Is the NYPD Reducing Traffic Violations? Hard to Say…


Continuing Streetsblog’s breakdown of the Mayor’s Management Report [PDF], we turn our attention this week from DOT to the NYPD section (page 119).

The MMR, if you’ll recall, is released every year as "a public report card on City services affecting the lives of New Yorkers." In addition to tracking felonies and violent crime, the NYPD section measures traffic violations — or, to be precise, how many tickets police hand out. As a performance metric, this leaves something to be desired, because it does not gauge the extent to which drivers actually break traffic laws.

"Cops really need to change the way they measure performance in terms of traffic infractions," says Paul Steely White of Transportation Alternatives. "They track the number of summonses, but what they don’t measure is the level of compliance."

By only counting tickets, NYPD cannot tell whether traffic violations are increasing or decreasing year-to-year, as the department does with felonies. Tracking reports of, say, grand larceny reflects the incidence of that type of crime fairly well, but tallying the number of speeding tickets issued reveals only a fraction of all speeding.

Counting summonses also gives the false impression that hazardous
traffic violations cannot be systematically addressed the same way the
department has drastically reduced violent crime.

"They could triple the number of summonses but that would have little impact on compliance," says White.

If the NYPD were to compile thorough data on compliance, a better case
could be made for greatly expanding automated enforcement measures such as red
light cameras. A 2001 study released by
the city comptroller’s office estimated that drivers burn red lights
1.23 million times every day in New York City. By comparison, the NYPD issued 874,929 summonses for hazardous moving violations in all of fiscal year 2007, according to this year’s MMR.

A table from this year’s MMR showing traffic deaths and enforcement data.

The NYPD took a step toward better measurement in 1998, when it launched TrafficStat, a system based on its highly successful CompStat accountability process. But TrafficStat focuses on hotspots where crashes tend to occur. It doesn’t capture systemwide data on violations.

"Once [compliance data] is ascertained, that’s only going to enforce the rationale for red light cameras," says White. He predicts that getting a better handle on the scale and scope of traffic violations will put pressure on Albany to act, since the NYPD won’t be able to crack down using manpower alone. "To get the political will to undertake an automated enforcement program, you need that data to show that this is citywide, every day."

NYPD has not yet responded to requests for comment and more information about how it gathers data on traffic violations.

Photo: Photo Gallery / Flickr

  • Dave H.

    Well, what is the best compliance metric and how would you go about compiling data using it?

  • Paul White


    for example, the NYPD could simply measure the rate of speed limit compliance on a given street. T.A. surveys indicate that on many streets, a majority of vehicles are in violation of the speed limit, and a significant % are exceeding the limit by 10 mph.

    increasing traditional summonses will help, but not as much as consistent automated enforcement and traffic calmed streets that are designed to be self enforcing.

    because there is zero data on compliance and the (weak) impact of summonses on compliance, the will to go beyond traditional summonsing is weak.

  • Komanoff

    While I appreciate Paul’s holistic approach, I’m uncomfortable with his statement, “They [NYPD] could triple the number of summonses but that would have little impact on compliance.”

    Wouldn’t it be better to argue for a “3-legged stool” approach of manual enforcement, automated enforcement, and traffic-calming design (w/ the latter including congestion pricing)?

  • ddartley

    Whatever other methods get used, A THIRTY MPH SPEED LIMIT IS OUT OF PLACE IN A CROWDED CITY.

    Yesterday, outside my window, (14th and 1st in Manhattan), I saw TWO collisions, one appeared to be car-on-ped.

    Local streets and avenues have to stop being treated as highways with the unlucky burden of a lot of traffic lights. That’s how most drivers treat them, and I always see cops ignoring it. How often do you see cops pulling people over for speeding on internal NYC streets? I don’t know if I ever have. And yet I have no trouble believing the surveys Paul cites; as long as there is room to zoom, it is what happens on every internal NYC street.

    In the 100% auto-reliant suburb where I grew up, the speed limit on local streets was 25mph.

    Where the f were the heads of NYC traffic engineers when they decided that the speed limit on narrow residential streets, with parking on both sides (for a kid to jump out from) should be 30mph?

    Once again (for the eighth time I guess), 30mph is the speed at which people IN cars start dying in collisions. That’s IN cars. New York City is saturated with people on the streets but NOT in cars.

    30 mph has GOT to end as the default local speed limit. Yeah, a lot of measures would have to accompany that, but that is one of the most important.

  • Paul White


    To build a case for building those three legs the NYPD has to move beyond its exclusive focus on summonses quotas (SQs) and expose the high rate of non compliance with basic traffic laws. My point was that SQs are input measures and no substitute for (output based) compliance data. Plus, SQ’s don’t work because they cause inconsistent enforcement. (cops will often meet their entire quota in one brief blitz, for example)

  • Paul White


    in recognition of your points, more and more european cities are reducing their limit to 30 km/h (18.6 mph) in dense city centers and residential areas.

    if such a limit were proposed for NYC, the cops and even the DOT may say that it is unenforceable. regardless of its ultimate enforceability or political viability, a 20 mph campaign– at least for residential streets– in NYC could do a lot of good.

