Thursday, September 8, 2011

Zombie Economics, Redux

For all of you who are dying to know more about John Quiggin's Zombie Economics, here is a review essay I have written. This is somewhat formal and un-bloggy, complete with references, and is not the Journal of Economic Literature review on Quiggin's book that I wrote, which is quite short. I do more in this essay than just critique Zombie Economics. It is in part a defense of contemporary economics, in response to various rabble-rousing, including Paul Krugman's 2009 NYT piece, or this more recent critique. There are also some financial crisis ideas in there.

If you find errors, please let me know, and I'm interested in your comments.


  1. Should it be "five zombie ideas" rather than three?

  2. A good reference on the Great Moderation would be Stock, James; Mark Watson (2002). "Has the business cycle changed and why?" They looked at a number of hypotheses, gave better policy credit for 10-15% of the improvement, but concluded that most of the decline in real volatility was due to "good luck".

  3. I liked it quite a bit. I have a two separate points about the book myself that follow up on your critique.

    For one, I find his policy suggestion that central bank intervene to stop a bubble puzzling. I wonder what leads him to believe central banks even know if something is a bubble versus a legitimate boom. Does he have some foolproof way to tell? If so, that would be a nice thing for him to publish. This issue is especially important since central banks who do not engage in commercial or consumer lending probably have severe informational deficiencies when undertaking an intervention in these markets.

    On progressive taxation, if you are really concerned about inequality, the issue is to tackle the underlying privileges that lead to such inequities. Rich people are sometimes rich because they are smart and talented, many of them are rich because of access to certain privileges (social and governmental). For example, rich white kids get good schools, poor black kids get bad schools. I think it's a really round-a-bout solution to fix this problem by transferring money. Instead, we should provide poor black kids good schools (whether they are financed by progressive taxation or not). There are a whole host of similar privileges that lead to inequities aside form schooling as well that we could tackle. After we have tackled these underlying privileges that lead to such enormous inequities, then we should look at progressive redistribution. One of the blind spots in the progressive movement is an unwillingness to tackle structural causes of inequality.

  4. Ted,

    Bubbles: Yes, this came up in discussions of "systemic risk regulation." Some people had the idea that a systemic risk regulator (the Fed or some other entity) should act to prevent asset bubbles from arising. But the word "bubble" gets thrown around a lot in ways that make it clear that different people have very different notions about what a bubble is. How then could you assign a regulator the task of preventing such a thing from arising?

    income distribution: Here is an interesting example. Someone like this:

    can get rich through something the government grants him/her, e.g. monopoly power. In a world where people get rich by knowing the right people rather than inventing things people find useful (e.g. Bill Gates), we don't get much innovation. There are better ways to solve that problem than with the tax system.

  5. page 7, 3rd paragraph, last sentence

    "it easy" ==> "it's easy"

  6. page 8, top

    "the deficit may be have arisen" ==> ~"be"

    let me know if this is annoying.

  7. You mention "the economics of crime" in that paper. I've read Bill Black bang on about economists neglecting that angle, but I've always been fuzzy on the particulars. I recall him blogging at New Economic Ideas From Kansas City which I believe got turned into New Deal 2.0. I first heard about him when he participated in this edition of Cato Unbound.

  8. typos:

    No, that's not annoying at all. Those need to be corrected.


    One person I know who works on the economics of crime is Peter Rupert (UC Santa Barbara). You can look up some of his crime papers on his CV:

  9. This is a confused review that may need substantial redrafting.

    For its stated objective, there are too many vague references and empty sentences such as:

    >It isn’t that the “market” guides us about >the “value” of the asset. The
    >price of the asset is determined by how the >market participants value it.

    Is the point being made here just an objection to the word "guide"?

    Is it rather naive to restate this as "how the market values it" and just insert the word "participant.

    Or is the dispute about using words like "price" and "value"?

    Anyway, overall this review is essentially a predictable song from one type of economics, that I suppose cannot see the wood for the trees.

    This comment is no defence of Quiggin, but an insistence on higher standards for discourse, and a welcoming of his project.

    [Although bloggists are free to print whatever they like on their own blog]

  10. No, John would identify himself. I like the part about "print whatever they like..." Yes, the typesetter and I worked in the basement all night on it.

    The "type of economics" we practice is serious economics, so get serious.

  11. On page 4: This is a very minor grammatical point, but MCA Act is redundant. You can probably be safe with referring to it as the MCA.

    The full name of MCA is "Depository Institutions Deregulation and Monetary Control Act".

    I am not exactly sure whether the MCA created the moral hazard problem (moral hazard is only a problem without prudential supervision) as much as the Federal Home Loan Bank Board allowing insolvent S&L's to appear solvent by allowing them to not follow GAAP and issuing income capital certificates.

  12. Nice.

    One minor comment: you try to argue that, when Bernanke said the fed could claim credit for the great moderation, he was indirectly claiming credit for old Keynesian economics, since this is largely what the fed does. In this particular instance this was very explicitly not the case. The literature would say that the fed deserved credit to the extent that in the 80's it adopted a policy rule that respected the "Taylor" principle; that nominal interest rate should react more than one for one with inflation. The argument for thr taylor prnciplele is very post 70's in that it is essentially about the conditions neededforntherento be' a unique rational expectations equilibrium. In other words, the Great Moderation would have come about because the fed had understood the lessons of the 70's and had become a new rather than old Keynesian.

