Expert Failure

Soul searching continues among macroeconomists who failed to predict the economic crisis. One culprit behind the great macro-flub: the “Control Illusion,” in which economists are blinkered by overconfidence in computer models just because they, as Levitt recently wrote, “solve a problem that is really, really hard mathematically.” Speaking of the illusion of control: when it comes to which experts to listen to, people seem to value confidence over expertise. (HT: The Morning News)[%comments]


Leave A Comment

Comments are moderated and generally will be posted if they are on-topic and not abusive.



View All Comments »
  1. ceolaf says:

    Doesn’t this speak to George W. Bush’s appeal?

    Always confident and full of bluster, rarely well informed. And shockingly popular for most of his life.

    Thumb up 1 Thumb down 1
  2. Butch says:

    There is always an internal tension between my two intellectual passions, engineering and economics. There are also elements in the practice of both that have much in common. A commonality pertinent to this issue is the use of computer models to solve extremely complex problems that have numerous poorly defined variables – the actual values being available only after the fact. Concurrent with this similarity is the problem that results when practitioners have only a superficial understanding of the underlying theory (or chose to ignore the complexity) and the limitation of the model being used. These individuals pick a set of inputs, run the model and assume that the output is “truth.” Point forecasts or failure to consider the applicability of the chosen analytical model and/or its sensitivity to variation of input is inevitably a recipe for disaster.

    Thumb up 0 Thumb down 0
  3. Caliphilosopher says:

    This doesn’t surprise me, especially since there is a failed view that economic facts are directly reducible to mathematical facts.

    Since it’s a social science (not a ‘hard’ science), economists who rely on computer models seem to forget that they have other variables for which they cannot account.

    Thumb up 1 Thumb down 0
  4. Eric M. Jones says:

    —Economic models all contain a conundrum:
    How can an economic model accurately predict the future when the very existence of a model that predicts the future changes the future? E.g. If my clever computer model says that a stock price is going to triple, how does the fact that I publicize that, change the behaviour of the stock prices?

    Almost any reader of this post can buy enough penny-stock in the Pink Sheets to severely disrupt the market and will get a call from his broker disallowing it. Well, my actions can hardly affect the price of IBM stock, but the casual remarks of the Fed chairman certainly can…and does.

    Hey, happy days are here again….guaranteed.

    Thumb up 0 Thumb down 0
  5. SpecialK says:

    Shouldn’t this same distrust of computer models apply to the global warming scientists that use computer models as a basis for trying to destroy whole segments of the world economy?

    Thumb up 1 Thumb down 0
  6. Joe Smith says:

    There is an arrogance in accepting responsibility – it assumes that you were in charge in the first place.

    The professions that should really be doing some soul searching are the lawyers and accountants who’s excuse for existing (and charging big fees) is professional review of business arrangements.

    Thumb up 0 Thumb down 0
  7. Caliphilosopher says:

    #5 –

    It’s not a distrust of computer models – it’s a distrust of an enterprise that attempts to reduce economic facts to mathematical facts without understanding that there are very few things that are directly and neatly reducible in that sort of way. The prediction that the temperature of the earth will rise does not causally assist in the rising of the temperature.

    Eric M. Jones has it spot on. Economics is a social science – climate science is something that deals with predictions of objects that have no (ir)rationality or anthropomorphic properties. That’s why physics, chemistry, mathematics, and biology work so much better with computer models than Economics.

    That’s the difference.

    Thumb up 0 Thumb down 0
  8. L Nettles says:

    And yet we are going to cripple our economy based on climate computer models. Will we ever learn.

    Thumb up 0 Thumb down 0