Friday, April 29, 2011

What I learned in econ grad school

I always find it interesting that criticisms of economics education focus more on the graduate side than the undergrad. Consider this broadside by Brad DeLong and Larry Summers:
For Summers, the problem is that there is so much that is “distracting, confusing, and problem-denying in…the first year course in most PhD programs.” As a result, even though “economics knows a fair amount,” it “has forgotten a fair amount that is relevant, and it has been distracted by an enormous amount.”...

The fact is that we need fewer efficient-markets theorists and more people who work on microstructure, limits to arbitrage, and cognitive biases. We need fewer equilibrium business-cycle theorists and more old-fashioned Keynesians and monetarists. We need more monetary historians and historians of economic thought and fewer model-builders. We need more Eichengreens, Shillers, Akerlofs, Reinharts, and Rogoffs – not to mention a Kindleberger, Minsky, or Bagehot.

Yet that is not what economics departments are saying nowadays.
This is interesting because, as someone who never studied econ as an undergrad (I was a physics major), I learned everything I know about macro from my grad courses. If there is an aspiring economist out there whose understanding of macro has been hurt by an overly narrow graduate curriculum, it would be me.

So, what did I learn in my first-year graduate macro course at the University of Michigan?

My first semester was on business cycle theory. (the second semester was all growth theory). We spent a day covering the basic history of the field - the neoclassicals, Keynes, Friedman, Lucas and the RBC people, and finally the neo-Keynesian movement. I recall reading the Summers vs. Prescott debate but not really getting what it was about. From then on it was all DSGE. We did the Ramsey model and learned about Friedman's Permanent Income Hypothesis. We spent a lot of time on RBC. We took a big break to learn value function iteration and how to numerically solve DSGE models by fixed-point convergence. Then we did Barro's model of Ricardian Equivalence, learned a basic labor search model, briefly sketched a couple of ideas about heterogeneity, touched on menu costs, and spent a good bit of time on Q-theory and investment costs. Finally, at the very end of the semester, we squeezed in a one-week whirlwind overview of Calvo Models and the New Keynesian Phillips Curve...but we weren't tested on it.

This course would probably have given Brad DeLong the following reasons for complaint:

1. It contained very little economic history. Everything was math, mostly DSGE math.

2. It was heavily weighted toward theories driven by supply shocks; demand-based theories were given extremely short shrift.

3. The theories we learned had almost no frictions whatsoever (the two frictions we learned, labor search and menu costs, were not presented as part of a full model of the business cycle). Other than Q-theory, there was nothing whatsoever about finance* (Though we did have one midterm problem, based on the professor's own research, involving an asset price shock! That one really stuck with me.).

At the time I took the course, I didn't yet know enough to have any of these objections. But coming as I did from a physics background, I found several things that annoyed me about the course (besides the fact that I got a B). One was that, in spite of all the mathematical precision of these theories, very few of them offered any way to calculate any economic quantity. In physics, theories are tools for turning quantitative observations into quantitative predictions. In macroeconomics, there was plenty of math, but it seemed to be used primarily as a descriptive tool for explicating ideas about how the world might work. At the end of the course, I realized that if someone asked me to tell them what unemployment would be next month, I would have no idea how to answer them.

As Richard Feynman once said about a theory he didn't like: "I don’t like that they’re not calculating anything. I don’t like that they don’t check their ideas. I don’t like that for anything that disagrees with an experiment, they cook up an explanation - a fix-up to say, 'Well, it might be true.'"

That was the second problem I had with the course: it didn't discuss how we knew if these theories were right or wrong. We did learn Bob Hall's test of the PIH. That was good. But when it came to all the other theories, empirics were only briefly mentioned, if at all, and never explained in detail. When we learned RBC, we were told that the measure of its success in explaining the data was - get this - that if you tweaked the parameters just right, you could get the theory to produce economic fluctuations of about the same size as the ones we see in real life. When I heard this, I thought "You have got to be kidding me!" Actually, what I thought was a bit 

(This absurdly un-scientific approach, which goes by the euphemistic name of "moment matching," gave me my bitter and enduring hatred of Real Business Cycle theory, about which Niklas Blanchard and others have teased me. I keep waiting for the ghost of Francis Bacon or Isaac Newton to appear and smite Ed Prescott for putting theory ahead of measurement. It hasn't happened.)

Now keep in mind, this was back before the financial crisis, at the tail end of the unfortunately named "Great Moderation." When the big crisis happened, I quickly realized that nothing I had learned in my first-year course could help me explain what I was seeing on the news. Given my dim view of the standards of verification and usefulness to which the theories I knew had been subjected, I was not surprised.

Around that time, I started teaching undergrad macro (under Miles Kimball and others), and was instantly struck by the disconnect between what I was teaching and what I had learned. Intro macro had a lot of history. Explication was done with simple graphs rather than calculus of variations. And undergrad macro was all about demand - never once did I utter the words "technology shock" in class. We taught Keynes and Friedman. Minsky got a shout-out, and we spent a whole week on the fragility of the financial sector, in addition to the week we spent analyzing the 2008 crisis.

In other words, Brad DeLong would probably have approved of the macro course I taught. He would probably think that the bankers, consultants, managers, executives, accountants, and policy researchers who even now are going through life looking at the economy through the lens of that intro macro class have been reasonably well-served by their education.

But all the same, I absolutely don't blame the grad-level professor for teaching what he taught. Our curriculum was considered to be the state of the art by everyone who mattered. Without a thorough understanding of DSGE models and the like, a macroeconomist is severely disadvantaged in today's academic job market; if he had spent that semester teaching us Kindleberger and Bagehot and Minsky, our professor might have given us better ways to think about history, but he would have been effectively driving us out of the macroeconomics profession.

