generally ended in 2000 in the United States; and the New York decline was twice as big. And that's enormously important. It's not just bigger than the rest of the country during this period of time, it's bigger than anything we've ever seen. But I want to go back to your conversation about the '70s because there are two different explanations we could have in the science of crime drops of this kind. One of them is to say let's take 42nd Street. The crime drop in 42nd Street around Times Square is more than 90 percent, but of course, Times Square is vastly different now.

Steve:          Right.

Zimring:          It used to be X-rated, now it's G-rated. It used to have mud wrestling, it now has Disney. So the fact that you had all those fabulous structural changes means that it's almost uninteresting that crime has disappeared as a problem there. But now let's talk about Canal Street, down in a different part of Manhattan, down in southern Manhattan.

Steve:          Where we are right now, by the way.

Zimring:          That's right. Now Canal Street hasn't changed enormously, in the population or social structure. You're still getting the same $5 dollar watches, except now they want $10 for them. The place looks very similar, and yet crime is also down by 80 and 90 percent on Canal Street. And that is very much different scientifically, because what that tells you is that the huge changes are in crime, but not in the nature of the city, and that turns out to be a much more typical picture of what happened in New York than the 42nd Street transformation. The biggest change that happened in New York over the last 20 years, wasn't the factors that changed crime, it was in crime itself. The thing, if you ask citizens what's different about this city, they will say it's a much safer city and they'll be right, but it wasn't that the population changed—where did all the criminals go? Well, it turns out they didn't leave for New Jersey. There were some important changes in police and policing and there were some population changes in only one borough, Manhattan— the number of African-Americans living in Manhattan went down 30 percent in 1990s—and the reason it did is that they got, the poor people got priced out of the borough of Manhattan. And all of those if we only had Manhattan crime data, would be important constraints on looking at things like policing. But 80 percent of New York's population is in its other boroughs. And when you go to the Bronx and Brooklyn and Queens, what you find is that populations is particularly high-risk and less economically advantaged populations haven't changed. And crime numbers are down almost as much as in Manhattan.

Castelvecchi:         You know, everyone knows Times Square has been cleaned up and the natural thing to guess could be, yes, okay, it's been cleaned up, but maybe it's just that all that crime that was concentrated in that area may be displaced and now all the prostitution and everything that comes with it is just happening somewhere else.

Zimring:          Somewhere else and therefore, the crime has also been exported. That's a fundamental assumption that we've always made about crime and criminals. It's been driving a lot of policy in the United States, and it's one of the reasons why we always doubted—and when I say we, I'm talking about professional social science—that police could make a big difference in crime rates. The reason that we doubted that cops would be a major preventive element of crime, is that cops can't be everywhere, and cops can't be there forever. So, what we would think is if you send two cops to 125th Street, that may mean there'll be no crime in 125th Street, but the criminals will go to 134th Street. Or if you put the cops there on Tuesday, that might be fine for Tuesday, but by the end of the week, there'll be just as many robberies.

Steve:          There was this assumption of conservation of crime, that there would be always be the same amount of crime, it would just move from time to time or place to place.

Zimring:          Yes, but indeed the late President Nixon put it that the chief cause of crime is criminals, and unless you can affect the number of criminals or lock them up, you're not going to have any real dents in the amount of crime.

Castelvecchi:         So the assumption was there are some people who are just inherently evil and who are going to commit crimes no matter what, and the only way to avoid that is to lock them up.

