Thursday, June 28, 2012

Economic Nonsense

In 1991, Paul Krugman published "Increasing Returns and Economic Geography," which the Nobel Prize committee cited in 2008 in support of Krugman's Nobel Prize in Economics. The paper is basically an exercise in normal economics. There are optimizing consumers and firms; there is a Dixit-Stiglitz monopolistic competition structure; there is an equilibrium in which prices adjust to equate quantities supplied and demanded in each market. If Krugman had submitted the paper for the 1991 SED meetings, I'm sure he would have been on the program, in spite of the absence of dynamics in the model. Indeed, no one would have had a problem calling Krugman's work macroeconomics - it fits well within the set of what modern macroeconomists like to think about.

Fast forward to 2012, and we all know what Krugman stands for, and he's been even more vociferous about these things over the last week or so than usual. One of his blog posts was this one, which is his response to some policy discussion in Britain about the state of macroeconomics. Part of what is in that piece, and other recent ones, is the argument - which we have heard from Krugman many times before, and will likely hear ad infinitum - is that the Hicksian IS-LM model is "spectacularly successful." I don't know whether to laugh or cry when I read things like that. How could anyone think of that crude tool as a success in this context? What could Krugman be thinking? Does the IS-LM model tell us what a financial crisis is, and what the policy response to such an event should be? Does it tell us what quantitative easing does? Does it tell us the extent of inefficiency in the US economy that monetary or fiscal policy might correct? Does it help us understand sovereign debt problems? Does it help us understand how to regulate the financial system? Of course not! Get a life, Krugman!

Krugman asks:
So why the sense that macroeconomics is a mess?
The answer to that question is, of course, that Krugman is doing a great job of propagating the fiction that it's a mess. But it's not. When I went from session to session at the recent SED meetings, I got the sense of a group of productive, young (for the most part), and thoughtful macroeconomists, working on, and making progress in understanding, the key economic problems of our time. Here's an example. In the Friday, 4-6 PM session on Macro/Labor, there are a couple of papers - one by Violante and coauthors, and the other by Alvarez/Shimer. The Violante paper makes some progress in defining what "mismatch" is, and measuring the extent of it. The Alvarez/Shimer paper is a study of how human capital accumulation matters for unemployment in the context of sectoral realloaction. Both papers are highly topical and teach us something about what is going on currently in the US economy.

The first key point is that macroeconomists are not in disarray. They are pursuing their research programs in a sound and coherent fashion, debating the issues, and making progress.

In case you were concerned about Krugman's focus on modern macroeconomists, he doesn't have much use for theorists either:
The other thing I’d like to say is that the notion that microeconomics is in much better shape is questionable, to say the least. I mean, it’s not as if the assumptions underlying standard micro theory are, you know, true – utility maximization? Really? Micro is consistent in a way macro is not, but for the most part it’s best viewed as a metaphor that’s helpful as long as you don’t take it too seriously.
The truthiness of utility maximization? I know I have to explain that to an Econ 101 student, but not to Paul Krugman. The first thing Krugman does in laying out the model, on page 488 of this paper is to specify a utility function, which is the objective function for all the individuals in Krugman's fictitious environment. Apparently he didn't feel the need to footnote that, and make excuses for, you know, the lack of truthiness.

The second key point is that modern "macroeconomists" do not actually think of themselves as different from "micreconomists." It's all economics, using the same set of tools - the tools that the 1991 Krugman could use to great effect.

Now, after insulting most of the economics profession, Krugman would like us all to sign on to his Manifesto for Economic Sense. For the most part, I think monetary policymakers do a good job of absorbing economic research and applying the ideas. For example, the FOMC is not perfect, but there are some good minds in the room when the committee meets. Fiscal policy leaves a lot to be desired. But injecting a dose of Krugmanite IS-LM into the mix is not going to help. I don't know about you, but I'm not signing up.

27 comments:

  1. Excellent post except that this: "Fiscal policy leaves a lot to be desired" was a good enough reason for me to sign it. One does not have to adhere to the IS-LM model to agree with the core of the manifesto.

    As for Krugman, he reminds me of Dr. Jekill and Mr. Hyde. Mr Hyde-Krugman uses the angry, populist rhetoric that has granted him mesiah status among his followers, while Dr. Jekill-Krugman feels guilty and tries to mend things on occasion. The question is, will, as in the novel, Dr. Jekill be lost forever to Mr. Hyde?

