Monday, October 10, 2011

Sargent and Sims

I woke up this morning to Per Krusell's voice on the radio, telling me that Thomas Sargent and Christopher Sims had won the 2011 Nobel Prize in Economics. Excellent!

Sargent, along with Neil Wallace, was among the first macroeconomists to recognize that Robert Lucas had done something important in 1972, and helped the rest of the profession understand that by developing the ideas. Sargent, Wallace, and Sims were instrumental in developing, in the 1970s, a model for cooperation in economic research between academics and central bankers at the Federal Reserve Bank of Minneapolis. Minnesota macro has since had a huge influence on the profession, and on the practice of central banking.

Both Sargent and Sims brought a strong econometric tradition to macroecononomics. Sims's work on vector autoregressions, beginning with Macroeconomics and Reality has been highly influential, and you can see Sims's influence in how people like Marty Eichenbaum, Larry Christiano, and Jordi Gali, for example, do their work. Modern quantitative work in macroeconomics, among New Keynesians and non-Keynesians alike, includes both estimation and calibration (from Prescott), a state of affairs I think Sargent and Sims are pleased with.

Sargent has been a key proponent of the use of mathematics and technical developments in other fields in macroeconomics, from dynamic programming methods to frequency domain techniques to robust control. In part, he has promoted the use of these techniques in several generations of textbooks for economics graduate students. Indeed, Sargent's key influence has been through his students. Any Sargent student can tell you about the "Sargent reading group," how it works, and how much they learned from it.

Both Sargent and Sims are economists with extremely high technical ability, but with brilliant insight into economic ideas and economic modeling. Sims is not only a top econometrician, but has made key contributions to the study of the fiscal theory of the price level and rational inattention.

The Nobel committee chose well this year. I think all of us should be pleased.


  1. I am ecstatic about Sargent and Sims' win.

    Very well deserved. I've been rooting for these guys for years.

    When Lucas got it, I was afraid that Sargent had been permanently passed over.

    Excellent choices.

  2. With Sargent, it's hard to isolate one or two key contributions (a key difference from Lucas), but his influence is very important. He doesn't go around tooting his own horn, and will typically be self-deprecating, often emphasizing what Wallace or Sims did, rather than his own contributions.

  3. Sims's influence is similarly diffuse. Example, see footnote 1 in Kydland and Prescott's "time consistency" paper:

    "The original objective of this research was to demonstrate the applicability of optimal
    control methods in a rational-expectations world. We recognized the nonoptimality of the
    consistent solution obtained by using control-theory techniques, but initially considered
    this a minor problem. Further thought, in large part motivated by C. A. Sims's criticism
    of our initial analyses, led us to the radical conclusions of this essay."

  4. My thinking was the same as Chris's above r.e. Sargent not being named when Lucas received it. In any case, for both Sargent & Sims, this is long overdue.

  5. Sargent and Sims are both outstanding theoreticians and technical innovators, but their receiving the award should not be interpreted as a vindication of "freshwater" macro. This is mostly a prize to the development of new technical tools, not of new economic ideas.

  6. They gave it to three people last time. Why not Wallace?

  7. "...the award should not be interpreted as a vindication of "freshwater" macro."

    That's OK. So-called "freshwater macro" does not need "vindication." Further, I think it's hard to separate the technical tools from the ideas, don't you think? There was a powerful set of ideas that came out of the Minnesota macro program that these people helped found. Why do you want to deny that?

  8. wonks anonymous,

    If it were up to me, I would have given Wallace the prize a long time ago, and I think Sargent would say the same. However, not everyone in the profession is aware of Wallace's contributions, and people who are aware don't necessarily get as excited about them as I do.

  9. I've always thought of Sims's research program as closer to Woodford than Prescott. But, then again, we are all freshwater now, aren't we?

  10. If by freshwater you mean skepticism about the non-neutrality of money, the believe that fiscal policy is never good, the neglect of short-run dynamics and its causes, or thinking that all that matters are real determinants, then no, we are not all freshwater now. This does not undermine the vale of the methodological contributions by freshwater schools. Economic ideas (the importance of expectations and incentives) is older than freshwater: that is Friedman and Phelps.

