Monday, March 14, 2011


My grandmother took economics from Stephen Leacock, who was better known as a humorist (or humourist) - something like a Canadian Mark Twain. Apparently Leacock was very funny in class. My grandmother thought that some people were "funny ducks," and if she could read this blog entry by Paul Krugman, (which she can't, having long departed this world) she would probably have a good chuckle. I'm sure people are guffawing today at the Minneapolis Fed, over having spent 40 years of hard work promoting ignorance.

Here we go:
Today’s freshwater economists don’t believe in Friedman-type monetarism;
First, "freshwater" is very passe. The descendants of Lucas, Prescott, Sargent, Wallace, etc., are everywhere in the world, and in some cases a couple of generations (maybe even 3 in some cases) removed from the source. Is Mike Woodford or Mark Gertler a freshwater or saltwater? Hard to tell. Second, I think we learned a lot from Friedman-type monetarism. The basic Friedman-rule idea (i.e. optimum quantity of money) is a powerful one, though there are good reasons why we think we should depart from it in practice. Targeting monetary aggregates seems a bad idea. 100% reserve requirements are unquestionably a bad idea. Some of Friedman works, and some doesn't, but you can't fault him for not being original.

The first post-Friedman generation bought into the Lucas-type argument that no anticipated shock to demand can have any real effect;
Many people who have been taught by Ed Prescott or have talked to him, have bought into the idea that "demand" and "supply" are the wrong language to be using for organizing our thinking about macroeconomics. Lucas's "Expectations and the Neutrality of Money" wasn't about "demand shocks" but shocks to the stock of fiat money, engineered by the central bank. Lucas wanted to show how one could observe Phillips curve correlations, but this Phillips curve would not be exploitable by the monetary authority. That's a powerful idea, though the monetary nonneutrality in Lucas's model is not something we take seriously any more. Lucas's paper is also a watershed in terms of its methodological contribution.

...the next cohort turned to real business cycle theory, in which recessions are basically like bad weather that both reduces a farmer’s productivity and induces him to stay indoors.
See this.

This is presumably the answer to my question about why Keynesians seem to understand New Classical models, while the New Classicals themselves apparently don’t: the Keynesians have thought long and hard about demand, the classical types have never done so, not even in the context of their own models.
I think most macroeconomists you run into these days understand the full array of models in use out there. Obviously Krugman has not been hanging out at macro conferences (or macro seminars in his own department), so he would not know that. Further, if you look through some of my blog posts, you'll see instances where I have to explain Keynesian models (New and Old) to the Keynesians.

Probably the most painful thing in Brad’s notes is Robert Lucas’s sneering dismissal of Christy Romer, whom he ridicules as someone who just made stuff up on the fly — and who then makes up his own version of Ricardian equivalence on the fly, and gets it completely wrong.
I wondered what this was all about, and looked at DeLong's notes. It's basically a lecture to Brad's undergrads, and he includes a slide with a quote from Lucas that seems to have been Bob's extemporaneous comments from a panel discussion, or some such. Funny thing to teach to your freshman class. I have a hard time imagining Bob sneering, but I'm sure it's possible. Maybe if Bob had Dick Cheney's face. In any event, Christina Romer richly deserves all the criticism she can get, if not outright ridicule. I supplied some here and here.

And finally:
But while the likes of Olivier Blanchard are indeed reconsidering their views, the people who got things completely wrong are showing about as much self-awareness and remorse as, well, Wall Street.
How can we have a discussion about this? There are some "people." Who exactly are these people? You can't debate anything if you don't name names. What is it that these people got completely wrong? What is right? What is wrong? Who is right and who is wrong, and why?

In Paul Krugman's mind lives a beast. The beast was born in his mind while he was a graduate student at MIT in the mid-1970s. The beast is Krugman's image of modern macroeconomics, and it bears no resemblance to any creature, living or dead. It certainly puts Paul in a foul mood. Maybe there is something he can take for that.