    ReplyDelete
  2. ""Fiscal policy leaves a lot to be desired" was a good enough reason for me to sign it."

    UK electors put the Conservative government in power. What were they expecting? US electors gave the Republicans control of the House, and would have given them control of the Senate in 2010, except that only a third of the Senate could have turned over. What were those electors expecting? First, political realities are such that Krugman and company can write all the manifestos they want and it will have no effect on fiscal policy, in either the UK or the US, which seem to be the chief targets. Second, the idea seems to be that we should do something - anything - even if we don't have a clue why. That makes me nervous.

    ReplyDelete
  3. The idea that we should do anything makes me nervous too. But don't you think academic economists have a responsibility to inform the electors about the nature of the problems they face (with the manifesto being one way of doing so)? Finally, I noticed that Chris Pissarides, among others, has signed it.

    ReplyDelete
    Replies
    1. "But don't you think academic economists have a responsibility to inform the electors about the nature of the problems they face..."

      Yes, and one of the problems we face is Paul Krugman.

      Delete
  4. Steve,

    I think that PK feels obliged to explain the virtues of having a clearly defined objective function to a wider audience, but knows it is unnecessary to do so in the JPE. PK understands utility maximization, but he is taking time to explain it to his readers, just like you would to intro students.

    ReplyDelete
    Replies
    1. Krugman appears to have forgotten what he knew. What he explains to his readers is that modern macro is all wrong, and his version of 1937 macroeconomics is right.

      Delete
  5. http://www.brookings.edu/~/media/Files/Programs/ES/BPEA/2012_spring_bpea_papers/2012_spring_BPEA_stockwatson.pdf

    Stock and Watson:

    Third, focusing on the recovery subsequent to the 2009Q2 trough, we estimate that slightly less than half of the slow recovery in employment growth since 2009Q2, compared to
    pre-1984 recoveries, is attributable to cyclical factors (the shocks, or factors, during the recession), but that most of the slow recovery is attributable to a long-term slowdown in trend employment growth. Indeed, the slowdown in trend employment growth is dramatic: according to our estimates, trend annual employment growth has fallen from 2.4% in 1965 to 0.9% in 2005.

    The explanation for this declining trend growth rate which we find the most compelling rests on changes in underlying demographic factors, primarily the plateau over the past decade in the female labor force participation rate (after rising sharply during the 1970s through 1990s) and the
    aging of the U.S. workforce. Because the net change in mean productivity growth over this
    period is small, this slower trend growth in employment corresponds directly to slowdown in
    trend GDP growth. These demographic changes imply continued low or even declining trend
    growth rates in employment, which in turn imply that future recessions will be deeper, and will have slower recoveries, than historically has been the case. In other words, jobless recoveries will be the norm.

    SW, IOW, we could let people print money on any available HP, laser or ink jet, whenever they need it, and still not risk inflation.

    Sign up!!!!!!!!!!!!!

    ReplyDelete
  6. maybe, but as Noah Smith aptly notes, really no theories or assumptions or models have been conclusively falsified (or more appropriately, no one is yet willing to admit it). A discipline where no one is ever wrong is not disciplined at all.

    ReplyDelete
    Replies
    1. Economics is hard. Even in sciences where experimental approaches are feasible, there can be fierce debates about what the experiments actually tell us. In economics typically all we have is the theory, and the available data, a lot of which is pretty crappy. That said, in the academic community I see plenty of honest work, people willing to learn, and weeding out of inferior theory. One of the inferior theories that got weeded out was IS-LM. No serious economists think about that thing any more.

      Delete
  7. As I pointed to Noah, the Popperian principle of falsification does not describe conduct in any science. Kuhn showed that in the end theories are not abandoned when they fail to conform fully with the data, but when more useful theories become available. Having said that, isn't the quantity theory taught in undergraduate macroeconomics (the one with well-defined money, output constrained by supply, and constant velocity) just as bad as IS-LM? Why is there no revolt against that?