Thus, DeLong and Summers are right to point the finger at the economics field itself. Senior professors at economics departments around the country are the ones who give the nod to job candidates steeped in neoclassical models and DSGE math. The editors of Econometrica, the American Economic Review, the Quarterly Journal of Economics, and the other top journals are the ones who publish paper after paper on these subjects, who accept "moment matching" as a standard of empirical verification, who approve of pages upon pages of math that tells "stories" instead of making quantitative predictions, etc. And the Nobel Prize committee is responsible for giving a (pseudo-)Nobel Prize to Ed Prescott for the RBC model, another to Robert Lucas for the Rational Expectations Hypothesis, and another to Friedrich Hayek for being a cranky econ blogger before it was popular. 

If you want to change economics education, it is to these people that you must appeal. The ghost of Francis Bacon, unfortunately, is not available for comment.

*Update:  I now recall that we also learned the Consumption Capital Asset Pricing Model (CCAPM). So that was about finance too.

Update 2: In my second year I took a macro field sequence, which taught me all about demand-based models, frictions, heterogeneity, and other interesting stuff. I don't want to make it sound like graduate school taught me nothing about how to understand the taught me plenty. It just all came in the field course...

Update 3: I've decided to remove the professor's name from this post. Although I tried to make it clear (and it should be obvious anyway) that one teacher is in no way responsible for the problems in the field of macroeconomics, I am still worried that some readers might interpret the post to reflect negatively on him, which is the last thing I want.

Update 4: I added a sequel to this post, describing what I learned in my second year.


  1. Nice post. I'm assuming you didn't just do Macro though, right? What about econometrics, microeconomics and other courses? How were they?

  2. Anonymous9:49 PM

    Moment matching and calibration are falsifiable. Ergo, testable.

    Bacon and Popper sleep well, even if University of Michigan student's don't understand what science is.

  3. Disclosure: Niklas Blanchard is not a RBC theorist, or practitioner...or even necessarily a fan of the approach to macro.

    However, ~98% of economic fluctuations in history were real business cycles (i.e. productivity/technology/wealth effect shocks). For the ~2% remaining, I say monetary disequilibrium.

    Note: I think that N-K menu costs, and long-run sticky wages as just as patently ridiculous as "great vacation" explanations of the Great Depression, or Recession.

  4. Alexander11:45 PM

    Moment matching and calibration do not qualify as the kind of rigorous empirical testing that someone like Popper would have wanted from a scientific discipline.

    If economists were really interested in testing macro models, they could derive the likelihood functions implied by DSGE models and use full-information maximum likelihood methods to test them. But we don't do that, and we all know why. As Prescott famously told Sargent, "your likelihood ratio tests are rejecting too many good models."

    Theorists arguably had good reasons to proceed as they did from the 1980s onward -- putting theory before measurement, and restricting empirical testing to relatively weak calibration and GMM exercises rather than full-blown tests of the models' statistical implications. But it's not wrong for people to ask whether this approach demands too little from the models in terms of empirical relevance.

  5. Wow, two out of five comments from freshwater dead-enders.

    I don't think there's any point in debating these people. The question, rather, is how do we come up with something better -- how do we turn that undergraduate macro (back) into a full-fledged research program?

  6. What a fantastic post!
    I couldn't agree more - I'm doing a PhD on Business Cycles in India, and the absence of common sense in research on the topic is striking (especially in recent decades) - and your essay is reaffirmation of the fact.
    Economics can be divided into a tale of two eras actually - there was the Era of Idea Generation (which saw a premature death around 1960) and we now live in the Era of Idea Validation (and this couldn't die soon enough as far as I'm concerned).

    1. Wow! Ashish, my thoughts precisely! . Idea Validation also explains why behavioral economics is coming to the fore front and there's been a cry to return to the essence of economics by doubting and testing the very principles considered imperative to Economics like the Rational Choice Theory.

  7. Anonymous12:59 AM

    As far as taking RBC models seriously. Mark Watson had a great paper in the early 90s that showed that if you evaluate RBC models by almost any halfway reasonable statistical criterion of fit, they failed spectacularly (Measures of Fit for Calibrated Models, Journal of Political Economy, Vol. 101, No. 6, 1993, pp. 1011-1041).

    Of course the response was predictable, along the lines of (In Lucas' words) "the econometric [likelihood ratio] tests were rejecting too many good models". I still find that an amazing statement ....

  8. Actually, I think the most influential "positivist" or reality-denial argument can be found in Friedman's 1953 bundle, where he basically says that reality should never get in the way of economic dogma/his own theories about wealth redistribution up the pyramid.

  9. In my first-year PhD sequence we had a quarter on micro with imperfect information. Then in the second year we had international macro and macro of imperfect capital markets. All very useful to understanding the crisis.

    I think it's completely false to say that we don't have models to think about what happened in 2008. We don't teach them to first-year grad students because we want to teach the basic tools. RBC models and PIH teach you the tools of modern economics. Once you have that down, you can add frictions to your heart's content. PhD students are smart. They can read up on the history of the discipline in their own time, or talk to their professors if they have questions.

    If you come from a non-econ background, it's your job to read up on that stuff and learn it, and one of the best ways to make sure you do that is to make you teach undergrad macro. You seem to have a handle on this stuff now and you're not out of grad school yet, so I guess it worked.

