"Econometrics" Is Bullshit

To: Donald Luskin

It was only this morning that I found these words at mises.org:

"It is a mistake to use, as journalists and some economists do, statistics without logic but the reverse does not hold: It is not a mistake to use logic without statistics."

Reading your correspondent Seater's words, I was struck with this:

"Science never proves anything; it only fails to disprove."

In the very next paragraph, he says:

"In that case, I challenge you to provide a superior method of establishing knowledge of any aspect of reality."

The thing that's curious to me is how people like this go about equating the skepticism that is necessarily implied in the first statement with what they also necessarily value as "knowledge" in the second statement. If we take the first statement seriously, then they are committing outright theft of a concept in the second, and they should be called on it.

I would presume that anyone competent to handle this discussion should be able to stipulate to the package of concepts contained in the statement: "The sun will rise tomorrow." What it means, of course, is a bunch of things like; the earth will rotate to a point where it will appear to any given observer at a given location as if the sun were separating from the horizon along a vertical axis. Something agreeably like that. Now, if an assertion like "science never proves anything; it only fails to disprove" means anything at all, then it must mean that the facts of physics which allow us to confidently look forward to tomorrow's sunrise are suspenseful. Note that this is not mere suspense over contingencies like whether a giant asteroid will come blasting along to knock us on our celestial asses, but whether the very standing of facts as facts merits human respect. That really is what it all must boil down to.

Having pointed out here, and very briefly, an anecdotal illustration in the long post-Enlightenment history of what I call "catatonic skepticism", let me point out something else very important:

I think you're missing something crucial when you finger the foolishness of "economists and journalists". The crucial thing is bureaucrats. Econometrics is a very handy device for diffusing the observations and principles of so-called "classical" economists, in order to carve out space for their technocratic machinations from the resulting confusion. The more doubt that these people can exploit, the more rationale for all their "exploring" and "experiments". There are serious political implications in econometrics. And I maintain that all this is a consequence of the deplorable state of epistemology over the past century or so: when there is no such thing as truth, then things like statistics will be pressed to service (such as it is), instead.

This is why Levitt has "breathed new life" into the bullshit. That book is a new line of credit extended to a desperately overdrawn account.

Now, see my subject header.

You say that econometrics "can be" a "valuable discipline". I won't argue that with you. What I say is that it should be kept in its place, which is: completely out of every discussion of "public policy". That's the "bullshit" that I'm talking about. A bit over forty years ago, now, Ayn Rand wrote:

"Political economy was, in effect, a science starting in midstream: it observed that men were producing and trading, it took for granted that they had always done so and always would -- it accepted this fact as the given, requiring no further consideration -- and it addressed itself to the problem of how to devise the best way for the 'community' to dispose of human effort."

It should be obvious to a blind person that econometrics has done yeoman's work in the cause. This is essentially because it doesn't have an ethical bone in its body of work.

This, above all, is my condemnation of Levitt and all like him.


Onward, then.