Mark D. White

Writer, editor, teacher

Mark D. White

There's a fantastic article in The Wall Street Journal today by Mark Whitehouse (!) explaining how, given the current crisis, more economists are finally starting to doubt the efficacy of mathematical modeling to predict movements in the economy. Of course, some want to build even bigger models, but others, such as Roman Frydman and Michael Goldberg, emphasize the impossibility of dependably precise and accurate forecasting, and through their "imperfect knowledge economics" recommend common-sense policy action that, rather than calculated to attain precise policy effects, is based on rules of thumb to push the economy in the right direction. (I must admit I was ignorant of this new work, not being a macro person, but I say the same things to my macro students every time I teach it, so I was very gratified to read this.)

Of course, the greater ethical question is why spend resources forecasting the economy anyway, but I'll expand more on this later…

Posted in , , , , ,

2 responses to “Questioning Economic Forecasting (Finally)”

  1. Stefan Avatar
    Stefan

    Thanks. Great points. At RAND they use “robust decision making” to assess which decisions might be most helpful in an unpredictable future http://www.rand.org/international_programs/pardee/methods/ Combines humans and computers.
    And Tetlock’s book “Expert Political Judgement” is a great work showing that human forecasting ability is even worse than that of mathematical models.

    Like

  2. Mark D. White Avatar

    Thanks, Stefan – I think all the concern with forecasting, especially for the purpose of policymaking, is woefully misplaced. I hope to expand on that soon.

    Like

Leave a comment