  10. People should be aware that philosophers of science, like Imre Lakatos and Spiro Latsis, have said that Friedman engaged in pseudoscience -- and that his research program was less scientific than Marxism.

    Friedman was always a bit anxious to respond to these criticisms.

    Here's Latsis:

    "Thus on the basis of Lakatos’s evaluation of Marxian economics by the same yardstick as ‘degenerate’ science because allegedly all its novel predictions were empirically refuted and their refutation not progressively resolved, on the same threefold graded criteria of ‘progressive-science’ and ‘degenerate-science’ and ‘pseudo-science’, Friedman’s economics was in effect evaluated as being even less ‘scientific’ than Marx’s, that is as ‘pseudoscience’ as opposed to ‘degenerate science’, because it had allegedly never even made any theoretically novel predictions that could possibly be refuted, a necessary condition of being ‘scientific’ in Lakatos’s normative methodology."

  11. Re: Niklas Blanchard said: "~98% of economic fluctuations in history were real business cycles (i.e. productivity/technology/wealth effect shocks). For the ~2% remaining, I say monetary disequilibrium."

    What wavelength are the fluctuations you are looking at?

    In the 19th and 20th centuries, at least, ~3/4 of ten-year wavelength fluctuations appear to be monetary disequilibrium. By contrast, ~9999/10000 of hundred-year wavelength fluctuations appear to be technology...

    Brad DeLong

  12. When I was an undergraduate at Michigan in the early 1970s, I took "Intermediate Macro" with Gardner Ackley. Masters and PhD students without the background were sprinkled about the class. Our main textbook was Keynes' General Theory. It was the tail-end of a long tradition, that died very shortly thereafter; Ackley himself abandoned having the students read Keynes a year or two after.

    In Ackley's class, we knew that the Phillips Curve was an artifact. We knew IS/LM was a kludge. I don't think we thought much of, or about, Solow, but (the) Phelps (volume) represented confidence in imminent progress for the research program.

    When I took Econ 101 (which was actually Econ 204 or something like that), I distinctly remember the Teaching Fellow remarking that pretty much all of economics was in Lipsey-Steiner somewhere in some form. That attitude, too, was the tail end of a long tradition that began with Samuelson's Foundations, and the textbook of Euclidean geometry that he derived from it.

    My Junior Seminar was led by a professor, who had been at Michigan 52 years, and was teaching his last class -- the seminar -- in the same room where he had taken his first class. He said that, in his day, real economists read Schumpeter's Business Cycles unabridged, (and may have worked on it or for Mitchell or Kuznets at the NBER, or for the Cowles Commission), , and he was the last professor, who had read it. (He would have understood that technology drives 60+ year Kondratieff waves, not Kuznet inventory cycles -- in other words, what Brad DeLong says in the comment above.)

    My senior advisor was Wolfgang Stolper, a man, whose great work, now forgotten I suppose, was a detailed comparison of the structure (in a Leontief-like sense) of the economies of East and West Germany.

    I think Samuelson's Foundations was experienced as a great relief, for graduate students. It appeared to be a codification of all of economics; it fulfilled the mathematical ambitions of the Marginal Revolution, and relieved economists from plowing through the blizzard of talk, talk, talk from the institutionalists. PhD students at Michigan in the 1950s and 1960s still had to demonstrate a reading fluency in technical German, a requirement that was increasingly waived in practice.
    For Samuelson's generation, his mathematical codification was a triumph, precisely because they understood it in a context of historical and factual knowledge. For, say, Krugman's generation, you didn't have to read anything before Samuelson; it was an unbreached barrier, hiding institutionalist traditions and knowledge. Krugman, himself, didn't read even Ohlin!, until after his own work on trade was complete. It was the beginning of a weird bi-furcation, in which Economists knew more about an increasingly sophisticated, mathematically complex Economics, and less and less about the actual economy.

    The great battle over stagflation was just getting started in my college days. Arthur Burns, Wesley Clair Mitchell's star pupil and the last great institutionalist student of the business cycle, was Fed Chair.

    I remember well the coming of "rational expectations". I couldn't understand the wilful determination to resurrect the classical model. If we knew anything, we knew the classical model (of general equilibrium governing the aggregate economy) was wrong. Money drove fluctuation in an aggregate economy, which was not inherently stable; so, we wanted a model of a self-stabilizing economy without money (or without money that mattered), why?

  13. Lolita2:59 PM

    I'm glad you did all the linking, because I'm not econ and it was looking quite like a foreign language at first. My spouse, who's an econ drop out, took one look and said "obfuscation is the point of econ language."

    I did learn a lot from your post, but I do have a question. Are there any models out there dedicated to predicting how our Wall Street overlords and corporate titans will behave, and that seek to counter or mitigate their effects on the rest of us?

  14. I think the interesting thing is that the DSGErs are sure that they are almost physicists by now. They have the form but not the substance -- the equations (sometimes the very same equations) but the variables have different meanings.

    Also, furthermore it is fun. I've matched a few moments myself.

  15. @Brad Delong: This may sound sort of smart ass-ish, but I'm taking the (much) longer view.

    There is good reason to take the long view, too! A theory of economics (especially a grand over-arching theory) should be able to get you from hunter-gatherers to consumer-traders consistent with the punctuated equilibrium found in the historical record, while making a minimum of ad hoc assumptions, and remaining stable without reaching outside itself.

    That is why I take Soddy and Georgescu-Roegen so seriously. Soddy, to my mind, is a pioneer in work concerning Minsky's FIH!

    As you already know; RBC, of course, has problems if you introduce money demand, Say's problem...thus it's not a workable grand theory.