P.S.: Here's a project for an ambitious person. Take the students (i.e. completed supervised PhD dissertations) of Prescott's, the students of the students of Prescott, etc. Add up publications, citations, weight by quality, whatever (you can use REPEC for this). Now do the same for Krugman, and compare.


  1. "I think most macroeconomists you run into these days understand the full array of models in use out there. Obviously Krugman has not been hanging out at macro conferences (or macro seminars in his own department), so he would not know that. "

    Are you talking about younger macroeconomists, immersed in research, but not much involved in politics, policy, the media, and blogs, as opposed to people who are like Taylor, Cochrane, and Prescott?

    Because I think Krugman may be talking about people like Taylor, Cochrane, and Prescott.

  2. That's the question. Who is Krugman talking about? Taylor is not much involved in research, but he's an interesting case. Obviously the guy is conservative and Republican, but when he did research it was very Keynesian. He's the guy responsible for the Taylor rule of course. Cochrane is also obviously a conservative, but he did research on the fiscal theory of the price level, which Woodford and Sims (both on the Keynesian side of the fence, more or less) got into in a big way as well. Prescott isn't much for getting directly involved with politics and the media, let alone blogs. Knowing the field is knowing the field. It doesn't have much to do with youth or old age.

  3. Krugman says: Today’s freshwater economists don’t believe in Friedman-type monetarism; they’re two intellectual generations of intellectual retrogression beyond that. The first post-Friedman generation bought into the Lucas-type argument that no anticipated shock to demand can have any real effect; when that model failed, the next cohort turned to real business cycle theory, in which recessions are basically like bad weather that both reduces a farmer’s productivity and induces him to stay indoors.

    What decade are we living in? Did everything not New Keynesian stop in the mid-1980's?

  4. Stephen,

    We all need to be more humble (including Krugman). It is not clear at all how much do economists really know about macro. Lucas wrote just a couple of years ago that the problem of the business cycle had been solved (check his paper "Macroeconomic Priorities"). No doubts that both Lucas and Prescott made influential methodological contributions but the economic ideas that made them famous are not convincing any more.

  5. Is Mike Woodford or Mark Gertler a freshwater or saltwater?

    Fair point, but the freshwater/saltwater distinction is a useful simplification of the real world, which is kind of what economics is about. Woodford and Gertler are both clearly in the Keynesian tradition, so they are saltwater. If you think of it as a continuum, Woodford in particular would be near the middle.

    100% reserve requirements are unquestionably a bad idea.

    Unquestionably? Really?

    Many people who have been taught by Ed Prescott or have talked to him, have bought into the idea that "demand" and "supply" are the wrong language to be using for organizing our thinking about macroeconomics.

    Perhaps but he's a RBC man, isn't he? In that paradigm, AD/AS clearly isn't helpful. To those working in a New Keynesian context it is. Woodford seems quite happy to use the term "aggregate demand", for example. Lucas may have not liked the term, but he was in fact talking about demand shocks.

    the monetary nonneutrality in Lucas's model is not something we take seriously any more.

    I'm interested in what you mean by this; who is "we". My understanding was that those of a freshwater persuasion took it seriously insofar as they take monetary nonneutrality seriously, which is not very. On the other hand, New Keynesians take monetary nonneutrality seriously, but don't take lucas misperceptions seriously.

    I think most macroeconomists you run into these days understand the full array of models in use out there.

    I'd like to think so too. Unfortunately, people like Fama, Cochrane, and Kocherlakota keep saying stupid things that make Krugman look correct, e.g.

  6. 1. What do you call me? I was Gertler's student, but Gertler and I both learned a lot from Aiyagari, who was a Wallace student. Gertler and I both wrote with Aiyagari, and Gertler, Bernanke and I (and Bruce Smith and others) shared a research program early on (roughly: financial frictions and macro). The core of Woodford's model is RBC - it came from Prescott. So what labels do you apply to who? I'm not sure.

