Perverse Incentives

Russell Michaels

Russell is inside his own mind, a comfortable yet silly place. He is also on Twitter.

Related Post Roulette

34 Responses

  1. Chip Daniels says:

    “A government policy that is intended to alleviate poverty by granting more money to unmarried mothers leads to an explosion of out of wedlock births. “

    Twice, in the space of a week.Report

    • Oscar Gordon in reply to Chip Daniels says:

      And still no evidence that it’s actually a problem today (if it ever really was a significant problem).Report

      • Dark Matter in reply to Oscar Gordon says:

        I have had four relatives change their behavior because of these policies, i.e. not get married to collect money. Each was open as to what was driving their behavior, 3 of them did eventually get married anyway, 1 did not.

        One of them did this… I think 3 years ago. That was the scariest, he’s as good with numbers as I am and tried to walk me through them in detail. We’ve had a lot of engineer to engineer numbers conversations over the years, he’s an impressive guy.Report

  2. Jaybird says:

    Related:

    The City pays the money from the wrongful death lawsuit, huh? Who does the city get the money from? Oh.

    Which means that the police have no incentive to change stuff like no-knock raids or bad warrants or anything. They have no skin in the game.Report

    • CJColucci in reply to Jaybird says:

      If the money had to come from the cops, rather than the taxpayers, who pay the cops, the victims wouldn’t get it, or not much of it.Report

      • Jaybird in reply to CJColucci says:

        Oh, there are compromises available.

        Here, check this out:

        At issue is a provision within SB 217 that allows for officers to be sued personally and held liable for 5% of any judgment or settlement against them or $25,000, whichever is less. Wiped away was a long-standing provision of qualified immunity, which means an officer acting under the authority of a government couldn’t be sued personally.

        Personally, I think the lower cap should be higher but baby steps, baby steps, baby steps.

        At least it passed and wasn’t vetoed.Report

        • CJColucci in reply to Jaybird says:

          Small enough not to harm the victims, at least not much. Big enough to create incentives? Doubtful.Report

          • Jaybird in reply to CJColucci says:

            Well, the article mentions that a number of police resigned in response to the law passing. Or, “interviews with chiefs of police, union officials suggest a number of them are (because of the law)”.

            So it seems like there are *SOME* incentives kicking in already.

            If Chiefs of Police and Union Officials can be believed. (Which is up in the air.)

            I don’t know how much police make but an immediate $25,000 being due (even on a payment plan) would make my life worse. Even if I could get a second mortgage for that amount in an hour and pay it off, I’d still have to pay off my second mortgage.

            Going to a mortgage calculator… $25,000 in a 10 year mortgage at 4.5% interest… that’s $259 and change every month for 10 years.

            Small enough to say “eh, that’s not an incentive”? Personally, I think the lower cap should be higher but baby steps, baby steps, baby steps.Report

            • Oscar Gordon in reply to Jaybird says:

              Now I see that as an opportunity for a new insurance market.Report

              • Jaybird in reply to Oscar Gordon says:

                Sell 100 policies that cost 1.5% of the payout to 100 cops and only cash out one every year? You’re rolling in money and can do this indefinitely.

                Pay out two every year? You’re raising your rates to 2.5% or going out of business.

                Have a year where you pay out 7? You’ve shuttered after the fourth one and…

                Wait. That might be interesting to watch.

                (Maybe sell insurance insurance to the cop insurance companies?)Report

              • Oscar Gordon in reply to Jaybird says:

                I’m more than willing to let insurance rates weed out bad cops, especially if we can limit or eliminate QI.

                Just like bad drivers and bad doctors can soon find themselves priced out of that activity, so should cops. Especially since insurance companies won’t be giving into Union demands that they purge records every few years.Report

              • Jaybird in reply to Oscar Gordon says:

                I love it. Let’s do it.

                If you thought the cops were ruthless, you should see the insurance companies.Report

              • ozzzy! in reply to Jaybird says:

                The whole point of insurance is that the expensive one’s and two’s DON’T kill the canary when singled out for their specific expense.

                This reminds me, sort of, Technology People re-figuring out why markets, regulations, and people have organized their markets in a certain way since forever.
                But, it’s quite enjoyable to read, so please, lets walk down this path together (you two taking each step one by one, together).

                *meant with no malice, really!Report

              • Oscar Gordon in reply to ozzzy! says:

                Right, so your average, good cop who can avoid being violent, etc. has low, easy to manage premiums, and if a cop starts to build up a history of justifiable complaints that result in awards, they start to feel the pinch in the premiums, and either they learn to chill, or they become un-insurable.Report

              • But isn’t this the kind of thing that the union would do for its members? Like group health insurance?Report

              • Oscar Gordon in reply to Michael Cain says:

                Sure, and that’s fine, I guess, as long as it comes out of union dues, and is not being paid by the taxpayer.

