Better Governing Through Data

A new initiative by Comptroller Scott Stringer aims to save New York City money and solve problems the way CompStat has helped the N.Y.P.D. reduce crime.

Comments: 32

  1. If CompStat worked, speeding and horn blowing infractions and simple assaults would be showing up on it. They don't.

  2. "The city’s tree-pruning budget was cut sharply in 2010, and injury claims from fallen tree branches soared. "

    Does the City pay people who are injured or have cars damaged by potholes? Does the State of NY pay people who get hurt hiking in the Adirondacks where there actually are a lot of trees and constant downfall? A lot is left unanswered by this article, except that lawyers living high in the sky are splitting the spoils of $300,000,000 a year with their 50% share of legal awards.

  3. If the City owns property and the trees on it, doesn't take care of the trees, and invites the public onto the property then the City is responsible if harm results from its negligence. While there may be an expectation of possible injury while hiking in the Adirondacks, there should not be an expectation of injury from falling tree branches in city parks.

  4. The last I heard, people who suffer injury or loss because of a pothole can file claims against the city, but only if that pothole has previously been reported to the city. In response to that rule, insurance companies, which may be on the hook for claims that aren't found valid, pay for cars to cruise the streets and report every pothole.

  5. ClaimStat seems to believe that the only valuable data are negative data. Wouldn't it also be valuable to identify geographic patterns of satisfaction, well-being or pride with city programs and infra-structures or do citizens of all cultures as well as government agencies see positive things the same way? I don't think so.

  6. Unlike ClaimStat, which uses administrative data which is collected anyway, the type of information that you describe above would require conducting hugely expensive surveys. Do you have any thoughts on where to get the money to fund such surveys? Just injecting some pragmatism into your otherwise well-intentioned suggestion.

  7. Actually, if you read all they way down to the middle of the first page, you'll see:

    "The Comptroller will also publicly acknowledge agencies that
    have taken innovative steps to reduce claims costs by launching ClaimStat “Star” awards."

  8. I don't understand. You're wondering why a system that aggregates and analyses data about lawsuits doesn't also collect data about citizen quality of life? These are entirely different things.

  9. That's what I'm talking about! Moving into the 21st century information age of facts, truth, and transparency! Awesome............well............OK, the hands of politicians, it could be tweaked, twiddled, and bent, Oh well.

  10. The NYC comptroller's office should be congratulated on its approach to better determining the causes of the city's immense bloodletting of lawsuit payouts -- enough money to fully fund the budgets of most cities; enough money to pay for half of Chicago's police dept.

    I served as a juror years ago on a liability case against the city. My experience, while anecdotal, gave me the lasting impression that it was basically another entitlement for those lucky enough to have the vaguest perch on which to hang a claim. In this case, the plaintiff lost before us and the attorney defending the city, while he believed in the city's argument, was nevertheless as astounded as the plaintiff , because juries then (and possibly now) almost automatically ruled for the plaintiff against the city.

    I'm also reminded of 1986. In 1980, Wollman Rink, a skating rink in Central Park now known as Trump Skating Rink, was closed for repairs that were supposed to take 2 1/2 years. Six years and an absurdly blown budget later, the rink still was closed -- union issues, as I recall. Donald Trump promised then-mayor Ed Koch, given a free hand and using his own people, that he'd have the rink open in three months. Koch agreed. Sure enough, Trump had the rink open in three months, on budget.

    Before the Donald gets TOO much older, NYC should sic him on the WHOLE liabilities payout process, with the objective of cutting payouts in half.

    Betcha he'd cut that process a whole SET of new orifices.

  11. What a clear way of finding the problems in a city ! Follow the money !! Claims from trees falling and sewers overflowing usually mean needing to spend more money for repair and prevention of problems. Note: the Maine State Legislature only approved a seat belt law after an ER physician did the math: an auto accident victim without a seat belt cost much much more in medical costs than an auto accident victim wearing a seat belt. Very simple. But that was over 15 years before the tea party victories in 2010. Can politicians now understand that a low minimum wage and poor health care, including increases in unwanted births, means lots more taxpayer dollars spent on child protective services, special education, psychiatric care, and criminal justice?

