Helpful Inequality

We find significant and sizeable negative peer effects arising from students at the very bottom of the ability distribution, but little evidence that the average peer quality and the very top peers significantly affect pupils’ academic achievements.

More here.  Thus we’d probably do better to isolate the worst kids in their own school or work hell; they’d be worse off but the gains to other students would more than compensate.  At the other end of the status spectrum, the number of new businesses we get seems limited by the number of folks personally wealthy enough to start new businesses.  So having more really rich folks benefits everyone via innovation.  Details here:

Since richer entrepreneurs make larger investments and expect to have more wealth in the future, it is the relatively poor entrepreneurs who decide to take more risk and would be more likely to exit from business in the future. As a result, the model predicts that survival of entrepreneurial business is positively related to entrepreneurial assets, which is consistent with empirical findings. … Since agents enter entrepreneurship with relatively low wealth levels, our model also implies that young businesses exhibit lower survival rates, and, conditional on survival, small (younger) firms grow faster than larger (older) ones. All these implications are in line with strong empirical evidence from the literature on firm dynamics.

I’m not saying these are the only issues for how much inequality we want, but they seem to me neglected issues.

GD Star Rating
Tagged as:
Trackback URL:
  • Doug S.

    Thus we’d probably do better to isolate the worst kids in their own school or work hell; they’d be worse off but the gains to other students would more than compensate.

    We already do. It’s called “special education”.

    • gwern

      Obviously they don’t do that enough in England; I suspect from what I’ve seen of American highschools, they don’t do enough either.

  • Jeffrey Soreff

    At the other end of the status spectrum, the number of new businesses we get seems limited by the number of folks personally wealthy enough to start new businesses. So having more really rich folks benefits everyone via innovation.

    Interesting point! This sounds analogous to the way a small tail of high energy particles in a Maxwellian distribution can dominate chemical reaction rates. What does P(business_formation|personal wealth) look like? IIRC the wealth distribution looks like a Zipf power law distribution.

    Also, is there any data that can be used to calculate the positive externality of firms’ innovation as a function of entrepreneurial assets?

    • It sounds more like we want lots more moderately rich folk, not a few more really rich folk.

      • In my world someone who can spend a million on a new business is really rich; perhaps you travel in different circles than I. 🙂

      • Michael Turner

        Schumpeter believed that thrift was an underappreciated factor in entrepreneurialism — that it was not a bad way to build one’s own “angel”-scale funding before you’re too old to consider starting a business, and not seen enough on the radar of finance to be taken seriously.

        One does not have to be really rich to start a business and be successful. A healthy appreciation of risk and a nose for what people will buy are far more important.

        Habits of thrift can help you limit what’s often a very dire risk for new businessess: personal overinvestment in the early stages. An old startup nostrum has it that you should start on your kitchen table, and when the business starts to grow, move to … your garage.

        There are a lot of people who are moderately wealthy (in the sense that they could live, modestly at least, on the yield from relatively safe investments, if they chose.) Many of them have gotten to the point of having a million dollars in readily liquifiable assets that they could (potentially) sink into a business simply by being a two-career couple, reasonably successful in some middle-class professions, while clipping coupons and kicking the tires at used-car lots, and managing their savings well. Of course, it might help if they inherited a house from their parents that they couldn’t make much use of themselves. But that also describes a lot of people.

      • gwern

        Robin: if someone with a million to invest is rich (let’s assume they have 10 million net), and being richer than that doesn’t help innovation and growth, then isn’t this an argument for breaking up all fortunes over that, such as the ~$70 billion Walton fortune? That’s thousands of new rich-people/investors, after all.

      • Granite26


        Being richer than 10 million net is the reason people do the investing in the first place.

