Skip to content
 

One fun thing about physicists . . .

. . . they’re not in awe of economists.

In contrast, economists sometimes treat each other with the soft bigotry of low expectations. For example, here’s Brad DeLong in defense of Larry Summers:

[During a 2005 meeting, Summers] said that in a modern economy with sophisticated financial markets we were likely to have more and bigger financial crises than we had before, just as the worst modern transportation accidents are worse than the worst transportation accidents back in horse-and-buggy days. . . . Indeed, for twenty years one of Larry’s conversation openers has been: “You really should write something else good on positive-feedback trading and its dangers for financial markets.”

That’s fine, but, hey, I’ve been going around saying this for many years too, and I’m not even an economist (although I did get an A in the last econ class I took, which was in eleventh grade). Lots and lots of people have been talking for years about the dangers of positive feedback, the risks of insurers covering the small risks and thus increasing correlation in the system and setting up big risks, etc.

I don’t think Summers, as one of the world’s foremost economists, deserves much credit for noticing this theoretical problem too and going around telling people that they “really should write something” on the topic. You get credit by doing, not by telling other people to do.

I think Steve Hsu (see above link) gets the point. No one’s going to go around saying that some physicist is a genius because he’s been going around for twenty years with a conversation opener like, “Hey–general relativity and quantum mechanics are incoherent. You should really write something about how to put them together in a single mathematical model.”

P.S. Just to be clear, I’m not trying to argue with DeLong on the economics here. He may be completely right that Rajan was wrong and Summers was right in their 2005 exchange. But I do think he’s a bit too overawed by Summers’s putative brilliance. In a dark room with many of the lights covered up by opaque dollar bills, even a weak and intermittent beam can appear brilliant, if you look right at it.

16 Comments

  1. I read Prof. Hsu's post you linked to, and I've read several posts by Prof. DeLong on this topic.

    Prof. DeLong's analysis impresses me more. Prof. Hsu's analysis in the post you link to is very surface level, and doesn't signal as deep a level of literacy on this topic as Prof. DeLong.

    I don't see the point of invoking that he's a physicist given how fluffy his analysis was on the topic at hand.

    It's too bad because I like cross-disciplinary imperialisms. I didn't see much to look at here, though.

  2. RGV says:

    Physicists aren't over awed, just occasionally jealous. Probably goes like, "Hey we are so much smarter than economists, but how come we matter so much lesser. Sigh, whats the world coming to these days…"

    Most physicists in wall street would have been better off in physics. But wait, there are no jobs there either.

  3. Numeric says:

    Summers. Forced out the woman who wanted to regulate derivatives in the late 90's (forget her name–she noted that Long Term Capital Management had $4 million to cover over $1 billion in bets and thought 250/1 was excessive. Sommers didn't.). Gave a manifestly untrue talk at Harvard about women not having the mathematical ability to do economics, among other things (mathematical economists are economists because their math professor told them not to go into math). Told Harvard to make a half-billion bet on some juiced-up derivative just before he left–they lost it all. As Chair of CEA, told Obama that $800 billion (which include a $100 billion extension of the alternate minimum tax) would work–Romer (another one of those pesky women) told him $1.2 trillion would be the minimum and a back-of-the-envelope macro calculation (see Krugman's various blogs) indicated $1.6 trillion.

    The fact that Summers is considered brilliant says more about economics and economists than his putative intelligence. Piaget claimed intelligence was the appropriate behavior for the situation and that's a far better definition than the ability to score near 800 on an SAT or write clever little models with no applicability and minimal mathematical sophistication.

  4. Numeric, I think you're straining mightily to make your narrative cohere. It seems to me Professor Delong provides a description of Professor Summers that has a more accurate feel to it.

  5. numeric says:

    http://www.tnr.com/blog/jonathan-cohn/77852/larry

    But we know, from Ryan Lizza's account in the New Yorker, that Summers didn't even present Obama with Christy Romer's proposal for a $1.2 trillion stimulus.

    I'm not going to provide references to the other
    comments I made in creating my "narrative", but in general, I've been too kind on Summers. Read

    Larry Summers and the Subversion of Economics

    http://chronicle.com/article/Larry-Summersthe/124

    It would not be hard to construct an argument that the funding of academic economics has made all of us worse off (except the economists). When the economists get ambitious and try to model real-world phenomenon they can start with modeling that seemingly counter-intuitive fact.

  6. louis says:

    The math economists use is in a whole other league then the math used by pure mathematians,
    and the larger part of economists has a very thin mathematical background.

    However, some of the theory folks have a very impressive background, surprisingly many imo gold winners. In recent years for example Sannikov, Manea, Carroll,…
    Mathematicians themselves consider imo gold medals as good indicators of math ability so I am not sure whether I completely agree with this statement by numeric "mathematical economists are economists because their math professor told them not to go into math".
    Economics is pretty ambitious in its goals and uses different ways to achieve this (empirics and theory). Maybe we expect too much from it…

  7. Jonathan says:

    Your vituperation, Numeric, even with links, gets in the way of clarity. For example, Ferguson is wrong when he says Summers claimed women were innately less capable. What he said was that women showed lower variance in skills than men, which seems supported by all the data. A consequence of this (holding means constant and equal) is that men will be better represented in the upper tail of the distribution as well as in the lower tail. Having seen Inside Job, and having some knowledge of the less-than-fair debating tactics that go into the making of such a work, it is unsurprising that the Ferguson piece doesn't present a balanced portrait. You have similarly overstated his powers as to Harvard's endowment.

  8. numeric says:

    Inconvertible.

    1) Summers opposed regulating derivatives in the Clinton administration. All accounts indicate that he forced
    Brooksley Born out because she wanted to regulate them.