  16. Interesting post. At the LSE in the late 70's, the alternative Econ intro. course was given by the marvelously eccentric Prof. Michio Morishima- he didn't actually come out and say that Keynesianism didn't exist, it was a sort of Cambridge mental fog thru which actual Economic processes in Germany under Hitler were guessed at and explained away, nor did he explain that Macro doesn't exist- there is Growth theory, there is a sort of Schumpeterian 'want- pattern' effect such that 'life-style' preferences for large classes of people change discontinuously under the pressure of a bust or a boom, maybe there is some sort of cliodynamic /Institutional/National 'Spirit' stuff which is also happening but Macro as such doesn't exist- especially in the silly Prices and Incomes Policy vs. Monetarism debate that dominated the end of the 70's in the U.K. True, w.r.t a credentialist crisis that peaked in the early 70's when way to many people ended up with degrees (was it to dodge the draft?) Macro looked good and attracted funding and financial dereg. meant Wall street needed Rational Exp./ Efficient Markets snake oil peddlers so it was an availability cascade fueled by a credentialist crisis and buoyed up by the convenient institutionalized chicanery of moment matching DSGE as you describe.
    This doesn't mean an education of this sort is wasted on bright people. Indeed, the reverse is the case- so long as a research program is an obvious failure, those acquiring their skill set under its aegis are better rather than worse motivated (if bright and sincere) to discover better approaches.
    But maybe that's best done outside the academy? Coase, a big hero of mine, started off in Commerce intending to be a lawyer.
    Keynes, my bete noire, started off as a Maths guy supposed to carry on from Marshall and Wicksteed and save Liberalism and- what happened? He turned into the worst sort of speculator- one who uses Govt. money or Institutional money while bluffing everyone around him that he was way smarter than they were because he pretended to find some really deep reason to find the simplest thing they said to be immensely puzzling. Coase was middle class. Keynes upper middle. Result, the man was a complete tosser.

  17. Niklas,

    Why do you think we should have a grand theory to address all questions at once? The map is not the territory. We have a road map for driving, a bike map for biking, a walking map for walking, a flight map for flying.

    Human economies are complicated enough that a theory that helps us understand how we went from hunter-gatherers to industrial revolutionaries is unlikely to have much to say about the niceties of counter-cyclical monetary policy.

  18. Anonymous4:38 PM

    Somewhere the ghost of Fourier howls with laughter every time an econometrician claims to be a scientist and to speak with mathematical rigor.

  19. Anonymous8:32 PM

    Regarding getting the B in the macro course -- I had a somewhat similar experience with a different twist.

    It was the mid 70's, and there was a required macro course which was dutifully taken. This was one of the courses where getting a B+ was a requirement for the program I was in. The course was offered by a professor (later a Nobel laureate) I studied hard, and when the grades came out, I got a B-. This was quite surprising to me as I thought I knew the subject matter adequately, if not well. So I had to retake that course, which I dutifully did -- this time with another "to be Nobel laureate" -- this time, right before the final exam, another PhD student came, and told me that he had been asked to give a few words of advise - he said his piece, about four sentences in total, and walked away. I went to the exam, changed the way I answered the questions, and when the grades came out, I had an A+

    So, what was it that was said to me?

    It turns out that both times, some key questions were worded thus -

    What would YOU do if ......
    What do YOU think would happen if .....?

    The key to the difference between a B- and an A+ was to not answer the question as worded, but rather to just lay out what would happen with various models -- I was told very emphatically in that short conversation - Do NOT offer your thoughts when asked that question on the exam.

    I can assure you I did not know any more macro than before, and neither did I have a better understanding of the various macro models.

  20. @Anonymous: HA. That happens to be EXACTLY the reason I got a B (well, B+). I did all the math just fine. The short answer questions, which in this particular class were an unusually large percent of the grade, were a different story. I just wrote my own thoughts and tried to be creative, when it turned out that what was necessary for full credit was pretty much just parroting of the exact language the prof had used in class, even though that language hadn't been the most elucidating.

    No, I'm not bitter, why do you ask? ;)

  21. Can't study people like you study particles. Economics will never be a science. The best analysis you will get will never be from orthodox neoclassical models but from heterodox and Marxist economists like Richard Wolff and Ha-Joon Chang.

  22. Anonymous1:35 PM

    "In a science incapable of reliable testing of theory, the reality of the assumptions is essential to judging the likely robustness of the results" (Martin Wolf)

  23. Anonymous NIklas Blanchard said...

    " @Brad Delong: This may sound sort of smart ass-ish, but I'm taking the (much) longer view.

    There is good reason to take the long view, too! A theory of economics (especially a grand over-arching theory) should be able to get you from hunter-gatherers to consumer-traders consistent with the punctuated equilibrium found in the historical record, while making a minimum of ad hoc assumptions, and remaining stable without reaching outside itself."

    That would be nice; for now let's have a theory which doesn't produce financial crashes which trash the economy.

  24. @Jon: (Paraphrasing) "Why should we have theories that bring us from hunter-gatherers to consumer-traders"?

    I view the importance of a coherent overarching theory as different from answering specific questions, but that any specific questions that are answered by a model should have implications that align coherently with such an overarching theory. I think that complexity economics (my preferred banner) offers a coherent theory...but that can be contested!

    I don't think Keynesianism passes this test (i.e. [old] Keynesianism does not provide an explanation for the trend rate of GDP). Nor do I think that RBC does, either.

    For the record, I view the Great Recession as a supply shock (to housing construction, and then finance [in an accelerator model]) that induced a demand for money that was not accommodated by the Fed. RBC breaks down under money demand.