    2. Unquestionably. Yes.

    3. When Woodford says "aggregate demand," that's just marketing. He's selling the ideas to Old Keynesians. "Aggregate demand" does not really mean anything in his model, which is about relative price distortions.

    4. The monetary nonneutrality issue is a complicated one. Prescott is on record as saying that monetary and financial issues are unimportant. But I don't think so. I think the nonneutralities work through redistribution effects and through what I call an "illiquidity" effect. No one is pushing the misperceptions story. I think that is dead.

    5. I'm not much concerned about who Krugman thinks is stupid, as he is generally clued-out. If you read Kocherlakota's speeches, you'll see that he is quite sympathetic to New Keynesianism. See this:

  7. 1. I would call you a freshwater economist, mainly because of your consistent hostility to anything Keynesian. I agree that the real world is complicated, and nobody really fits the archetypes exactly. To be honest, I'm not entirely clear about the implications of New Monetarist models. Financial frictions usually fits in with a Keynesian view of the world, where there's a role for the government to manage aggregate demand. You seem to think the role of the government is more limited, which puts you in the freshwater camp.

    As for the core of Woodford's model being an RBC model, well yes that's why I would put him in the middle if it were a spectrum. But I think it's missing the point because what's important in Woodford's model is monetary policy, whereas in RBC models that's irrelevant. Followers of Woodford thought government was responsible for the great moderation due to enlightened monetary policy. According to the Prescott view, that's baloney. So there's a big difference.

    2. Where's the proof? I'm tempted to agree that 100% reserve requirements are a bad idea, but better men than I have thought otherwise, so I have an open mind.

    3. Oh, please! He's using it because it's what he means, because economists understand the difference between aggregate demand (money, government spending etc.) and aggregate supply (production).

    The relative price distortions are important because of their implications for aggregate demand.

    4. The way I see it, there is some evidence of a Phillips curve type relationship and evidence that some fraction of economic fluctuations is due to monetary disturbances. My guess is that Prescott would accept this and would point to the Lucas model as an explanation. What is dead is the idea that misperceptions are a leading cause of business cycles, but that's because the proponents of that view have either moved to the RBC explanation or to New Keynesian models.

    5. Krugman may be clued-out but the statements were clearly stupid. Fama and Cochrane claimed that increases in government spending necessarily lead to lower investment via the identity S+T=I+G. This is wrong because the increase in G may lead to an increase in output, hence increasing S and T. Even without government, if S<I and interest rates are fixed then output will increase until S=I.

    Kocherlakota was stupid because he ignores the way the Fed sets interest rates, which involves the liquidity effect. He only focuses on the Fisher effect. Maybe he doesn't even believe in the liquidity effect. In any case for the Fed to keep the Fed funds rate low as the natural real rate increases implies increasing the money supply, in order to purchase securities in order to prevent the increase in nominal rates that would otherwise occur. Kocherlakota implies that the private sector would observe the money supply increasing and would respond by lowering inflation expectations, which is ridiculous.

    Again, the error is to misunderstand the economics behind the identity. In this case, Kocherlakota takes the real rate as given, assumes the Fed sets the nominal rate, and concludes that inflation expectations are just the difference between the two. In fact, the Fed controls the real rate in the short-run, due to the liquidity effect. Low real rates (below the natural rate) lead to higher inflation. In the long run, the real rate must be the natural rate. Hence you end up with the nominal rate being determined by the natural real rate plus the inflation generated by the Fed. In other words equilibrium is only achieved once the Fed allows the nominal rate to rise. To achieve an equilibrium with deflation, the Fed would have to slow down the rate of money growth, which would imply an interim period of high interest rates (a la Volcker).

  8. I'm just wondering if the new generation of active macroeconomics researchers is less scared of questioning the assumptions and theory of RBC, Ricardian equivalence, etc., more willing to look at and learn Keyensian and new Keynsian economics.

    This is as compared to the young and active group from say 10, 15, 25 years ago, when it was a lot less healthy to your career to question RBC/RE and look in Keynesian directions.

