Brad DeLong’s Methodology
Yes, I am using that word correctly. I am here talking about DeLong’s views on method. So remember kids, if you’re giving a job market talk, don’t say, “In this paper my methodology was to regress unemployment against Glenn Beck’s ratings.” If I am on the hiring committee, you’re done for. (Of course, I won’t be on the hiring committee, so it’s a moot point.)
Anyway, Mario Rizzo tips us off to this amazing statement by DeLong:
One of the embarrassing dirty little secrets of economics is that there is no such thing as economic theory properly so-called. There is simply no set of foundational bedrock principles on which one can base calculations that illuminate situations in the real world. Biologists know that every cell runs off instructions for protein synthesis encoded in its DNA. Chemists start with what the Heisenberg and Pauli principles plus the three-dimensionality of space tell us about stable electron configurations. Physicists start with the four fundamental forces of nature. Economists have none of that. The “economic principles” underpinning their theories are a fraud–not bedrock truths but mere knobs twiddled and tunes so that th right conclusions come out of the analysis.What are the “right” conclusions? It depends on what type of economist you are, for three are two types. One type chooses, for non-economic and non-scientific reasons, a political stance and a political set of allies, and twiddles and tunes their assumptions until they come out with conclusions that please their allies and their stance. The other type takes the carcass of history, throws it into the pot, turns up the heat, and boils it down, hoping that the bones and the skeleton that emerge will teach lessons and suggest principles that will be useful to voters, bureaucrats, and politicians as they try to guide our civilization as it slouches toward utopia. (You will not be surprised to learn that I think that only this second kind of economist has any use at all.)
I feel utterly vindicated. For a few years now, when people ask me, “Are these Keynesians stupid or evil?” I say, “No, I think it’s that they don’t believe in economic law. So they aren’t lying in their minds. They can go to sleep at night because they think, ‘For all we know, my policy prescriptions just might work. Those free market guys can’t be sure they’re right either. This is just a big game we play with the public.'”
Now I had guys like Krugman in mind when crafting such responses; I really do think Krugman is clever enough to realize he can justify any policy response he wants, by making some ad hoc assumptions about political feasibility and then whipping up a two-period model.
But DeLong explicitly admits that he doesn’t believe there are economic laws. Very refreshing.
By admitting he believes there are laws of economics, it appears that DeLong is being honest, but he is merely revealing a deeper level of dishonesty. He wants readers to think that he is a robot merely examining the hard facts of history. But he does not do that, unless of course he really is a robot. All scientists, even physicists, approach historical evidence through theory. DeLong simply doesn’t want us to know what theory he uses. As Mises wrote, no one can approach the vast amount of data in historical research without some kind of theory because there is too much data. We have to filter it and we use theory to filter.
An example of a real automaton is neural networks software. NN programs really do approach data without any theory in mind. NN programs can create models of historical data if you tell them what your response variables are and what your predictor variables are. Then the program uses various differential equation algorithms to break the data apart and put it back together in ways that best predict the response variables. If theoriless modeling on historical data could work, NN programs could do it better than any PhD economist. Try it some time. There are free NN programs on the net. I used the best one on the market for many years in corporate work. It’s made by a small Tulsa company called StatSoft. What you’ll find with NN programs modeling macro economic data is that they give you a range of good networks based on the error rate. But each network will provide very different forecasts. So most NN modelers will average the 10 best. NN does a far more accurate job of forecasting than does traditional statistics, but it works best if you approach it with sound theory. And even then, you may not have the data you need to build good models, as Hayek often wrote.
It is safe to say the DeLong is simply being dishonest here. He even deliberately misidentifies which of the two branches he belongs to.
I guess we’re supposed to be the first kind of economists. “The ‘economic principles’ underpinning their theories are a fraud” kind of economists.
However, I’ve never found Mr. DeLong to show any familiarity with concepts such as “acting man”, “subjective value”, “economic calculation”, or “distortion of the capital structure”.
I guess he need to bother even familiarizing himself with those concepts since we’re part of the “Marx-Hoover-Hayek axis”. What else is there to say about that? See:
http://mises.org/daily/3463
and
http://www.thefreemanonline.org/featured/mainstream-macro-in-an-austrian-nutshell/
Typo alert:
I meant:
I guess he need NOT bother even familiarizing himself with those concepts since we’re part of the “Marx-Hoover-Hayek axis”.
It seems like after listening to your talks I keep hearing “methodology” used everywhere (and incorrectly). Everytime somebody says something like “my methodology is to walk my dog to the park”, I keep expecting them to say “my methodology is to use the facts of reality” or “my methodology is to use divine revelation from God”.
I think they’re evil AND stupid. Krugman and Delong at least, I am charitable enough to regard most Keynesians as simply stupid. They are both guilty of things that cannot be accounted for by dogmatism alone.