  12. If a city develops the image of being an easy mark, there are no shortage of members of the ambulance chasing set who will right there, scratching at the door for more. Many claims, of course, will be reasonable, but the dubious ones rise in such a climate when lawyers, trolling for clients, figure out the city would rather pay than fight. That means our city "leaders" need to not just believe any allegation just because it's been made.

  13. The next step is to use a data-centric approach to some national problems. There should be a federal commission to collate and analyze information on health care delivery, for instance. There's an enormous amount of data out there but it's fragmented and disconnected from all the decision-making. This would be one of the hidden benefits of a single-payer system — a coherent health care delivery policy that is based on the most accurate data available.

    A tiny example: currently hospitals are rewarded by Medicare for minimizing their readmission rate, the goal being to push them to do better discharge planning. Because needed support services (VNA, adult day health programs, even meals on wheels, transportation to followup appointments, etc.) are stretched too thin, their solution is completely counterproductive. They keep people who enter via the ER on "observation status" for 3-4 days, even moving them upstairs to a floor that is identical to an inpatient unit, and because they are never officially admitted to the hospital then when they bounce back a week later it doesn't affect the "readmission rate." It's a completely rational solution for the hospital in the short run, as long as one ignores the bigger picture.

    Having data on the costs of this kind of thing and being able to use it to shift funding towards outpatient support services would save billions of dollars. The key is to have the data in one place and for policy makers to actually use it for decision-making.

  14. This is so interesting and so smart on the part of Stringer. I have been reading recently about Charles Merriam and Franklin Jameson, both at the University of Chicago at the turn of the century....both were proponents of using data to analyze public policy and history, many ways their advocacy gave shape to municipal reform in the early 20th century.
    An enlightened city must be built on a strong foundation of knowing from where the money comes and where it goes. Merriam published the first major review of municipal finances in 1907.

  15. The sad thing about all these "stat" initiatives is that they're news

    Modern effective businesses have been doing this for decades

  16. What is the essence being done here? Processing complex network relationship information with greater speed, accuracy, and power. That = better survival.
    We could do that with big data and relationships across geo, eco, bio, cultural, and tech networks. There's a 7,000-square-mile dead zone where the Mississippi Rivers enters the Gulf of Mexico. We could have bots patrolling the river, taking samples, reporting in real time, etc. Match this against code that delineates what the range of water quality should be.

    Human reach has exploded; instead of 300 men in the rain forests with stone axes, we have 10,000 with bull dozers, chain saws. The impact is huge, but our cultural coding mechanisms can't process that level of complex relationship information with sufficient speed, accuracy, and power.

    Our cultural genome is complexity inadequate. Agriculture, and additional accrued human knowledge, yielded the transition from simple hunter-gatherer social structures to the exponentially more complex city-state structures. This complexity increase rendered the primary cultural coding mechanisms of tribal structures — moral and religious codes — insufficient. To facilitate the navigation of the more complex information architecture of city-states, we added writing, legal, etiquette, and monetary codes.

    We need to do this again. The idea that monetary code could possibly calibrate relationship Value with accuracy at these levels of complexity is ludicrous. Imagination people!

  17. The more data the better may probably be helpful...but only when firmly grounded in empiric findings, in reality. Data at toxic levels may lull us into complacency or forego unexplored yet pertinent information (remember what happened with Greenspan's misreading of the economy with too much data?). And trying to save a penny now, just to have a huge payout later (in lawsuits for possible harm), is silly. Case in point, the police department; people are afraid of the police, as excessive force has become routine; a paradigm is urgently needed, a better educated force, grounded in trust within the community and with no automatic race-profiling and its adverse effects; and last I checked, we should be considered innocent until proven otherwise (and without coercion). The money lost could be better used for education, health and infrastructure, but of course this goes without saying.

  18. There is a culture of claim paying without asking too many questions. It seems there is such a deluge of claims, the city couldn't fight all the bogus ones as the additional manpower required would outweigh any financial benefit. But perhaps that is what's required to change the payout culture.