    @Barry: "That would be nice; for now let's have a theory which doesn't produce financial crashes which trash the economy."

    We will always have financial crises with modern finance. Soddy's model explains why (as does Leitaer's and Minsky's, although I'm the biggest fan of Minsky's framework). The most I see that financial regulation doing is shifting risk around. You can do that better or worse, but in the end, something is inevitably going to crash. And who is the party most able to absorb the risk? The collective "taxpayer", of course...

    ...the '87 stock market crash didn't trash the economy, nor did the failure of LTCM. The tech crash had an adverse effect, and so did the housing collapse. Why? Just look at the trend rate of nominal spending!

  25. Meta-economics, I always thought, should be about the market for Economic theories and credentials.
    Perhaps, if meta-econ was explicitly taught along side Econ- i.e. what is the credentialist demand for this shite and what the availability cascade derived supply of this crapfest- Econ would rise out of its existing heteronomy- or harlotry- and...what? Admit its not chemistry but alchemy, not physics but metaphysics, not good Aristotelian house keeping but chrematistics.

  26. "For the record, I view the Great Recession as a supply shock (to housing construction"

    Mein gott.

  27. Have you not taken an econometrics class yet? Basically those classes are all about rejecting, or failing to reject, hypothesizes. Those theory classes just tend to explain models that describe economic systems. The econometrics of those models are meant to be implicit. That is what is meant by "tweaking" the parameters. People do the same thing in physics.

    For example, "I think that the acceleration of an object to the center of the earth is proportional to time squared. This theory is based on things I have observed and been told by others. I will now collect data based on distance and acceleration of a falling object and do a line fitting to . I find it to be 9.8 m/s^2. I can now use statistics to see if my guess is statistically significant."

    To be honest, the methodology of economics (or numerical data based social science), is virtually the same as the physical sciences. The main difference is that in physics, you can know everything, because you are looking at simpler systems. You are in the weeds. In economics, you can't really know everything, so there is a lot of proxies used and estimations. Your building block is a person instead of a particle.

  28. @Niklas Blanchard

    The LTCM fallout didn't trash which economy? I think you'll find some folks in current current account surplus countries in wild disagreement with you. More to the point, referencing a tautology as evidence for why these bumper-sticker events didn't affect the US economy is pretty dismal even for economics. It's enough to make one long for moment matching.

    But hold the phone- you claim to be a big fan of Minsky or at least his FIH while also believing that the financial crisis was the result of an exogenous shock (precipitated presumably by some yet to be specified cosmic phenomenon, details forthcoming in the upcoming edition of astrological economics), whose monetary fallout was insufficiently blunted by Federal Reserve easing??? Is this for real?

    Just out of curiousity, what was the right amount of balance sheet ballooning then? Would $4tn of toxic waste subsidies have done the trick or should the Fed have hit out for a cool $8tn? And if this is all the Fed's fault in the first place, why is it "inevitable" that markets will crash? Surely we just need someone in there who understands 'complexity economics'.

    In any case, that you furthermore claim to believe that financial regulation just "shifts risks around" confirms that there are really only three possibilities here: a) that you've actually not read Minsky, b) that you have read him, but his words broke the sound barrier whilst flying thirty thousand feet over your head, or c) that you have read him, but have chosen to willfully misrepresent his work as being the opposite of what it in fact is (perhaps, and I'm just spitballing here, due to the utility to you of tying his prescience and related prestige to your discredited ideas about economic dynamics...?). Suffice it to say, while I prefer the answer be a or b, I wouldn't under any circumsntances be caught long either premise. Which of course has inferences for c. Ehem.

    IMO, economics is a victim of the fact that its enlightened implications are hostile to entrenched interests (climate science is also this, but that is a recent phenomenon, and it being a hard science, or collection thereof, is far less easy to corrupt). This explains why those that can integrate pdes over more friendly if utterly inane, (errr... what are we're going with on this thread? 'empirically unsupported'?), visions can see the top of their professions while those willing to sort through economic history for ideas that have proven to have some actual value are relegated to the sidelines (apropos of Jamie Galbraith's post-crisis piece). As another helpful addition to this case in point, Mr. Blanchard also apparently believes that taxpayer bailouts are the obvious way forward, ostensibly (and of course, given their worldview, inexplicably), because they 'can bear the risk'.

  29. Also Dr. Blanchard, your comment about Keynesianism's* failure to satisfactorily reproduce trend growth made me think of something I recently read by someone whose views- unlike those of the bubble ensconsed exponents of modern economics- very much have been tested empirically over decades. Quoth the Grantham:

    "I briefly referred to our lack of numeracy as a species, and I would like to look at one aspect of this in greater detail: our inability to understand and internalize the effects of compound growth.... Four years ago I was talking to a group of super quants, mostly PhDs in mathematics, about finance and the environment. I used the growth rate of the global economy back then – 4.5% for two years, back to back – and I argued that it was the growth rate to which we now aspired.

    To point to the ludicrous unsustainability of this compound growth I suggested that we imagine the Ancient Egyptians (an example I had offered in my July 2008 Letter) whose gods, pharaohs, language, and general culture lasted for well over 3,000 years. Starting with only a cubic meter of physical possessions (to make calculations easy), I asked how much physical wealth they would have had 3,000 years later at 4.5% compounded growth.