                Even group rates will go up if you have enough ‘bad apples’. Also, the insurers may set an un-insurable criteria, so the cop either becomes un-insurable, or the Union is obligated to self insure.Report

              • Ozzzy! in reply to Oscar Gordon says:

                So there are individual reviews And exclusions for pre-existinG conditions that can somehow be determined? By what dataset (and if you can get that dataset, why go through the rigamarole of insurance to enforce social change)?Report

              • DavidTC in reply to Ozzzy! says:

                And exclusions for pre-existinG conditions that can somehow be determined?

                Uh, yes, by officer complaints? A small percentage of officers have large percentages of the offenses, which are directly correlated to the complaints large enough to get judgements against. It’s really not hard to figure out who the (very) bad cops are.

                And no one has to go review anything. Price everyone the same at the start. All insurance companies have to do is demand officer complaints get forwarded to them.

                They’ll just soon start raising rates appropriately.

                By what dataset (and if you can get that dataset, why go through the rigamarole of insurance to enforce social change)?

                Because assholes do not want us to be able to do this, because they either deny or do not understand the extent of the problem.

                If the ‘fire the bad cops’ solution was plausible, don’t you think we would have done it? Years ago?

                This is using their own nonsense back at them. No one wants to be on the hook for this, and saying ‘police officers should have to pay for it, but they can get insurance’…everyone see how that works. People doing risky things that could hurt others _often_ have to get insurance, that seems reasonable.

                What is being left unsaid here is…insurance companies rapidly would not even insure a large chunk of police forces, and might out opt out of certain forces _altogether_. There are some police forces that officers literally would be uninsurable at any reasonable rate.

                In fact…everyone seems to be assuming the insurance companies would actually create a product, for anyone. Are we _sure_ about that? Insurance companies usually dislike insuring _deliberate actions_ of people. Mistakes, like malpractice insurance, are not the same things as deliberate actions. I am not sure insurance companies are going to be particularly willing to say ‘If you shoot someone to death for no reason and get fined for that, we will cover your costs.’…I mean, speaking of perverse incentives.Report

              • Ozzzy! in reply to DavidTC says:

                This is the jump I am missing David. Please explain this part.

                “What is being left unsaid here is…insurance companies rapidly would not even insure a large chunk of police forces, and might out opt out of certain forces _altogether_. There are some police forces that officers literally would be uninsurable at any reasonable rate.”

                Why are they insurable now and not after whatever magically makes them uninsurable?Report

  3. DensityDuck says:

    “The biggest reason why these perverse incentives do not get fixed is because, usually, the bureaucrats and the supporters of the original policy will tie themselves into knots to avoid admitting the policy is the cause of the problem the perverse incentives encouraged.”

    There’s also the fact that the policies genuinely do help people, and that some people have in fact created lives that are dependent on the benefits the policies provide, and that just eighty-sixing the policies causes genuine trouble for those people. And simply saying “well they’ll just have to SORT THINGS OUT” is a response but not one that’s going to get a lot of traction.

    You can put in sunset clauses, “no new applicants” things, and that works, but you will have to deal with people saying “well why don’t I get those benefits, why don’t I get a slice of that cheese”; that said, it’s easier to justify hard-heartedness toward people who haven’t yet built lives on these benefits.Report

  4. Pinky says:

    “The biggest reason why these perverse incentives do not get fixed is because, usually, the bureaucrats and the supporters of the original policy will tie themselves into knots to avoid admitting the policy is the cause of the problem the perverse incentives encouraged.”

    That’s only one of the reasons. First of all, policies often have lagged effects, and even more lagged indirect effects. It takes a long time for corrective legislation to be proposed, and there’s hardly 100% agreement and trust even when it is. Since these are indirect impacts, they’re typically harder to measure. And as you’ve noted, the central system isn’t going to understand the impacts as well as the market will. So there’s no reason to assume bad faith on the part of the administrators.Report

    • Philip H in reply to Pinky says:

      So there’s no reason to assume bad faith on the part of the administrators.

      And yet we bureaucrats get tagged with all sorts of bad faith all the time, especially for implementing policies someone doesn’t like.Report

      • Jaybird in reply to Philip H says:

        I think that “if you do X, it won’t work” getting a response of “yes it will!” and then it not working, enough times, can change the nature of the tagging.

        The first time? Sure. No need to assume bad faith. It’s probably bad faith to assume bad faith.

        The umpteenth time? Well… maybe we should start listening to the naysayers.Report

        • Philip H in reply to Jaybird says:

          well if that’s what I was talking about, you MIGHT have a point.