  19. You, sir are correct. But, what you are asking for would also require some reforms to "Tort Law". Sound simple doesn't it? Well guess what, changing Tort Law requires that the State Assembly CHANGE the tort laws. I,ll give you ONE guess which State Assemblyman (who is also a Lawyer) has, to my knowledge, at least three times refused to take the matter up. one=====two====three=====wait for it====four====five=====Okay its Sheldon Silver. That's right dear old Shelly the lawyer will not allow even a discussion regarding changing Tort Law.

  20. The City pays for many reasons: they can be liable; City employees may not be very good either at their jobs or as witnesses at deposition or at trial; jurors and judges look at the City as the "Deep Pocket" and so the City picks up the tab for joint and several liability situations; judges want to move their dockets, so they browbeat the lawyers and Comptrollers reps into settling cases; the Law Department is overworked and underpaid, but more telling gets little or no support in terms of responses or resources or respect from the various departments; and under the previous administration and this one, cases are settling for more than they are worth because of "political considerations".

    Tort Reform may help slightly, but the City needs to figure out what is it they want, and how much are they willing to pay to get it. It may turn out, that's it's more economical to continue things the way they are, than to spend lots of cash to fix what's wrong.

    For over 40 years, I've heard the same litany, and nothing changes. Stringer's "program" is one in a long line of panaceas. In case you're wondering, I've never worked for the City, but have observed how things operate in the Courthouses around town.

  21. "After adjusting for the crime rate, the report found that several precincts in the South Bronx and Central Brooklyn had far more claims filed against their officers than other precincts in the city. What does that mean? It’s hard to know, "
    Well if you cannot relate the data to any actual actions then what purpose is it other than to convince other people you are doing something.

  22. You ask, " What does that mean? It’s hard to know, " Personally, I do not think it is so hard to know. It can mean one of many things, one of those is that in those precincts the police tend to be, shall we say, over zealous in their pursuit of perfection!

  23. Larry Hoffman: Have you stopped to consider that a culture of entitlement in parts of those precincts could be at the root of these claims?

  24. NYPD has "reduced" the crime rate in the general population, but are committing more crimes themselves. Is that part of the "data?"

  25. your remark while possibly being true, is OFF the topic of the article.

  26. Considering that according to the Cities press releases the Police "CrimeStat" system has helped to reduce crime, it is worth, at least, the old college try. If it does not work, we lost only money ( like the city really cares as long as they have some taxpayers) if it does work, perhaps the Tax Payers might beable to get a small break out it.

  27. This approach only works when the governing body is willing to listen to factual evidence instead of dismissing it in favor of an ideology that they prefer to believe.

    Thank you, Scott Stringer. You serve us well.

  28. Management of government programs is woefully lacking.

    Nothing is gained by anyone trying to increase efficiency and reduce bureaucratic waste. so this is rarely done.

    When was the last time you read about the firing of unnecessary government employees?

  29. Pollitics, opinions and getting elected is the driver of policy. Facts? "Facts are stupid things" said the great conservative mythical hero Ronald Reagan. And he certainly lived the idea.
    Most of our congress ignores facts on global warming, immigration, income inequality and most other items and act on what is best for them.

  30. Another good outcome will be to identify which people and which law firms repeatedly are suing the city. I wonder what percentage of payouts go to fraudsters.

  31. As a software developer for the past 30 years, I'd advise people to keep their expectations of what data can do in check. Those cool stories about insights gleaned from data are merely the sizzle on the steak; you never hear about how you have to go to Wyoming and buy the cow, transport the cow, feed the cow, vaccinate the cow, house the cow, slaughter the cow, butcher the cow, and then finally, cook the steak before you smell that sizzle.

  32. In a way you are right. Data is an essential tool in marketing government.
    "the way CompStat, the fabled data-tracking program pioneered by the New York Police Department, reduces crime."
    In fact crime fell nationwide at the same time. Flogging good statistics to make it appear you are responsible for them is nothing other than disingenuous marketing.