  11. "I've always thought of Sims's research program as closer to Woodford than Prescott."

    Yes, certainly that's true more recently, and to some extent long ago as well.

  12. the scientific backgroung piece by at:

    uses a New Keynesian model to illustrate Sargent's structural estimation method (section 3).

    Clearly they are recognizing his technical contributions, not his economic ideas...

  13. I'm not sure what you are trying to say. The document you link to puts Sargent's work squarely in the research program of Phelps and Lucas. The passage you seem to have in mind is this:

    As a result of this new research program, macroeconomics changed course
    rather drastically. Views on policy were transformed, and the awards to Lu-
    cas and Phelps were largely motivated by the policy implications of their
    work. Views on business cycles also changed, due to the contributions of
    Lucas and Sargent as well as later work by Kydland and Prescott. More
    recently, Keynesian ideas Keynesian ideas have been revived in a “New Key-
    nesian Macroeconomics”, which builds directly on Kydland and Prescott,
    with the addition of various frictions such as sticky prices and wages. Mod-
    ern empirical macroeconomic research relies heavily on structural-estimation
    methods, of which Sargent is the main architect.

    This all puts his work in context. But again, I don't see how you separate the methods from the ideas in the research program, of which Sargent was a big advocate. What exactly are you trying to say?

  14. I am just pointing out that section 3.2 of the document (p. 25) uses a simple New Keynesian model to illustrate the structural estimation methods devised by Sargent. Sargent is obviously not a New Keynesian economist and would not agree with the implications of models proposed by New Keynesians. Hence, it seems that the prize is mostly celebrating his technical insights and less his ideas about the working of the macroeconomy. I think this is also true for other "freshwater" macroeconomists that have been awarded the prize.

  15. That's a strange idea. If A uses B's idea, and B gets a prize for his/her idea, but B disagrees with A, then we have to say that B got the prize not for his/her idea, but for his/her "technical insights."

  16. Well deserved prize, but does anyone else feel like Neil Wallace is the odd man out?

  17. Yes, the way the prizes are given out now - 2 or three at a time - it's now harder to give it to Neil Wallace.

  18. Why do economic models always support the political biases of the modelers?

  19. Benjamin keeps flogging that horse, but he isn't getting any smarter.

  20. Benjamin,

    Bold statement. Always? In this case, what do you think the political biases of the modelers happen to be? Do you know how Tom and Chris vote?

  21. Benjamin,

    What are the political biases of Sargent and Sims?

    And, more provocatively, even if you were correct, why would that matter? If such a model is deeply flawed because of political bias, why would that model or framework become so successful that it dominates the profession and earned the practitioner a Nobel prize? (Assuming again, that you are referencing Sargent and Sims)

    I can think of a particular Nobel prize winner who likes to talk about a model, the assumptions of which are entirely based on his political bias, that the profession has discarded to the trash bin. (His name isn't Sargent or Sims.) This anecdote certainly fits my meme. Do you have counter-examples?

  22. Why is such an observation important?

    Okay, economists want the public and policy-makers to accept insights based on models.

    But when an independent observer (me!) notices that models from AEI-freshwater types always come to certain conclusions, and models from saltwater, Brookings types always come to a different set of conclusions, what am I supposed to believe?

    I believe economists carefully craft their models--what they choose to measure, what start and finish dates, what weight to give different data--to support their biases. I further assume if a model comes up with the "wrong" result, it is forgotten or tinkered with until the right result comes out.

    That said, I this this bias extends to both halves of the aisle.

    I am also dubious about the quality of the data. A fragile, esoteric model may be based on flawed data.

    One parting note: All of us would like to think the whole double-blind FDA testing protocols are sound, and on paper they look good.

    The sad result is that many drugs make it all the way through the multiple test stages, but then prove ineffective on the general population. And this is in a field where one could argue the immediate data is better--it is harder science.

    I like math, especially statistics. But esoteric and fragile economic models strike me as very dubious.