          But its not. I”m talking about Congress X passing Law Y, Signed by President Z that tells Agency A to do a thing. Agency A does that thing – since its a law – and then is promptly accused of Waste, Fraud and Abuse by those on the opposite side of the political aisle from either or both of the Congressional majority and the President.

          All the damn time.Report

          • Jaybird in reply to Philip H says:

            I imagine that that’s less of a problem with a law that has passed and been signed than with, say, an executive order.

            After all, if Agency A just says “hey, we’re following OSHA”, they can get away with that.Report

            • Philip H in reply to Jaybird says:

              And you’d be wrong. Its generally worse with statutes because someone in Congress lost a vote and gets all butt hurt about it.

              Your other example is also non-sensical as following the regulatory mandates of other agencies is a legal requirement for feds. We don’t get to pot out of that stuff unless the statute says so, and we are often easier to punish as we can’t deploy our armies of lawyers to fend off inspection reports.Report

              • Jaybird in reply to Philip H says:

                I’m not denying that congressman whatshisnuts will grandstand.

                I’m denying that the defense of “hey, we’re just following the law” will insufficiently pass the buck.

                I know that it’s pretty much impossible to get a law to sunset, let alone be repealed. But saying “we’re just following the law” does a good job of changing the argument to “well, that shouldn’t be the law!”

                Leaving only the grandstanders to keep yelling “can you believe that they’re following the law?!?”Report

  5. Obamacare was debated for months. Then there’s this one:

    https://en.wikipedia.org/wiki/Tax_Cuts_and_Jobs_Act_of_2017#Procedural_concerns

    The 400-page House bill was passed two weeks after the legislation was first released, “without a single hearing” held. In the Senate, the final version of the bill did not receive a public hearing, “was largely crafted behind closed doors, and was released just ahead of the final vote.” Republicans rewrote major portions of tax bill just hours before the floor vote, making major changes in order to win the votes of several Republican holdouts. Many last-minute changes were handwritten on earlier drafts of the bill. The revisions appeared “first in the lobbying shops of K Street, which sent back copies to some Democrats in the Senate, who took to social media to protest being asked to vote in a matter of hours on a bill that had yet to be shared with them directly.”

    Report

  6. Kazzy says:

    “Information differences tie into everything people do in a capitalistic economy. How an individual values a good or service being the most prominent. Two people go to the store. One values the premium salad dressing at $10.00, so sees the $8.00 price tag as worth it. Another values it at $5.00, so instead buys the Kraft brand for about $4.00 instead. This is also situation specific. Say you go for a walk, one in 90 degree weather and another in 40 degree weather. Your valuation of a bottle of water could be multitudes higher after that 90 degree walk. You’ll pay the $5.00 for the water from the hot dog cart when you likely would not after the 40 degree one.”

    Are these really differences in information? Or differences in who values what and how much?

    As for perverse incentives, the Freakonomics guys highlighted one of the most interesting and obvious ones while studying a daycare. The daycare closed at a particular time and you were expected to pickup your child by then. If you were late, you were met by a frustrated teacher staying after hours to watch your kids. To cut down on this, the daycare enacted a fairly common “solution”, charging a late-fee if you showed up after closing. The effect? Late pickups went UP. The belief was that when the cost to parents for their tardiness was guilt at keeping their child’s teacher after hours unpaid, most folks weren’t willing to pay that. But when it was a few bucks? Easy-peasy.

    Oddly, this worked both ways. I worked at a daycare with such a policy and typically had the late shift. If a parent was late, it was $5 plus $1/minute for every minute after 5. As the person staying back, I got the cash. I was earning probably like $15/hour? Maybe more? I can’t remember. But $5 for 5 minute? Or even 2 minutes?! That was like triple my rate! And $5 to a 22-year-old is a couple beers at the dive bar. So I didn’t exactly meet the parents with my grumpy face if they were late.Report

    • Aaron David in reply to Kazzy says:

      The difference in information about the salad dressings comes from a different place; each purchaser has information about what quality/price is acceptable to them. But only they have that specific information. In other words, they all have a difference in information.

      It isn’t the best way to phrase it, but I think that is what he is saying.Report

      • Philip H in reply to Aaron David says:

        Given your further description I can agree that is what he’s saying – but its not how the average person would read that phrase.

        And sadly that’s an information difference type that we can’t really overcome.Report

    • Russell Michaels in reply to Kazzy says:

      Yes. Value is subjective. How is that hard to understand?Report

  7. Anthony says:

    Perverse incentives can erode social norms pretty quickly. At the time, people said that welfare for unmarried mothers would go from being young widows (many of whom were soldiers’ widows) to causing an explosion in bastardy. Proponents said that would never happen. It *did* take several decades to erode that social norm to zero among the poor.Report