Zimring:          That's right, and if that were the case, then everything that we've been discussing in New York couldn't have happened. Because what you have is that New York has reduced safety crime by more than four-fifths. It hasn't had major changes in population. It is locking up 28 percent fewer people in 2010 than it was in 1990. So under those circumstances, the persistent and predictable criminality would have to be here, and what we now know—and this is not a matter of complicated math; this is simple arithmetic—is that it isn't here. So that fundamentally, the New York story tells us that major assumptions that we've been making, not just in New York but all over the United States and the western world, are wrong. And that the decisions to commit crimes and the things that influence crime rates are much more—if I can use a couple of buzz words—situational and contingent than we thought. And what that means is that if you can reduce the robbery rate on Tuesday, you've probably reduced the robbery rate this year. And if you can make a robbery that was going to happen on 125th Street, not happen there, it probably won't happen. And it is that kind of vulnerability of crime and danger to situational and contingent and temporary solutions, which is I think both the most important and the most hopeful news that comes from the New York story. But quite frankly, we haven't been noticing how basic the change is or how important it is for changing our policies. And we've had a similar kind of enormous insight about drugs. The big fight in the United States between 1985 and the mid-1990s—the period of the war on drugs and the drug/crack panics—was between what I call the legalists, who say, "Look the only way you're going to reduce the violence and social disorganization and crime that is associated with drugs is to reduce the use of hard drugs"; and the so-called harm reduction school, which said, "No look, let's just look at what the biggest problems created by drugs are." If we're worried about HIV transmission, then let's have needle sharing, for instance. If what you're worried about is drug violence, let's find out where that comes from and rather than going tooth and nail after all drug use. Well, let me tell you two things I think I can prove on the statistics of drug enforcement in New York City. One of them is that the police department never had a high priority of eliminating all illegal drug use. New York City has been the capital of heroin abuse for the last 80 years in the United States. It was a major center and still is a major center of cocaine use. But what the New York City police department did was go after two things tooth and nail. One of them were public drug markets—these were streets where ordinary citizens were not safe to be because they had become outdoor drug bourses. In the first 10 years of the new regime, really from 1994 to about 1999, the police had a 137 percent increase in the narcotic squad, and its principal job was to eliminate those public drug markets. This is…

Steve:          This is during the Dinkins then Ray Kelly.

Zimring:          That's right. This is Dinkins followed by Bratton, followed by Kelly and…

Steve:          Those are our police commissioners, Bratton and Kelly, by the way.

Castelvecchi:         Dinkins?

Steve:          Dinkins was the mayor before them.

Zimring:          And Lee Brown was the police commissioner who was matched with… okay, and so all of a sudden the number of people on the police department interested in narcotics went hugely up in the '90s, and drug violence went way down.

Steve:          But drug use did not go down.

Zimring:          Ah! That is exactly the point. Drug violence is down over the last 20 years by more than 90 percent. Drug use is quite steady, if you're talking about heroin. It went down for young people late in the 1990s and then went back up a little bit as a general population figure; so that you have a tremendously important contrast between steady rates of illegal drug use as measured by drug overdose death rates, as measured by urines of arrested people, emergency room mentions, all kinds of things. So you have drug violence going down and drug use staying stable, and you also have the police making an important decision in this environment. The largest increase in the 1990s of police in any department were narcotics. Between 1999 and 2006, all of the extra cops that were associated with narcotics disappeared. They were, in 2006, fewer officers in the narcotics squad in New York City than were on that squad in 1990. And this is very important because it proves that those police commissioners were getting exactly what their priorities were—that the reduction of violence and the reduction of the public lack of safety that came from the public drug markets. Now why did drug violence go down by more than 90 percent? Well if I'm selling drugs illegally out on the street, and you're selling drugs illegally out on the street, we’re both going to compete for the best corner. And if each of us is also carrying a loaded gun, that's going to become a drug killing. If I'm selling drugs from my apartment, and you're selling drugs from your apartment, where's the turf battle? So this is a situation where the implications of harm reduction are greatly different and much easier to achieve than an all-out war on drugs. Now, what I think the New York story shows is a triumphant demonstration of harm reduction as a policing priority by a police department which is supposed to be conservative and hard line.

Castelvecchi:         What is the definition of harm reduction?

Zimring:          It is saying that we are going to take as a priority not the number of people taking drugs, but the number of people dying as a result of drug violence. The streets in New York that aren't safe for other people to walk on because of drug sales and drug violence. And I would say the needle sharing example, HIV transmissions that happens because of the use and reuse of syringes. It's taking those priorities and trying to solve those problems, instead of saying it's all or nothing. We're either going to eliminate drug use or we'll lose the war.

 Castelvecchi:         So it's harm reduction in contrast to zero tolerance.

