And You Thought IS-LM Analysis Was Dead

The old Neoclassical Synthesis is resurrected once again, over at Krugman's blog.  (Notice that his discussion of the budget deficit's impact on interest rates assumes the LM curve is nearly horizontal below full employment.) 

What fine work from a recent Nobel Laureate.  Hansen would be thrilled.

Comments

[Greg Ransom]

Can you say IDIOT SAVANT?

Paul Krugman is the poster boy of what the AEA Committee on Graduate Education was talking about in 1991 when they first described the graduate school "idiot savant" -- and he's the guy Klamer and Colander were writing about in _The Making of an Economist_ when the wrote about the "rocket scientists" who knows no history of economic thought, or much of anything besides the latest math fads and some econometrics.

the IDIOT SAVANT.

It's what for dinner from the top graduate schools in economics.

[Jayson Virissimo]

Greg, I disagree about Krugman. Most who have worked with him do not consider him to be very mathematically inclined. His background is in history, not mathematics. Most of the math he does know was learned specifically in the context of economics. I have experience with so called "idiot savants" in economics departments, but Krugman doesn't fit the bill.

[Peter Boettke]

Jayson is correct, Krugman is not considered a high powered theorist. He is a good writer comparatively, and he is politically engaged so he attracts a wide-readership. What he is not, is a scholar of economic ideas (there is little reward in this profession for that) nor is he a charitable reader of opponents.

He is very similar to John Kenneth Galbraith from a generation or two past, except that his period writing "technical economics" last a little longer than Galbraith's.

There is also a disjoint between his contributions to economic "science" and his policy discourse. The phenomena of Krugman is as worthy of study as Krugman himself.

The presentation of ideas in this article are so out of date --- it is pre-1980s macroeconomics that Krugman is championing. Can anyone say Bob Lucas or Robert Barro?

Pete

[liberty]

My question: if he is truly pre-1980s as far as the profession goes, as far as academia goes, why does this not seem to hurt his reputation, and why don't other economists call him on it? Why does it seem like the mainstream political opinion of today has slid back to Keynesianism (eg. the Obama team) and economists have done nothing to stop it?

If it really is gone from graduate programs, as people keep telling me, then why is it OK for the cover of the NY Times, for the official government analysis, and for the State of the Union speeches (that will come)?

In other words, if this stuff has been debunked, and is long past, no longer considered legitimate by the experts in the field, why is it what the government is basing policy on? If government science policy were pre-Darwin, or Lysenko, would this not be an issue? It is especially strange if we passed over this phase, and then have suddenly returned to it... is it not?

[Josh Hendrickson]

Even Hicks abandoned the IS-LM framework.

[Stewart Browne]

I love that he labels this post (and other similar drivel on his web site) as "Wonkish." It's the classic blowhard argument. "If you don't agree with me, it's because you're not smart enough to understand!"

[Greg Ransom]

I'm no expert on Krugman, but I do know he's an MIT guy from the 1970s, with all that that entails.

I do know that his claim to fame as an economist has to do with participating in the mathematization of ideas which some very, very good economists say were already out there, but were not in mathematical form.

I do know that Krugman was a math cruncher at the Council of Economic Advisors

And I've read enough of Krugman to know that he is a very, very shallow guy when it comes to the history of economic thought, the history of ideas, the history of America, and in just about any other dimension.

So he counts as an "idiot savant" in my book -- even if there are many better paradigm cases.

Krugman is a massively influential guy, and he is worth a closer look. Any good Krugman bios out there?

[Greg Ransom]

"why don't other economists call him on it?"

Ed Leamer and Robert Barro and Greg Mankiw have fairly forcefully all but said that Krugman is an idiot on this stuff -- and completely outside his professional competence.

Fairly gutsy stuff for academics, who aren't known for courage.

[Ben]

Idiot Savant, without all that pesky Savant baggage.

[Bill Woolsey]

To me, it seems more like that Finance guys, some of whom hold econ PhDs, were the most guilty. The "quant" guys on wall street. (Some of them were physicists and mathemtaticians.)

I just read a claim that investments in mortgage backed securities were made based on risk assessments formed from analysis of the prices of credit default swaps. This might be fine if those selling the swaps understood and made good judgements about the risk. But someone actually has to use their judgement about the real world. Mixed in with this was the claim that risks were determined based on past performance. (I hate to admit it, but when I first heard about this "black swan" business, I just couldn't believe that anyone would think that past variances could be used to completely forecast future performance. I was just that out of it regarding what those quant people do on wall street)

I guess one lasting impact that Mises had on me is that I never have had a notion that I have a lot to offer investors or businessmen. We had our own local Madoff a few years back. He was an economist and also the "local" forecast guy. He ran a ponzi scheme. Some of his investors claimed that he used his clever economic tools to beat the markets. Well, he never told any of the other economists in town about this. My view was that we would have all laughed and known he was a crook. Maybe not.

In macro, it was Barro, Lucas, etc. who may well be good economists, but who made mathematical formalism the holy grail. If you can't express an idea in the context of a intertermporal, stochastic model based on constrained maximization, it isn't scientific.

I think simple tools that provide partial insights are highly useful, expecially compared to what noneconomists think about these things.

I _like_ the "IS curve" as a way of thinking about interest rate fundamentals. Just like Krugman, with potential income being the constraint (he calls it "full employment.) But then, I like "supply of saving" and "demand for investment" curves too. You just can never assume that such simple tools tell the entire story.

Frankly, I don't see a lot of "Austrian" work on the determination of the natural interest rate. The fundamental nature of the interest phenomenon, yes. How money creation distorts interest rates, yes. What sorts of things determine the "undistorted" interest rates--not so much.

Krugman has been claiming that the "LM" curve is horizontal at zero. Well, I think there is something to the zero lower bound on nominal interest rates.

I think Barro has many good insights, but believing that Ricardian equivalence applies in the real world isn't one.

I am sure that Lucas has many good insights, but continuous market clearing wasn't one of them.

Perhaps it is just me. I am one of those economists who knows little beyond what I gathered in graduate school. But one of the things I gathered (from Buchanan) is that Barro and company prove what constrained maximization isn't what economics is supposed to be about. And that Ricardo didn't believe in Ricardian equivalence. And from Yeager that "old" classical macroeconomics didn't assume continuous market clearing, and neither should we.

My major criticism of Krugman's arguments is that the fact that the single interest rate story is especially problematic when a stampede to safety inceases the divergence between different interest rates. He points to this when it is convenient for his argument (for big government) but he just mentions it as a complication here.

In my view, Krugman is trying to understand the real world. He agrees with Ransom (and Kling, I think has said the same) that macro has been mathematical masturbation for a good bit of time.

It is obvious that many economists don't believe in Ricardian equivalence (generally, that people always save enough to pay future taxes.) Still, many people were willing to play with models that assumed that... because.. I don't know. Solving constrained maximization models is what we do? It is scientific? Wow? look at all the interesting results you can generate? (like deficit financed govenment spending raises current output because people start working hard now to pay the future taxes?)

[thesis paper]

Wonderful article, thanks for putting this together! "This is obviously one great post. Thanks for the valuable information and insights you have so provided here. Keep it up!"