  • ddartley

    Paul, I’m sure you’re right, and not only would the police call it unenforceable, but it actually would be hard to enforce.

    But too bad–I don’t think they do speed enforcement anyway.

    The speed limits should come down even if the only early likely effect is that the public knows they’ve come down. There should be an end to official sanctioning of a speed limit that’s deadly in its specific environment.

    Even just 25 would be an improvement! (And so would a few of those exotic things they call “speed limit signs.”)

  • 30 mph is (apparently) unenforceable too, but that doesn’t stop us from having it.

    I like to daydream about suing to close a street because of all of the unenforceable traffic laws that are broken on it. The judge would say, “Golly, this street is out of control — we’re going to have to close it!”

  • lee

    My favorite speed limit is at a shopping center in Douglaston where the posted limit is 2 1/2 mph.

    But yea, speed limit enforcement is terrible in this city. Punishing speeders with violations is a way to raise revenue. Redesigning streets to prevent speeding doesn’t.

  • Larry Littlefield

    (More and more european cities are reducing their limit to 30 km/h (18.6 mph) in dense city centers and residential areas. if such a limit were proposed for NYC, the cops and even the DOT may say that it is unenforceable.)

    It is enforcible on one-way streets with signal timing. Just set the signals to clear in succession, at 18 mph, and put a sign next to each traffic light announcing the fact.

    The “enforcement” would be the jerk in the SUV on the side street, ready to move out the moment the light turned green. And since the formal speed limit would not change, there would be no need to go to Albany for permission.

  • Ace

    Red light running (including illegal turns on red and those no-stop fast sweeping turns into a crosswalk), speeding, and parking violations could all be enforced through automated cameras.

    Big brother? Well to live outside the law you must be honest.

  • ddartley

    Larry, you know more about this than I do, and I should look this up before I flap my gums on it, but I don’t feel like doing that:

    I don’t think NYC needs Albany’s approval to change speed limits on its internal streets. NYC would need Albany’s okay to mess with the FDR, Henry Hudson, etc., but I don’t think so for, e.g., the grid. I could be wrong.

  • My source for this is something I read on Gridlock Sam a while ago but can’t find there any more, so correct me if I’m wrong, but my understanding is that the lights on the major avenues are timed at 35 mph even though the official limit is 30. So no wonder most people are speeding!

    The good news of that is that a newly traffic-calming-minded DOT could re-time the lights and slow everybody down without going through a 20mph campaign. Maybe?

  • HF

    A few points:
    1) NYC doesn’t need the state’s permission to change speed limit AFAIK
    2) AASHTO suggests streets be designed for a design speed 15% (if I recall) higher than the speed limit. WTF is up with that?
    3) If streets were designed for at max the speed limit, if not less, then enforcement wouldn’t be as big a problem – we need the design along with the enforcement (manual and automatic).
    4) I think a lot of European countries have higher speed limits on major streets and much lower speed limits on residential or otherwise non-through streets. Let’s do it.
    (Only downside is even more sign clutter to distinguish different speed zones)

  • Spud Spudly

    I definitely see the point that NYPD should be measuring compliance with traffic laws and not just how many summonses it issues. But it’s silly to compare it to felonies and say that NYPD accurately measures felonies in the city and therefore could do the same with traffic violations (traffic violations would in fact be much easier to track because they take place out in the open).

    NYPD doesn’t know if felonies are actually increasing or decreasing every year, it just knows how many are being reported. It’s not really out there actively detecting felonies that are otherwise unreported. And since nobody’s out there reporting traffic violations, it’s a silly comparison.

  • JK

    No, comparing traffic crimes to felonies is apt. The FBI conducts an annual national survey in which members of the public are interviewed to see if they have been crime victims. The survey is fine grained enough so the NYPD does have a pretty good idea of what the “real” crime rate is.

    Similarly, as you suggest, the PD/DOT could statistically sample light running, speeding, reckless driving, cell phone violations, failure to yield etc in representative survey areas, to determine the actual level of motoring misbehavior.

  • Spud Spudly

    Yeah, sure, and I’ve got a bridge to sell you. Nobody really knows what percentage of rapes, assaults and robberies are unreported. When one drug dealer robs another, is it reported to the NYPD? And is the victim ever interviewed by the FBI? How about for date rapes and incest? Is every street fight on the police blotter? Does every homeless person who gets jacked in the middle of the night call 911? Puh-leeeeeze. NYPD tracks the crimes reported to it and the FBI at best makes a semi-educated guess about the rest.

    Not that this diminishes the main idea that traffic violations could be assessed more accurately. It’s just an example of TA’s occasionally loopy communications style.

  • vnm

    I agree with Paul White’s suggestion in principle, but this is a lose-lose proposition for the cops. They’re not going to act on it unless there’s an upside for them.