        In a non-Libertarian view, the reason they’re there is to serve as a carrot for everybody else, not because they are useful in an of themselves. (Same as why Tim Hartford’s CEOs get paid dumb amounts… Spur on the Veeps with hopes of advancement)

  • gwern

    There’s a non-gatewayed version of the paper here:

  • In “The Wire” that’s what Bunny Colvin tries to do in a Baltimore middle-school after he’s booted from the police for creating a “Red Light District”. It’s denounced as “tracking” and stopped.

    Alexis de Tocqueville and other 18th/19th century aristocrats also spoke of some of the downsides of equality, but we don’t hear that message so often now.

  • gwern, thanks, I’ve switched to that link.

    TGGP, yes I’d forgotten about Bunny’s school venture.

  • “We already do. It’s called “special education”.”

    In New Jersey, at least, we spend twice as much per special-ed pupil than on regular education.

    • Captain Oblivious

      Not trying to claim that the special-ed students aren’t deserving, but I have to point out that they (almost certainly) won’t be the ones inventing new technologies to help solve our energy problems, global warming, etc… it’ll be the on-level or “gifted” children who do that… so maybe we ought to focus our limit resources there?

  • justin

    We’ve still got to make it attractive to the “smart ones” to invent new technologies to solve our energy problems, global warming, etc., if they care enough to do that (if such activities are lucrative, attractive, status providing, and allowed to gain in supply given already strong contrary interests virtually everywhere else..)… etc.

  • Tristan

    The problem with separation is that while it does benefit the top kids, it denies the special-ed kids the possibility to become top kids later. The other problem with creating a special education ghetto is that you just raise the floor, creating a new class of “normal” kids who used to be in the middle and are now at the bottom relative to their peers. So negative-peer effects (aka, class-clowning, slacking, and bullying) arise from them, and they get pushed down into special education, too.

    The problem then is that no matter how much wealth or “positive peer effects” arises from the firms/kids at the top, it matters less when there is a huge pool of people labeled failures, who, because they’re surrounded by bullies, slackers, etc., will have a huge problem getting out of that ghetto.

    I say, if you’re going to separate out the bottom kids, reserve it for the really grievous negative-peer-effect producers, i.e. the bullies. The slackers who just throw paper airplanes or whatever can be dealt with via normal disciplinary methods.

    • Ben

      The “special-ed kids” are much less likely to be highly productive. IQ during youth is highly predictive of adult IQ, and IQ is a big driver of economic productivity.

      In order that as few bright kids as possible get stuck in the ghetto, they should be IQ-tested early and separated accordingly.

      Robin’s point is that, however bad it is for the kids at the bottom, they could be more than adequately compensated with the additional wealth created by the more efficient set-up, so paragraph two is a bit of a non sequitur unless you can back it up with a study.

  • quanticle

    In a Rawlsian sense, such a system would be unjust. If people started in a Rawlsian original position, they would not accede to such a system of segregation, because the veil of ignorance would not allow them to choose whether they are amongst the few consigned to that special “educational hell”. Therefore, no rational individual would accede a major social institution that performed such segregation.

    As Rawls states, “The first principle of a social institution is justice.” Without justice, considerations of efficiency are meaningless. Its like discussing the elegance of a scientific theory that has been proven to be inconsistent with observation.

    • gwern

      If there are grades or any sort of assessment which are used to decide admissions, then isn’t that system segregating?

      Every system will perform some sort of segregation and discrimination, as long as resources are not equal to desire – that is one of the classic definitions of economics, isn’t it? The allocation of finite resources. What social institutions could not segregate in that sense?

      If they all do, then it is Rawls’s theory that is inconsistent with observations; if some such segregation be ‘justice’ and others not be, the problem is only pushed back – why is the ‘academic hell’ any worse than special ed, or remedial classes or any of the current destinations for those disruptive or really bad at classes?

  • MZ

    Thus we’d probably do better to isolate the worst kids in their own school or work hell

    Many school systems around the world start stratifying students by the age of 14 or 15, some going directly into the work force, some going to vocational schools, and others going to college preparatory schools. I think I would have been better off in high school if I wasn’t required to take electives in art, home economics, and wood shop. I’ve never used that “knowledge” as I don’t even remember it.