    ReplyDelete
    Replies
    1. The quantity theory was done in too. Friedman was the most prominent quantity theorist, and he derived a policy recommendation from it, which is in his 1968 Presidential address. Central banks around the world took that to heart, and it didn't work. It's hard to find a quantity theorist these days. Most of them are retired.

      Delete
  8. Yes, but quantity theory is still being taught and some of our graduates (and electors like Ron Paul's supporters) still take it to heart.

    ReplyDelete
  9. I certainly don't teach it, except as history of thought.

    ReplyDelete
  10. I think I need to take another look at your Intermediate Macro textbook! :)

    ReplyDelete
  11. Random thoughts on limited self-correction in Macro, for the sake of controversy:

    1. The quantity theory is wrong because velocity isn’t constant. Velocity fluctuates with liquidity preference, which rises with the level of our “disquietude.” Isn’t this an elegant illumination of why a massive increase in the monetary base hasn’t produced inflation or an increase in investment?

    2. Unemployment only requires the existence of an asset that can’t be produced by labor, e.g., money.

    3. New classical theorists have helped themselves to GE theory without paying much attention to the processes that lead from disequilibrium to GE.

    4. New classical theorists use production functions that ignore the “Cambridge critique” and Paul Samuelson’s generous concession that Anglo-Italians were right.

    5. Rational expectations theorists have ignored Kenneth Arrow’s point that, outside GE, RE loses its plausibility.

    ReplyDelete
  12. Greg,

    1. The center of the quantity theory is the money demand function. Friedman's claim was that money demand was a stable function of a few variables, and he went from there. You don't need constant velocity. Of course the problem is that, no matter how you define "money," you money demand function will not be stable. It's pretty easy to see why the quantity theory is not helpful, particularly under the current circumstances. But that's beating a dead horse. Who is talking quantity theory at the moment?

    2. That statement makes no sense.

    3. All we have is equilibrium theory. "Disequilibrium" is meaningless. In fact, any Keynesian theory has a clear notion of equilibrium.

    4. Everyone uses production functions. The fact that aggregation is a thorny issue doesn't make modeling using a production function useless.

    5. RE is a useful tool. Live with it.

    ReplyDelete
    Replies
    1. Stephen,
      3) What about various post-keynesian models, Leijonhufvud's general disequilibrium, etc?

      Delete
  13. Stephen,

    1. “Who is talking quantity theory at the moment?” My point was that the rejection of the quantity theory should have been accompanied by a renewed appreciation of Keynes’s point that intractable uncertainty plays a large role in the demand for money.
    


    2. “That statement makes no sense.” Try this, then: Unemployment develops “because people want the moon; -- men cannot be employed when the object of desire (i.e., money) is something which cannot be produced and the demand for which cannot be readily choked off” (GT, 235). Pretty good explanation of where we are now. No?

    3. “All we have is equilibrium theory. ‘Disequilibrium’ is meaningless. In fact, any Keynesian theory has a clear notion of equilibrium.

”

    This doesn’t really address the point I raised, which is that insufficient attention has been paid to the process by which you get from one equilibrium to another. See, e.g., Fisher’s “Disequilibrium Foundations of Equilibrium Economics,” SMD, and many of Frank Hahn’s papers.

    4. “Everyone uses production functions. The fact that aggregation is a thorny issue doesn't make modeling using a production function useless.” It’s not just aggregation problems, it’s re-switching and all that.

    "The phenomenon of switching back at a very low interest rate to a set of techniques that had seemed viable only at a very high interest rate involves more than esoteric difficulties. It shows that the simple tale told by Jevons, Böhm-Bawerk, Wicksell and other neoclassical writers — alleging that, as the interest rate falls in consequence of abstention from present consumption in favor of future, technology must become in some sense more 'roundabout,' more 'mechanized' and 'more productive' — cannot be universally valid" (Samuelson, “Summing Up”).

    But things go forward as if nothing happened.

    5. “RE is a useful tool. Live with it.” Is it? I think it rather blinds us to the fact that people hold different expectations about future prices, quantities that can be sold at these prices, etc.

    Apologies for the length, but I thought you might enjoy the quotations.