    Under the proposed scenario, the NYPD spends a lot of money in order to appease its critics and provide future fodder for them. I can just seem them saying: “You’re going to make me spend all this money so you can say we’re not doing a good job?”

    Unless T.A. has some huge wellspring of political capital its built up with the cops that I don’t know about, I think someone else is going to have to do the study of lawbreaking, using random sampling techniques. The police aren’t going to do it.

  • Ed

    I think traffic-calming design is the way to go. Better enforcement is a stretch – we can’t have traffic cops on every street. And cameras really are too “Big Brother.” Better to design for slower speeds, and continue to seek tougher penalties.

    Spudly – Even if felony stats only count crimes reported, they are still a good measure of whether felonies are rising or declining. As opposed to say, just counting the number of felony convictions, which is essentially what counting traffic tickets does.

  • Dave H.

    Thanks Paul and Ben. This discussion is a tremendous help for some projects going on here.

  • Spud Spudly

    Yeah, you’re probably right Ed.

    vnm: There are traffic studies done in this city all the time by both public and private interests, and it’s doesn’t require a lot of interpretation to tell whether a traffic violation has been committed. So NYPD wouldn’t necessarily need to do it.

  • JK

    Let me try again. The FBI interviews a statistically robust sample of the population and asks them if they have been a crime victim, what type etc. This gives an overall picture of the incidence of crime across the population. Is every crime captured. Of course not. The survey describes overall picture and trends. Since crime is a big deal the FBI survey has been studies exhaustively, and compared to many smaller more detailed studies. It has some problems but generally holds up. Back to traffic safety, similarly nobody cares if every one of the million plus a day red lights run are counted. The point is gauging how common light running is, and whether there is more or less of it. As a number of folks suggest, this might be something that the DOT or an outside contractor hired by Mayor’s Operations is better equipped to measure than NYPD.

  • MaybeNot

    One of the premises running through the comments here is that speeding (above 30 MPH) is rampant on city streets. In my experience and observation, I don’t think this is true. It occurs, of course, but I don’t think it is the most common or most dangerous infraction on streets. Moreover, the cost of data acquisition and analysis vs. the potential benefits wouldn’t justify switching money from programs with much clearer benefits. Let’s reduce the ped and biker crashes caused by (usually legally) turning vehicles, the DUI related crashes, the red-light/stop-sign/failure to yield accidents. These all have much better established connections to accident rates than speeding, and are far more amenable to existing design and enforcement tactics (except maybe DUI).

    A further point on data — the level of enforcement resources has been steady for quite a while. I think that rates of enforcement actions (summons, arrests, etc.) in combination with accident report rates are a good enough proxy for trend purposes. I think trying to get to the “real” accident rate would be enormously expensive and have very little analytic value above what already exists.

  • Jim

    the speed enforcement issue could be enforced by the NYPD calibrating theit speedometers on their marked cars. The speedometers are calibrated from the factory but the calibration must be certified every few months to make sure they are accurate. This is only done on Highway patrol cars.

  • JK

    One thing the police and city should do is stop putting so much emphasis on fatalities as a surrogate for overall street safety — injuries would be a better indicator. Charlie Komanoff has done a good job reporting on a major Harvard study, and other research, which suggests that improved emergency medicine is responsible for about half of the last decade’s drop in deaths due to motor vehicle crashes and homicide. This said, a number of studies suggest that only 1/3rd to 1/4th of pedestrian injuries are reported.

  • 20 mph is the only safe speed limit for the nyc streets. The line of sight has changed since the limit was original implemented.
    Speed camera are available that are mobile and effective.
    i wish someone would start a tort about the dangerous high speed effect on pedestrian and bicycles.
    nyc lawmakers went to another state to sue a gun store for selling guns that were smuggled into nyc, why not a tort to lower the speed limit, while we are at , need a law suit to enable gas free travel on the state roadways.
    taxi and automobile industry is so threatened by a tiny motor on a pedicabs that they successfully lobbied against them.

  • mickey


    speeding is a good place to start because it is in fact rampant, and is a main factor in more than 1/3 of crashes.

  • Rob

    Speeding is epidemic, especially on outer boro residential side streets. With less traffic and no lights I regularly see people, especially livery cabs zooming around Astoria at 50+ mph.


How Meaningful Was NYPD’s Weekend Speeding Sweep?

A few thoughts on the NYPD “anti-speeding initiative” the department conducted last weekend, when police issued 736 citations to drivers across the five boroughs. First, it’s a good sign that NYPD seems to be responding to public pressure to tame speeding drivers, and that the department sees PR value in highlighting this enforcement effort. These speeding […]

NYPD Failing to Keep Kids Safe From Traffic at Bronx School

The frequently-ignored "One Way" sign at Briggs Avenue and East Moshulu Parkway. Image: NY1 A report from NY1’s Susan Jhun today describes the dangerous conditions at an intersection right next to P.S. 8 in the Norwood neighborhood of the Bronx, where parents and students constantly contend with cars darting the wrong way down the block. […]