    Now, these were trained mathematicians, so I teased them: “Come on, make a guess. Internalize the general idea. You know it’s a very big number.” And the answers came back: “Miles deep around the planet,” “No, it’s much bigger than that, from here to the moon.” Big quantities to be sure, but no one came close. In fact, not one of these potential experts came within one billionth of 1% of the actual number, which is approximately 10^57, a number so vast that it could not be squeezed into a billion of our Solar Systems."

    Needless to say, the argument is not much deminished by the particulars of the trend growth rate chosen...

    It always amazed me that economists built a house on the sand of two-hundred years of the experience of an emerging (now rapidly declining) hyper-power in a newly industrializing/computing age. Great data it seems also correlates with other things. See, e.g., the date at which historical German bond market data generally becomes available.

    * As a by the bye, I agree with Minsky that Keynesianism is a bastardization of the teachings of Keynes, though certainly a less grevious assault than that accomplished by Mr. Blanchard on Minsky here.

  30. Anonymous9:54 PM

    EJMR responds:

  31. @Majorajam:

    I was unaware that I had said anything so controversial that it would provoke your ire. The LTCM fallout, and the subsequent Asian debt crisis did not trash the US economy. Are you claiming that the economy would fall into a severe recession/depression under a constant growth rate in nominal spending? I can't really tell.

    I'm unsure what you term as "the" financial crisis, but I generally reference Sept/Oct 2008. The real shock was the fallout in sub-prime in very few markets, which trashed some vulnerable players...but overall, the economy didn't tank until monetary policy became too tight relative to the needs of the economy in late 2008. That caused problems with all kinds of nominal debt contracts, accelerating the financial crisis. That story is not opposed to the FIH.

    Are you trying to make the (wildly incredible) claim that financial regulation has the ability to eliminate risk with zero effect on the level of investment in the short run? If so, then say it...but just warning you, I won't be able to keep a straight face. I don't think taxpayers should hold the bag, I'm saying that in reality "they" often do, because "they" can.

    I'm kind of confused about what you're actually disagreeing with, where you didn't simply uncharitably interpret what I was saying because you seem to believe that I have "bad motives"...or I'm the "bad guy"? I give you credit for being more cordial than calling me a "freshwater dead ender"...but still.

  32. Well at least you did some formal macro. I did once as an undergraduate...

    My problem is that very few people read.

    Also I once emailed Brad De Long about being a PhD student. No reply.


    PS. There are some darn useful models out there you were not taught in grad school...

  33. Jasiek8:04 AM

    High time we try again and read Chapter 21-III of General Theory.

  34. As to financial regulation, the entire thrust of Minsky's FIH belies your assertion that 'financial regulation just shifts risks around'. FIH is a world in which risk is NOT some set static value, but is cyclical with the investment cycle, and with an amplitude in direct proportion to the length and intensity of the booms and unfettered transformations of the financial/monetary/real landscape they create. 

    More specifically, it is Minsky that made the keen insight that for profit financial intermediation is the enemy of a stable economy; that, left unchecked, financial innovation empowers banks and non-bank entities alike to create money at will entirely outside of the control of the central banking authority. He furthermore noted that private intermediaries have every profit incentive to do just that- to innovate and grow the money supply (see China's peculiar newfound proclivity for stockpiling base metals, the erstwhile, continuing and eye-wateringly massive shadow banking system, and just about as many parallels as the length and breadth of the history of financial capitalism). And he noted how each stage of speculative finance begets yet more dangerous and unstable dynamics. Hence deleterious ends. Hence ritual financial crises. Hence the FIH.

    But really, don't take my word for it. Here's Minsky:

  35. "In both Keynes and Schumpeter the in-place financial structure is a central determinant of the behaviour of a capitalist economy.  But among the players in financial markets are entrepreneurial profit-seekers who innovate.  As a result these markets evolve in response to profit opportunities which emerge as the productive apparatus changes.  The evolutionary properties of market economies are evident in the changing structure of financial institutions as well as in the productive structure.  In the Theory of Economic Development Schumpeter called the banker/financier the ephor of market economies.  The ephor was a magistrate of Sparta who contained and controlled the kings.  In Schumpeter's vision it is the banking structure of a capitalist economy which controls and delineates what can be financed, and only that which is financed enters the realm of the possible. But nowhere is evolution, change and Schumpeterian entrepreneurship more evident than in banking and finance and nowhere is the drive for profits more clearly the factor making for change.   But in an evolutionary system the power and efficacy of the ephor is also endogenously determined.  To understand the short-term dynamics of business cycles and the longer-term evolution of economies it is necessary to understand the financing relations that rule, and how the profit-seeking activities of businessmen, bankers and portfolio managers lead to the evolution of financial structures..."

    "...The welfare state big government managerial capitalism largely but not completely divorced business profits (cash-flows) from private investment...It followed that the margin of safety which entered into the building of liability structures which reflected earlier experience were too big: the safe level of indebtedness was higher in the postwar economy than hitherto...the independence of operating corporations from the money and financial markets that characterized managerial capitalism was thus a transitory stage.  The emergence of return and capital-gains-oriented block of managed money resulted in financial markets once again being a major influence in determining the performance of the economy.  However, unlike the earlier epoch of finance capitalism, the emphasis was not upon the capital development of the economy but rather upon the quick turn of the speculator, upon trading profits...As managed money grew in relative importance, more and more of the market for financial instruments was characterized by position-taking by financial intermediaries. These positions were bank-financed.  The main financial houses became highly-leveraged dealers in securities, beholden to banks for continued refinancing.  A peculiar regime emerged in which the main business in the financial markets became far removed from the financing of the capital development of the country.  Furthermore, the main purpose of those who controlled corporations was no longer making profits from production and trade but rather to assure that the liabilities of the corporations were fully priced in the financial market...The question of whether a financial structure that commits a large part of cash flows to debt validation leads to a debacle such as took place between 1929 and 1933 is now an open question..."