  • Mendel Schmiedekamp

    The education article is in favor of tracking – after all much of what it is talking about is optimizing the average, not raising the top. The educators see the 5% gifted kids as a special education class that money can be saved on, by combining with the remaining 90%, and they want to increase the overall result for that 90%.

    What I find fascinating is that for the females this does work (perhaps because it helps over-ride the noted tendency of women to adopt an unintelligent role around puberty?), but for the higher performing, but second tier males it causes them to under-perform (perhaps due to competitive pressure? Or maybe it is further sexual politics at work?). Naively the conclusion might be to throw political correctness away and have a male-only honors program, but it’s not clear that this would actually work.

    In either case, I couldn’t tell whether the bottom 5% was before or after major learning disabilities were accounted. Even if not, from the special ed teachers I know, those students take a great deal of time, so by removing the typical tracking I suspect that is one theory that could explain the results.

    And having worked with some of these kids before, I’ve concluded that nearly all of the clever kids in any school are split between the gifted and the special ed.

  • michael vassar

    Adam Smith made the point about needing lots of moderately rich people in Wealth of Nations.

    A person with three million dollars can spend a million on a business. America has millions of such people and they aren’t generally considered really rich.

    My guess is that the optimal entrepreneurial wealth level is about $10M. I think that’s roughly where the Gates and Buffet families were coming from, and its fairly close to the old meaning of “millionaire”.

  • Philo

    You don’t need to be rich yourself if you can persuade a rich person (or a banker, or a manager of a venture capital fund) to lend you the money.

  • Seven Years Of College Down The Drain (7YrsDD)

    Robin: if someone with a million to invest is rich (let’s assume they have 10 million net), and being richer than that doesn’t help innovation and growth, then isn’t this an argument for breaking up all fortunes over that, such as the ~$70 billion Walton fortune? That’s thousands of new rich-people/investors, after all.

    gwern, I’m not sure how the thousands of new RP/Is follows at all from breaking up the fortunes > $10 million NW. Even at the per-household not per-capita level, I don’t see how the per-household number gets to even $10,000 per household. (Assuming for the sake of argument $1 trillion in aggregate worth in fortunes > $10 million and only 125 million households to receive proceeds or shares.)

    • gwern

      7YrsDD: You completely misunderstand me. I’m not sure how that’s possible since I clearly did not envision giving the money away per capita or something silly like that.

      If you break up the Walton fortune into 7000 chunks and give 7000 people selected for entrepreneurial possibility 10 million a piece, and this theory is correct, wouldn’t the theory say that we would then see a wave of entrepreneurship from those 7000 which we would not see from the Waltons?

      I think any counter-argument would have to invoke negative externalities – focusing on the negative effects of the confiscation/break-up itself, or the distribution; but any such counter-argument can be defeated by rewriting the argument to focus on preventing the creation of such giga-fortunes and favoring smaller fortunes. (For example, inheritance laws could be modified so that giving anyone more than 10 million is heavily penalized but any amount under that is favorably treated.)

      Granite26: many people do not invest for those reasons. It’s hard to imagine that the investors in many techie-beloved startups like Tesla seriously expect to make billions of dollars off it; many tech millionaires invest to ‘do good by doing well’ or because the goal is crazy-awesome (John Carmack & Armadillo Aerospace come to mind).

      And even if there is a negative effect, we don’t know that it would always and at every level of break-up be worse than its gain. To go back to my example, would breaking up and redistributing the Walton fortune do *so* much damage that the benefits of 7000 entrepreneurs looking to invest in something won’t exceed it?