    I hate to sound like a Luddite of the Austrian School, or someone content with their own bias, no matter what the models say.

    But increasingly I feel the models are just false armor to shield personal biases, in an attempt to look scientific.

    On top of that, right now the Fed should print a lot of money. Like Milton Friedman told the Japanese in the 1990s.

    Are there any examples of a major nation, after a real estate-recession, rapidly emerging from contraction while keeping inflation under two percent? Japan? The USA 2008-2011?

  23. Clearly you are not that familiar with the models. Being scientific just means that you might have to work hard to justify a particular government intervention. The basic modeling approaches that modern macroeconomists use have no particular policy implications. For example, a basic real business cycle model says you can explain some key features of business cyles in a model where there is no suboptimality the government needs to be correcting. That doesn't say there is nothing for the government to do, as we are left with the same externalities and distortions that we have always known imply that the government should be intervening. Indeed, Woodford took basic RBC, introduced sticky prices, and developed a framework for thinking about optimal monetary policy. Some people are curious scientists, and they use models and data to learn about the world and how it works. Other people seem convinced of particular "truths" and use models and data to try to convince us. So what? We can thoughtfully evaluate the arguments of both sets of people and we sort them out and make progress. Why so cynical?

  24. "But when an independent observer (me!) notices that models from AEI-freshwater types always come to certain conclusions, and models from saltwater, Brookings types always come to a different set of conclusions, what am I supposed to believe?"

    Neither AEI nor Brookings use models. Don't throw real scientists -- either from salty or freshy schools -- in with the political claptrap spewed by AEI. What they do with our models is not our concern.

    I'd say you should know better, but you seem to be an idiot.

  25. "AEI-freshwater types"

    Who are these freshwaters anyway? Bob Hall made this up in the 1970s to apply to people who worked at Chicago, Rochester, and Minnesota, primarily. Maybe add Carnegie-Mellon. Even then, those schools were not monolithic - the individuals at any one of those places found plenty to argue about. I don't know what "freshwater" is supposed to mean in the generations that followed Lucas's. I don't know anyone who is allied with AEI.

  26. Stephen:

    Okay, I will try to be less cynical. Happens when you get bald.

    I am not cynical, but indeed earnest and optimistic, about the direction that Market Monetarism is taking.

    And I don't understand the near obsession with maintaining ultra-low rates of inflation. It seems to have become a religion. There is a moral value imputed to zero inflation, as if it is the primary goal alone and by itself.

    I think prosperity, growth, innovation, and economic and commercial freedoms are even more important.

  27. Benjamin,

    The so-called Market Monetarists, as I understand them, do not want more inflation -- they want stable inflation expectations (or at least in the case of Scott Sumner, stable expectations of nominal income) and an explicit target for monetary policy. To the extent to which there is validity to their argument, you are doing a disservice to it.

    Also, does the fact that I think sticky prices are unimportant mean that a model I construct is biased? It is not clear that sticky prices imply anything important. See this:

    There are a bunch of ad hoc assumptions baked into the New Keynesian framework (e.g. the Calvo fairy). If I think that models should be based on individuals who are not "veritable slugs" to use the phrase of Williamson and Wright, I will model things differently and get different conclusions. Does that mean that I am biased?

  28. "And I don't understand the near obsession with maintaining ultra-low rates of inflation. It seems to have become a religion. There is a moral value imputed to zero inflation, as if it is the primary goal alone and by itself."

    We know, it is clearly apparent you don't understand much. Inflation is costly and therefore should be low, if possible. A wide range of models imply optimal inflation rates at or below zero. Not bias, you arrogant twit, science.

    "I think prosperity, growth, innovation, and economic and commercial freedoms are even more important."

    Nobody cares what you think is important. The models say that changes in the growth rate of the economy swamp everything in terms of welfare, so perhaps you've accidentally hit on something here. The rest of your preferences aren't even well-defined objects -- what is "economic freedom" anyway? Usually that is code for "low taxes", which may or may not be optimal, and spouted by dimwits like Ron Paul and Grover Norquist.