    I know from my finance PhD days that 25, 30 years ago in finance it was very unhealthy for your career to not just assume very high market inefficiency, and today it's very different, far easier to publish behavioral/limited rationality work.

    In other words, has the field of macroeconomics gotten more realistic and interested in theory that actually fits reality and is useful for actual positive economic practice/policy than it was 10, 15, 25 years ago?

  9. Anonymous:

    1. If what I write comes across as hostility to Keynesian economics, then I'm not being understood. I think Keynesian ideas are interesting, and it would be nice if they were correct. I'm just not convinced, and I like to ask questions. Coordination failure models were quite tight. I can't say the same for New Keynesian models. I once worked on credit rationing. Is that a Keynesian idea or something else?

    2. "what's important in Woodford's model is monetary policy" An important question is whether Mike has actually captured the essence of central banking in his models. I don't think so.

    3. "According to the Prescott view, that's baloney. So there's a big difference." Not much point in discussing Prescott in this context. He doesn't think about monetary policy - his contribution is elsewhere.

    4. "Where's the proof?" Friedman did not think much about financial intermediation and its economic benefits. 2 problems with 100% reserves: (i) to implement it, you have to say what is "money" and what is not. Impossible, and once I define it, someone will find a clever way to circumvent the requirement (as has been done with sweep accounts). (ii) Reserve requirements are a tax on intermediation, and reduce efficiency, eliminating a whole segment of intermediation activity with 100% reserves if they are effective. I don't know about the "better" men. They must not be so great.

    5. "Kocherlakota was stupid..." Please, don't ever use the words "Kocherlakota" and "stupid" in the same sentence. Maybe we could arrange to have you spend some time with Narayana, and you will see what I mean. Believe me, he understands these issues, and you are misunderstanding him, or putting too much stock in what Krugman says, if you think he doesn't get liquidity effects vs. Fisher effects.

  10. About your post scriptum, a good starting point could be the following Family Tree of Trade Economists:
    which includes PK, his students, etc.. Unfortunately, it does not include EP. If someone is aware of anything similar in other fields (I mean in economics, because in math this stuff has a long-established tradition), please speak up.

  11. 1. I once worked on credit rationing. Is that a Keynesian idea or something else?

    It depends on whether they imply a role for demand management. The defining feature of Keynesian models is that government needs to use fiscal or monetary policy to stablize output and increase welfare. Classical economists think such intervention is ineffective, unnecessary, or counterproductive.

    5. Even a genius can say stupid things. His remarks were unambiguous, and wrong. The only way it makes sense is if he doesn't believe in the liquidity effect, and hence doesn't belive the Fed controls real interest rates. That view may be correct, but it is not how most people at the Fed see things.

  12. Anonymous:

    1. Herein the problem. You seem to be saying that to be truly Keynesian, you have to reverse engineer the policy conclusion. That would just be bad economics from the point of view of some people.

    2. I had read Kocherlakota's speech, before Krugman/Thoma/Rowe/DeLong got on his case, and did not find anything out of the ordinary there. I then wrote this:

    I don't want to rehash the issues, but the ensuing interaction was very strange. Kocherlakota was making some statement about the long run which was very standard Irving Fisher, some people (including Krugman) were willfully sowing confusion, the result being that no one really learned anything.

  13. Max,

    Good start. Someone should do the Minnesota family tree. Actually, I mentioned three generations above, but there are more than that. I am actually third-generation Minnesota (Wallace-Aiyagari-me), and I have grandchildren. Ricardo Lagos would be third generation (Wallace-Wright-Lagos), and he must have students. There would be large trees associated with Prescott-Rogerson or Prescott-Azariadis-Cooper, for example.

  14. 1. No, I'm saying if your view of the empirical and theoretical evidence leads you to Keynesian conclusions, your a Keynesian. Remember, the Keynesian revolution was a reaction to the prevailing laissez-faire policies advocated by the classical economists. The freshwater view, began by Friedman and then continued by Lucas, Prescott, Barro etc. was to resurrect the classical view that government should not attempt to manage the economy.