  36. "...In the present stage of development the financiers are not acting as the ephors of the economy, editing the financing that takes place so that the capital development of the economy is promoted.  Today's managers of money are but little concerned with the development of the capital asset of an economy.  Today's narrowly-focused financiers do not conform to Schumpeter's vision of bankers as the ephors of capitalism who assure that finance serves progress.  Today's financial structure is more akin to Keynes' characterization of the financial arrangements of advanced capitalism as a casino.   The Schumpeter-Keynes vision of the economy as evolving under the stimulus of perceived profit possibilities remains valid.  However, we must recognize that evolution is not necessarily a progressive process: the financing evolution of the past decade may well have been retrograde." (Hyman P. Minsky, Schumpeter and Finance from Market and Institutions in Economic Development: Essays in Honour of Paulo Sylos Labini, 1993)

    That was 1993. Just imagine what he'd have said about 2007 era funny money market finance!!!

    Now then, does this- from the mind that developed the FIH from a far broader framework that realized that institutions matter, that uncertainty matters, and that above all, a clear eyed examination of incentives matter- would have promoted cowboy capitalism complete with a promise of bailouts, i.e. which as Mr. Grantham has pointed out, is a working definition of moral hazard? Is it really necessary to ask the question? 

    Speaking of which, by your lights the taxpayer shouldn't bailout reckless leveraged speculators earning massive sums until the bust, and we shouldn't regulate financial markets, AND the monetary/fiscal authority should save the economy from the systemic crisis provoked by such casino capitalism, the extremes to which we have reached neither Keynes nor 20's era bankers could never have imagined. Can you spot the inconsistency or do you want me to? 

    Seriously sir, you and your ilk are furnishing the logic that provides air cover for decades of obscenely misbegotten policy that has benefited the few at massive expense of the greater good, and you want me to explain to you why I'm impolite? Really? Where I come from, everyone bears responsibility for the effect their actions have on the greater good, and villains great and small are owed no courtesy.

  37. This was posted first, but for some reason has yet to appear:

    @Niklas Blanchard

    What do we disagree about? I have yet to find a single exception. You said: "The LTCM fallout, and the subsequent Asian debt crisis did not trash the US economy." Please reread what I wrote. The US isn't currently a current account surplus nation, is it?

    You said: "Are you claiming that the economy would fall into a severe recession/depression under a constant growth rate in nominal spending?" Not quite, but I am saying that citing nominal spending growth rates as causes of turns in the business cycle is asinine. Please note that I try not to assign italics randomly.

    As to "the crisis", (and there was but one that goes under that moniker alone), sub-prime was the very tip of the Ponzi finance spear. As anticipated by FIH, the tip of the spear is always the lead element in an endogenously eventuated financial crisis, such as this was. The recovery in asset prices and the real economy, such as it's been, was a consequence of the intervention of what Minsky called Big Government and Big Bank. As a consequence, both are tottering the world over.

    I have no idea in what sense you are using the term "shock". If I ride a skateboard down a half pipe and it picks up a head of steam and then peters out in the air above the subsequent rise, that loss and eventual (free fall) reversal of momentum is not in any way a shock as I am familiar with the term. Yet these are the precise dynamics described by FIH. Perhaps you can enlighten me.

    As to the subsequent tanking of the economy, please tell me what normally happens when an entire banking system is revealed to be insolvent many times over in a prototypical "Minsky Moment" such what as 2008 was (and more rightly the culmination of a series of Minsky Moments beginning with subprime in 2007) and what monetary policy can do to prevent that. Perhaps you're suggesting the monetary authority should subsume the entire banking system in such a circumstance? Of course, this is what it largely did, except to a degree less than required to overcome the complete loss of confidence in private claims and the asset prices they financed to the hilt.

    Really, when the, whatever it was, 60 some odd trillion dollar CDS market failed after Lehman, what was the outcome of that to be? When the $3tn money markets failed after Lehman, what was the outcome of that to be? When the $1 some odd tn repo market failed after Lehman, what was the outcome of that to be? When real money wouldn't touch with a barge pole leveraged loans, CDOs or any ABS for that matter, what was the outcome of that to be. How was the central bank to overcome the tearing asunder of huge swaths of the capital markets? These were MASSIVE sources of liquidity for households and firms, not to mention the barnacle leveraged speculators that were likewise grease for the wheel.

    The broader point of course is that it was euphoria itself that gave so many of these claims their value. That money is, after all, what we believe it to be. And that when that belief turns to disbelief the bottom falls out, and a (massive) pullback of intermediaries replete with debt deflation dynamics is inevitable. That is FIH. What you are peddling, by contrast, in no way resembles Minsky in the god forsaken least.

  38. Anonymous11:13 AM

    A whole lot of words spilled to no effect. None of you have, or will ever have, any effect on the profession (except you Brad, and that effect has been uniformly negative). Return to thine holes, whiny brats, and leave economics to the adults.

  39. More bold words from Anonymous.

    He might be right that the profession is invulnerable both to challenge and to events in the world. It's an institutional question. In some areas scholasticism survived until the 19th century, fighting to the bitter end against Humanism, the Reformation, the Enlightenment, democracy, and the the scientific world view.

  40. Re @Jon: The argument that students should be taught the math and then left to their own devices to learn about economic history is absolutely bollocks that most econ students are happy to live with because they're more comfortable dealing with Mathematics than the more abstract history of thought.