    2) Summers attacked (with his "brilliant" intellect) the 2005 presentation by Rajan which was exactly correct. Summers called him a "Luddite" (and I'm accused of vituperation?).

    3) Summers signed off on a less-than-adequate stimulus, going so far as refusing to pass on Romer's estimate of at least 50% more to Obama.

    Academics tend to confuse clever arguments with wisdom, primarily, I think, because they are selected for the ability to make these arguments rather than wisdom. Summers clearly lacks even the fundamentals of any real insight into how the economy works. In particular, the privileging of capital historically has always lead to economic disasters (various crises throughout the 19th and early 20th century, culminating in the Great Depression), then, starting with deregulation in the late 70's, S&L, East Asian, and now the Great Recession. The point of Inside Job (which should be obvious) is existence determines consciousness–"brilliant" people are doing too well under an increasingly unstable financial structure to present intellectually honest arguments (among the lesser-known aspects of the nearly 3-fold increase in income inequality (measured as percent of income going to the top 1%) over the last 30 years is that professor's salaries have, in real terms, doubled. In other words, they've been bought off and they've become shills for the financial system which has rewarded them.

  9. Jonathan says:

    Well, let me try to "convert."

    1) Maybe. But so what? You have to run the counterfactual — how would things have turned out if Born had her way? And I don't think anyone knows the answer to that for a whole host of reasons, including the fact that Summers was far from alone in his opposition.

    2) Correct after the fact. And "Luddite" is, in this context, an ugly phrase. But no one has ever accused Summers of being nice. The question is whether he had the better of the argument, in which being proven right in hindsight is but one factor.

    3) And how much stimulus would Congress have passed? Has Summers described his reasons for not passing on Romer's estimate? How about its conflict with Romer's previously published work on the elasticity of GDP with respect to stimulus? Or something else?

    I'm not a defender of Summers — really, I think he's a bit of a jerk, and I know a thing or two about his personal life which are unsavory. Further, his politics are not mine, either. And I agree that academic brilliance > policy brilliance. All I'm saying is that your charges are somewhat underargued — and that his denigration of women is just factually incorrect. http://www.thecrimson.com/article/2005/1/14/summe… "Goldin said that Summers’ support for women in academia is well-known… 'The reason Larry gave this talk is that he’s extremely interested in the way that institutions can enable individuals to perform to their maximum. And it bothers him when individuals do not perform to their maximum,' Goldin said. She added that Summers is 'really dedicated to changing institutions' so that women can attain leadership roles throughout academia.

  10. numeric says:

    I'm certain Custer complained about carping from his men at the Little Big Horn as hindsight, or criticisms that he should have waited for the rest of the Army as counterfactual, had he known what that word meant. Economists contradict themselves all the time, but it was clear from history that this was a financial crisis recession, and those take 3-5 years on average to work through (the Republicans realized that in that infamous powerpoint presentation _before_ Obama was inaugurated, which is one reason they decided on total obstruction).
    As far as better arguments on derivatives, Buffet referred to them as instruments of mass destruction, but Buffet was the most successful investor in post WWII America and not an academic. I'm sure an academic can whip up a dandy mathematical model which shows that they can only improve things and we live in the best of all possible worlds but it clearly isn't true.

    As regards to Summers's attitude towards women, I've given three examples where women seemed to be on the short end of the stick. But the conceit of Summers is that this tighter "variance" somehow disqualifies one from being a better researcher is remarkably self-serving. One of the attributes of "good" research is that one can make predictions from it. Summers predictions were lousy and there is ample reason to suggest that his self-interest colored his predictions. I don't think it's too much to suggest that his biases (either conscious or unconscious) affect how he views women as researchers or colleagues or as government bureaucrats.

    Incidentally, there is a way to test this–it's done all the time in psychology. Just set up a random experiment where academics are given a paper written by a male or female author (otherwise identical paper) and ask them to rate its worthwhileness (obviously, this couldn't be done on any one individual, but it would give systemic effects). Everytime this is done in psychology women end up below men.

  11. TGGP says:

    For what it's worse, I agree with H.A that Hsu's take seems rather superficial compared to Delong's (Delong does risk sounding partial by saying he owes his success to "running an emulation of Larry Summers in my brain" though). Has Rajan written up an account of his disagreement with Summers?

  12. Andrew Gelman says:

    Tggp:

    I defer to DeLong on the econ, and he might very well be correct that Summers has good ideas about macroeconomics, but I'm not at all convinced by DeLong's defense of Summers as quoted above. Summers is/was a well-known researcher in econ. To say that he went around talking about a well-known problem (positive-feedback trading), recommending that other people work on it . . . well, that's not so impressive. You get credit for doing the work, not for spending 20 years suggesting that other people work on a problem.

  13. TGGP says:

    I agree that there's nothing impressive about telling other people they should do something. It merely defends Summers against any charge that he wasn't aware of such issues or didn't think they were worth spending time on. Instead, he just thought Rajan's particular viewpoint was wrongheaded.

  14. Andrew Gelman says:

    Tggp:

    Could be. But maybe Summers shouldn't have been so sure that viewpoint was wrongheaded, if he hadn't actually been working in the area himself.

  15. " You get credit for doing the work, not for spending 20 years suggesting that other people work on a problem."

    I think you get credit for both. What sense is it to give no credit for suggesting other people work on a problem?

  16. Andrew Gelman says:

    If someone discovers or even popularizes a new anomaly, that's fine. But taking a problem that everyone's aware of, and encouraging other people to work on it . . . maybe that's good citizenship but it's hardly anything impressive. As I wrote above, no one's going to go around saying that some physicist is a genius because he's been going around for twenty years with a conversation opener like, "Hey–general relativity and quantum mechanics are incoherent. You should really write something about how to put them together in a single mathematical model."