    ReplyDelete
  14. Greg-

    Let's say I wants to know how the extension of unemployment insurance will alter employment rates, consumption, investment, and output, all else equal. Let's now say that we want to think about people who will, if we radically change the rules, think about how to best respond to it. And let's try not to allow the modeler to choose the expectations of households in ways that give them wahetevr answer they want--so we restrict them to be at least not be routinely wrong. With these caveats, let's say we want a quantitative answer--e.g. "a 6 month extension will raise/ lower x, y,z, and q by x',y',z', q'.

    What would you propose, and why it is better than what Violante, e.g., is doing right now?

    People are trying to answer questions. You are telling us that they can't be answered the way we are trying. Granting taht you're right, put a model on the table. If it aids your vanity to know that we're wrongheaded, that's fine. But unemployed people deserve better.

    So either help us get better answers--propose them to economists, send your work to conferences just like the rest of us, and the send and revise your work according to the demands of unreasonable and anonymous referees, just like the rest of us--or please leave us alone.

    ReplyDelete
  15. "according to the demands of unreasonable and anonymous referees"

    That's a good one! :)

    ReplyDelete
  16. " Apparently he didn't feel the need to footnote that, and make excuses for, you know, the lack of truthiness."

    Hmm, you seem to forget that the paper was written for a particular cult that believes that (1)utility maximization is truthful (2) if it is not truthful, it is the most important ingredient to understand decision.

    By the way you better name your blog "The anti-PK manifesto" Not that I agree with his analyses; but his and your approach: theory deprived of stable empirical anchors, tests of falsification, relevance, and substance in pursuit of simplicity is destined to fail by the standard of real world. I understand the game of publishing and doing economics. What I cannot understand is the sheer confidence in the belief that the real world must be incorrect if it is not consistent with theory of optimization and cold blooded strategic thinking. It won't be long before we witness economics joining the rank of astrology unless it makes itself relevant.

    ReplyDelete
  17. To Anonymous:

    By your reasoning civil engineering should also join the rank of astrology since it is based on the falsified Newtonian theory. The point is this. Optimization yields some interesting results. It explains why consumers buy less and producers offer more of a product when it becomes more expensive. It explains why Alaskans do not increase their consumption spending when they receive payments from Alaska's Permanent Fund, and why mutual funds managed by highly-paid experts do not outperform index funds. So regardless of whether each individual truly optimizes, theories based on optimization have strong explanatory power. PK knows that, and this is why he used them. Like with Newtonian mechanics, it is true that theories based on optimization cannot explain certain phenomena. But unless you can come up with a theory not based on optimization that has even more explanatory power, then you are just another Monday morning quarterback!

    ReplyDelete
  18. CA,

    Quick question: if index funds outperform actively managed funds, which they do, why do so many investors, including virtually every pension fund, continue to invest in actively managed funds?

    p.s. Yes, I do agree that optimization is useful in some contexts.

    ReplyDelete
  19. Greg,

    the best I can do is convey the responses I have gotten from fund managers themselves. Many say that what they really provide is education and psychological support, especially during troubled times. They highlight investment strategies to their clients, explain to them why bad things happened when they happen, and talk about how and why things will get better. You can't get that with an index fund. Others take advantage of the fact that their clients are not well informed about how much it costs them to have the fund actively managed and the (lack of) benefit from doing so. Finally, some people, like myself, were not given an option by their employer to choose an index fund.

    ReplyDelete
    Replies
    1. CA,

      I think you're right about the psychology. It does, however, pose a challenge for an optimization approach. I suppose one could say that many "investors" are trading off higher returns for the peace of mind offered by investment managers (whether this peace of mind is well-founded (rational) or not.

      I think persuasion, to use a polite term, plays a big role too. But once you admit the effect of persuasion/advertising, etc., then you're at the limits of those paradigms that envision even modestly informed agents who come to the market with a pretty good idea of what they want (i.e., where advertising is just supplying information).

      p.s. There is a firm, IFA, that provides advice along with a menu of index fund portfolios.

      Delete
    2. Greg,

      I have no problem conceding that. But isn't this what bounded rationality models that also assume optimization (under cognitive constrains and costly information) are all about? It seems that any behavior can be modeled as the result of optimization if the "right" constraints and objective functions are chosen. That's what makes the assumption so powerful. As of now I do not see a better way of modelling decision-making. In any case I am sympathetic to the viewpoint sxpresed by Caballero in the Journal of Economic Perspectives, that someone linked in the post following this one.

      Delete