    I've heard the same argument being used by students in my undergrad honours program as an excuse to not take a course on the history of economic thought, and frankly, I think its absolutely ludicrous for economics students at ANY level, UG, PG or Ph.D, to go through a program without conducting an in-depth study of economic thought in their field. I'm sorry, but if you plan on completing about 10 years of education in Economics loading up on the math and completely ignoring everything from Smith to Keynes, then you're exactly the kind of economist we don't need.

    I see more and more economists who are Mathematicians who know some economics, whereas I think we need more economists who have the required knowledge of math instead.

    I realize that Math has become an integral part of Economics over the last 3 decades, but even then, I'd rather that it be a mere tool used in the study of Economics than be the main show. This might be disconcerting to some of you, who try to figure the world out in mathematical terms because that's where you feel safest/most comfortable. But as one of my professors said, "Math in Economics should be like a washing machine. I need it, but it doesn't run my life. I put stuff in, and stuff comes out. Simple as that"

  41. @Majorajam:

    Since wealth is not physical cubes of some some substance, but rather is the satisfaction of the wants of individuals, your debunking of economic growth is nonsensical. A block of steel and the automobile formed from it (assuming the car has a buyer) have the exact same mass, but quite a bit of wealth has been created from the process.

    A "growth rate of 4.5%" is simply a measure of a monetary aggregate, not a volume or a mass, and it has presuppositions behind it that make it non-applicable to non-monetary economies and slave societies.

  42. In reply to
    "Senior professors at economics departments around the country are the ones who give the nod to job candidates steeped in neoclassical models and DSGE math. The editors of Econometrica, the American Economic Review, the Quarterly Journal of Economics, and the other top journals are the ones who publish paper after paper on these subjects, who accept "moment matching" as a standard of empirical verification, who approve of pages upon pages of math that tells "stories" instead of making quantitative predictions, etc. And the Nobel Prize committee is responsible for giving a (pseudo-)Nobel Prize to Ed Prescott for the RBC model, another to Robert Lucas for the Rational Expectations Hypothesis, and another to Friedrich Hayek for being a cranky econ blogger before it was popular.

    If you want to change economics education, it is to these people that you must appeal."

    This from Max Planck:
    “An important scientific innovation rarely makes its way by gradually winning over and converting its opponents: it rarely happens that Saul becomes Paul. What does happen is that its opponents gradually die out, and that the growing generation is familiarized with the ideas from the beginning.”

  43. Anonymous12:29 PM

    I chuckle at the idea that we'll all just die out and be replaced by nitwits like Niklas Blanchard or this idiot grad student. Not going to happen.

  44. another economist6:02 AM

    Anonymous, economics is a science. We have no need for your kind. Perhaps Fox News is hiring.

  45. Anonymous1:07 PM

    Have you come across the Von Neumann quote (vis a vis momeent matching):

    "With four parameters I can fit an elephant, and with five I can make him wiggle his trunk."

  46. Anonymous said...

    " Have you come across the Von Neumann quote (vis a vis momeent matching):

    "With four parameters I can fit an elephant, and with five I can make him wiggle his trunk.""

    That's the whole problem with simply fitting a model to the data.
    And if I were the 'anonymous' who said "Moment matching and calibration are falsifiable. Ergo, testable.", then I wouldn't put down my name, either.

  47. Great post.

    Teaching macro must have been a bitch without having taken the course yourself.

    On empirical verification: It might be valid to avoid a long detour into how to actualise various theories. For example how do you figure out if a price rose because of supply down or because of demand up? Both supply and demand are private information inside either the heads of consumers or somewhere inside a firm's knowledge base. Yet the resolution of the diamonds/water paradox was a significant step forward.

    (Still, they could have at least referred you to some papers that justify that this stuff is true. I remember one cool paper I skimmed found Gibson goods in a rat lab -- thus, if reproducible, "proving" the Lagrangian maximisation model over "dumb psychologist" models like attraction theory or amelioration or things with names I don't remember. On the other hand, Townsend & Busemeyer developed an anti-economist behavioural model that has dumb-learning humans walk around in realspace, thus invoking mathematical sophistication.)

    On the calculation end: hadn't you heard before starting grad school that economists can't successfully predict anything? You're preparing to enter an intellectual field that needs improvement.

    On frictions: I think 'friction' is probably the wrong word because it assumes the model is right, "minus a constant drag". In my view the model is probably just wrong-headed, puro.

    On finance: Didn't you know economics isn't about money?! :) There's a funny video game that depicts an economist dodging dollar bills. You can also hear Larry Summers tell a funny story about trying to trade one of his macroeconomic models -- I believe it's on Yale Financial Econ podcast/video series by Bob Shiller. (LS would be a guest speaker on one of them, you can get the story there.)

    on CAPM: did you learn that it's wrong?

    on DSGE: You know what pissed me off? The use of real numbers. It's well-known that integer optimisation problems have different solutions than real problems, and yet our agents are able to buy a transcendental number of eggs.

    This must be an interesting time to be in graduate school -- or maybe it's plus/minus a few years that would position you best to change the field, I'm not sure.

    I'll leave you with my possibly mis-quotation of David Graeber: "The definition of power is not having to justify your assumptions."

    (eg, Ricardian equivalence, rational actors)

  48. Anonymous9:23 AM

    There is an implicit assumption here, that grad teaching in physics or molecular biology or chemistry is better.
    Based on my experience, I would say no.
    The errors are different; perhaps the first law of grad school - pedagogical errors are conserved.
    Ezra S Abrams