Statistics and mathematics — of no avail in solving our disagreements

25 March, 2015 at 20:29 | Posted in Economics | 3 Comments

Statistical Science is not really very helpful for understanding or forecasting complex evolving self-healing organic ambiguous social systems – economies, in other words.

leamer1 zoomedA statistician may have done the programming, but when you press a button on a computer keyboard and ask the computer to find some good patterns, better get clear a sad fact: computers do not think. They do exactly what the programmer told them to do and nothing more. They look for the patterns that we tell them to look for, those and nothing more. When we turn to the computer for advice, we are only talking to ourselves …

Mathematical analysis works great to decide which horse wins, if we are completely confident which horses are in the race, but it breaks down when we are not sure. In experimental settings, the set of alternative models can often be well agreed on, but with nonexperimental economics data, the set of models is subject to enormous disagreements. You disagree with your model made yesterday, and I disagree with your model today. Mathematics does not help much resolve our internal intellectual disagreements.

Ed Leamer



  1. “complex evolving self-healing organic ambiguous social systems”

    I wonder how “self-healing” got in there — seems oddly ideological beside the other, more neutral terms. And, I wonder, too, about the absence of “reflexive” which would seem particularly problematic to my naïve mind.

  2. Needs o substitute ‘mathematics’ with ‘econometrics’ then hard to disagree.

  3. This is a very, very important post. It is a fantastic demonstration of the processes by which economics remains in a coal pit.

    Leamer is a professor of economics and a professor of statistics. His conclusion is correct, more or less. But here are the fields he doesn’t know much about: philosophy, social science, and science. As a result he makes the following statement:

    “In experimental settings, the set of alternative models can often be well agreed on, but with nonexperimental economics data, the set of models is subject to enormous disagreements.”

    This suggests that he believes the lack of experimentation is what sets economics apart. But this is silly. “Organic ambigous social systems” have been extensively studied by, for example, Jane Goodall, without conducting a single experiment.

    The basic problem, and it is very widespread, is that the debate about “is economics a science” is conducted by people who equate “science” with “what I learned in high school about physics”. Thus the debate presents a false choice.The high school part isn’t the problem. The problem is that there are a ton of sciences that don’t do much experimentation and are nothing like Newtonian physics. Economics is one of those sciences, but no one is ever going to see this if the critics of mainstream economics don’t see it first.

Sorry, the comment form is closed at this time.

Blog at
Entries and comments feeds.