  • From:

    From a social and cultural point of view, mixed in with rule breaking and exploration in High School, separating people makes a lot of sense. Hard to arrange it just so, but possible. Right now, most communities either force everyone together or have a cliff where kids are thrown off of into the pit of the “other” school. The nature and quality of the “other” options is what separates good districts from bad districts, but few or none do it right yet. Ideally, you need to keep fluidity so that one false move or bad quarter doesn’t doom you to the dungeon. As long as there is rapid mobility between realms, it is a great idea. Unfortunately, often the insidious things start small and aren’t clearly a problem until long after they have become infused in a clique. Making negative effects of attitude, habits, and choices apparent much sooner would be very good. Feedback loops longer than a few weeks are useless to a large chunk of teenagers. And any extremely discontinuous penalties lead to fatalism in some significant subgroup.

    The problem is that many with wealth aren’t A) adventurous and B) don’t feel enough social responsibility to start businesses or otherwise directly foster the creative engines. Many have been convinced that they are doing that by investing in the stock market or some other indirect feeding of the Big Business Machine. I feel that, in the average case, that is exactly like thinking that you are directly helping the homeless guy you saw today by paying more Federal taxes.

    I firmly feel it is my responsibility to, after doing as well as I can with my children and significant others, to use my skills to A) do the most good and, given A, B) make the most money so that I can find new ways to do or help other people do A. Having wealth creates a responsibility to perform at least as much as having high ability. Being neutral or negative in effect on the world is the only real sin there is.

    For any wealthy individuals who truly pursue and directly support entrepreneurialism in a consistent and significant way, I would vote for zero taxes during those years they are cogent and active. When they are successful, I’d even vote for a rebate on some of the tax revenue they created (probably for some specific situations). For those hoarding and riding their wealth, I would double their taxes (or something along those lines). I totally applaud the former and despise, to some degree, the latter. And I despise societies that make the former harder than it is naturally, often by directly protecting big business thereby enshrining reverse economies of scale out of fear or ignorance. (Japan perhaps? France probably — their socialistish environment for small businesses is onerous.)

    On the relatively poor entrepreneurs starting businesses, they tend to disappear more often, however they are more driven and are able to take existential risks more to some extent. Different businesses may require different types of actors for competitive success.


  • Kynes

    Oh brave new world, that has such people in it.

    • Ha, you are pattern matching on the wrong thing. I’m suggesting steps away from homogeneity, not toward a rigid class system with any kind of predestination. Everyone has a wide range of where they could end up capability wise. Much of what actually happens is governed by influences that have little to do with your natural ability.

  • Ken Rhodes

    MZ wrote >>I think I would have been better off in high school if I wasn’t required to take electives in art, home economics, and wood shop. I’ve never used that “knowledge” as I don’t even remember it.>>

    MZ, my experience is the opposite of yours. I, too, had to take shop courses I’ve never used, but I think they were VERY valuable, because I came to understand how things are done that, as a software engineer, I would otherwise never have done, and how people work that I would otherwise never worked with. I don’t propose this as an economic advantage, but just an advantage for “living in the real world.”

  • wkwillis

    1. Pay kids to learn.
    No further problems in high school because the kids are motivated to learn as fast as they can. Before high school kids learn so little that the lower level of schooling simply doesn’t matter.
    You do have the problem that very high earnings enable brainiac fifteen year olds to kill themselves driving their convertibles the wrong way down one way streets while a half naked fifteen year old girl is drinking Johnny Walker and doing coke in the shotgun seat…but the sacrifice is worth it, in the opinion of the fifteen year old brainiacs.

  • william

    “Thus we’d probably do better to isolate the worst kids in their own school or work hell; they’d be worse off but the gains to other students would more than compensate.”

    Yes, perhaps, but the missing part of this statement is that those who benefited from this arrangement would then have to compensate those who got screwed.

  • Pingback: Overcoming Bias : We Need The Very Rich()

  • Pingback: Why Do We Need the Rich? | John Goodman's Health Policy Blog |

  • Pingback: Why Do We Need the Rich? | Healthcare()