    Now, I guess which camp you fall in partly depends on your personal outlook and philosophy, but I wouldn't accuse anyone of trying to reverse engineer the policy conclusion. Take Barro, for example. He started out as a Keynesian and rejected the theory on scientific grounds, so he now advocates a classical approach. Or Taylor, who clearly has problems with demand management yet is regarded as a Keynesian.

  15. 2. The problem is he wasn't just talking about the long-run. He said:

    "[Deflation] would require the FOMC to make the surprising mistake of ignoring the long run in its desire to fix the short run."


"central banks then respond to deflation by easing monetary policy in order to generate extra demand. Unfortunately, this conventional response leads to problems if followed for too long."

    "if the FOMC maintains the fed funds rate at its current level of 0-25 basis points for too long, both anticipated and actual inflation have to become negative."


"If the FOMC hews too closely to conventional thinking, it might be inclined to keep its target rate low. That kind of reaction would simply re-enforce the deflationary expectations and lead to many years of deflation."

    In all cases, Kocherlakota was saying that the short-run policy of keeping the Fed funds rate near zero would lead in the long-run to deflation. At best this is a very controversial view, contrary to what most economists believe. At worst it's a simple fallacy. Since he presented it as uncontroversial and didn't present any new evidence to contradict the standard view, one has to assume it's the latter. Indeed, he seems to accept the standard view that the Fed can influence the real rate in the short-run, which pretty much proves his claim was wrong.

    Now, none of this means he is dumb: he may have just misspoke. For all I know, he has by now realized it was wrong and is embarrassed. But this sort of error plays into the story that Krugman is trying to peddle, that freshwater economists are technically brilliant but kinda clueless about the implications of their models for the real world. And when you refuse to acknowledge these mistakes out of some sort of tribal loyalty it just adds weight to Krugman's claims.

  16. "Many people who have been taught by Ed Prescott or have talked to him, have bought into the idea that "demand" and "supply" are the wrong language to be using for organizing our thinking about macroeconomics."

    I really don't get this... in every model where there are well-defined markets, demand and supply are useful constructs to organize thinking. In the RBC model for instance, households supply capital, and firms demand it, households supply labor, and firms demand it. This obsession with labeling demand and supply as old-style-macro objects sounds a bit elitist to me.

  17. anonymous,

    Tribal loyalty my ass. My only loyalty is to good ideas and good economics. If you still don't get this, or are unwilling to try to understand Irving Fisher, I can't help you.


    No, the problem is with the "aggregate demand" and "aggregate supply" language. You write down a model, and then the way you play with it is to ask what happens to the endogenous stuff when you change the exogenous stuff - policy rules maybe. In traditional textbook AS/AD, the AS includes what is going on in the labor market, for example. But of course the labor supply decisions are being made by the same people who are on the demand side in the goods market. What is aggregate demand and what is aggregate supply? Further there is the question of what is exogenous. Some discussion you hear (from DeLong for example) seems to proceed like this. Y = C+I+G, and each of C, I, and G is exogenous. We want Y to be constant, so if C or I falls, you make up for that with more G. I don't find that very helpful.

  18. So, it's official. You delete posts that challange your understanding of economics. So much for being loyal to good ideas and good economics. You obviously place personal loyaltity and pride above scientific integrity.

  19. Sorry, got a little grumpy there. I was losing patience. I tried to make it clear above that I was not really interested in discussing Kocerlakota's speech from last summer. I thought the argument he made was well-reasoned and consistent with what I know about monetary economics. But I can't read the guy's mind, so it's pretty awkward putting myself in his place and arguing as some kind of Kocherlakota representative. It's like playing the White House press secretary or something. Read through my posts, and you'll see me criticizing any number of people who I know well - Kocherlakota, Prescott, whoever. That's actually a Minnesota tradition too. Neil Wallace even likes to criticize himself excessively. Pretty healthy, don't you think?

  20. Steve Williamson: You are a genius!