Zimring:          Exactly. And of course, zero tolerance is one of the wonderful myths of New York City policing. There are a lot of things that I say in my analysis in the Scientific American article that haven't been established as crime reduction in New York. One of them is all the stopping and frisking that's taking place as part of otherwise quite effective policing. We know that it's part of effective police maneuvers, but we don't know that it adds value. We don't know whether we couldn't get away with hot spots lite just as much as the heaviness. But then when I come to the question of so-called broken window or order maintenance or zero-tolerance policing.

Steve:          Which got such press, I mean, the idea being that if a neighborhood had a broken window, it instilled a sense of lawlessness among the residents of the neighborhood that led to a breakdown of civilization; but if you fix the window, then everybody was on board with the idea that we're all in this together, and we are all going to live in a peaceful society.

Zimring:          That's a wonderful demonstration of it. And the people who wrote this wonderful article, James Q. Wilson and George L. Kelling in 1982, then said, "By the way, where you are going to concentrate this effort is not going to be in the places with the very highest crime rates; they're hopeless. But its going to be in the marginal neighborhoods, the ones that are on the edge."

Steve:          Which could go either way.

Zimring:          That's right. And so one of the great urban legends is this is what the New York City police did. From day one this isn't what the New York City police did.

Steve:          The big things were the public drug markets and what else about the policing?

Zimring:          The next big thing that obviously worked was so-called hotspots policing—and this is we haven't changed the subject from order maintenance. What hotspots policing tells you is that crime is intensely concentrated, and it's concentrated in places where it keeps happening. And so what the New York City police did with this new police manpower and with all their aggressiveness, is they go to the places where the crime rates are highest and they go there with maximum force and for long periods of time. Rather than considering those as hopeless neighborhoods, they are the primary priority.

Steve:          And that crime that didn't happen in the hotspot doesn't go somewhere else.

Zimring:          That's exactly right.

Castelvecchi:         It’s not as if criminals have a to-do list, a quota, "I'm going to commit this number of crimes per month." If they don't commit crimes that month, they won't necessarily catch up the next month.

Zimring:          If they don't commit crimes that day, they probably won't commit those crimes. It turns out that situational and contingent: depending on who I meet today and what they want to do; depending on how good it looks to commit a crime instead of smoke a joint or have some wine; or doing a burglary or listening to some music. Most of the serious and violent crime in the United States is not the way that people make a living, and it is not deeply motivated, it's superficially motivated. And the good news about that is that if it's superficially motivated, then little changes in environments can make very big changes in safety crime.

Castelvecchi:         So, what can other cities in the world learn from New York's experience? What can the São Paulos and the Manilas and the Cape Towns of the world do to bring down their crime rates?

Zimring:          Okay, first of all, police matter. And also police should be concentrated in the areas where the priority crimes are the heaviest—hotspots are a proven success. Secondly, that police, and anything which assumes the variability of crime in the short term are probably better investments of fixed resources than incapacitation and imprisonment; that the unpredictability and variability and situational focus that New York crime has taught us, suggest that the investments we make in preventing crime on Tuesday and not sending people up for 150 years of expensive state custody are probably going to have higher payoff. I think, thirdly, there is the lesson that harm-prevention strategies, particularly if they focus on violence, are probably more cost effective than generalized wars on drugs. And that again is going to focus more on city streets than on building prisons in Ossining.


Steve:          Franklin Zimring's article, "How New York Beat Crime", is available in its entirety free for a limited time on our Web site for anyone smart enough to be a fan of Scientific American on Facebook. We'll get back to our interviews with Nobelists from Lindau soon. In the meantime, get your science news at our Web site,, it'd be a crime if you didn't. Hey! Follow us on Twitter, where you'll get a tweet each time a new article hits the Web site. Our Twitter name is @SciAm. For Science Talk, I am Steve Mirsky. Thanks for clicking on us.

Share this Article:


You must sign in or register as a member to submit a comment.

Give a Gift &
Get a Gift - Free!

Give a 1 year subscription
as low as $14.99

Subscribe Now! >


Email this Article