    The reference, I believe, is thenpaper by clarida, gali and gertler


  13. This certainly looks like a more serious and substantive response than the blog post a while back - I haven't yet seen the JEL piece. I'm flat out at present, but will try for a brief response soon, and maybe something more extensive later.

  14. RV,

    There is actually nothing "New Keynesian" about the Taylor rule. Taylor made it up in the early 1990s and argued that it worked well (i.e. according to standard "old" quadratic loss functions) in some Old Keynesian macroeconometric models. The Taylor rule as an optimal policy rule is not a robust implication of New Keynesian models, though you can do some sleight-of-hand to get it. Thus, the Taylor rule is actually Old Keynesian. The fact that it entered New Keynesian economics was just another facet of how the approach was sold to the Old Keynesians.

  15. John,

    Yes, to be fair, there's much more in the book than can be addressed in a little blog post, and there are serious issues involved, so it's serious. I had to look up "I'm flat out" in urban dictionary, which gave the equivalent as "I'm buggered, mate," i.e. too much to do (not flat out on one's back, or some such).

  16. SW,

    What's this famous story about the 76 Lucas Critique being left behind at the Minnesota Fed? I've never heard about this bu am curious.

    - C

  17. I didn't mean the Taylor rule as formulated by Taylor. Just the principle that nominal interest rate should react strongly to inflation (or expected inflation) if you want to have a unique solution to the model. You get something like this from most standard linearized New Keynesian models (see for example Woodford, chapter 1 or 2 I think) and is at the core of much of New Keynesian critique of 70's monetary policy.


  18. C,

    This was an anecdote of Sargent's that I heard in 1995. Lucas was there, and did not deny it. Apparently Lucas was visiting the Minneapolis Fed, left for the airport, and left some notes behind, that someone had to send on. Sargent claimed it was a draft of the Lucas critique paper, and he made some joke about it that I can't remember. At the time, of course, losing your notes would be serious business.

  19. RV,

    That's a tricky business. In Woodford's book, he wants to come up with a simple way to formulate monetary policy, and wants something that looks like what the Fed actually does. He knows that there is a literature on indeterminacy under nominal-interest-rate-pegging, and wants to get around that. In the framework he is working with, it actually makes just as much sense to think about the government following a price-level rule, whereby the relative price distortions are corrected with a rule that makes the price level contingent on the complete history of exogenous shocks. Woodford wants to invert that using the first-order condition that determines the nominal interest rate to give an interest rate rule that the central bank follows. But he wants this to look more like a Taylor rule, which is actually expressed in terms of endogenous variables, and he goes through some contortions to give you something like that, and then proves determinacy. The key point is that there is nothing about the Taylor rule that follows naturally out of Woodford's model, and it certainly is not a general principle that tells us what an optimal monetary policy should be.

    A useful point is Benhabib/Schmidt-Grohe/Uribe's one - that Taylor rules in general do not give you uniqueness.

  20. I looked at some of Peter Rupert's paper, but it was on things like unemployement insurance or wage subsidies and their effect on crime among the unemployed and employed. Your paper was about financial crime causing a crash. That sounds more like Black's focus. He often references the S&Ls since he was a regulator at the time.

  21. Yes, none of the economics-of-crime work I know about actually addresses itself to financial crime or, I think, white-collar crime more broadly defined.

  22. In a 2005 interview published in MACROECONOMIC DYNAMICS (v. 9, 561-582), Sargent tells the anecdote about Lucas leaving the notes for his `critique' paper at a 1973 U. Minnesota conference. The day after the conference, Sargent retrieved the folder with the notes for the paper; he refers to this rescue of the manuscript as his contribution to the paper.

  23. Black specializes in white-collar and particularly financial crime. I couldn't find a publications page at his university, but his ssrn page is here.

  24. Stephen,

    The closest paper I know to the economics of financial "crime" is Akerlof and Romer's 1993 "Looting" paper. I don't agree with everything in the paper, but it's interesting none-the-less:

  25. I hadn't seen that. I just looked at a bit of the introduction. Some of this is standard moral hazard, but they get at some other issues in there too. I didn't look at the model.

  26. Stephen,

    First of all, thanks for the overview of developments in the field from your vantage point.

    Second, I'm intrigued by the following,

    "Hsieh and Klenow (2010) provides a nice summary of how received economic research views the determinants of income differences across countries. According to them, about 10-30% of the differences are accounted for by human capital, about 20% by physical capital, and the bulk - about 50-70% - by total factor productivity (TFP)."

    Am I correct in concluding from this finding, particularly the importance of TFP in explaining income differences, that a person's income depends more on where he or she was born than on his or her own efforts?

  27. When Hsieh and Klenow do that accounting, that is in the aggregate, i.e. look at a country, measure total human capital, total physical capital, total TFP, and aggregate real income, then make the comparison to another country. For an individual, if we think of his or her human capital as determining his or her income, relative to the other people living in his or her country, then it seems that is not the major determinant of his or her income relative to the income of someone living in another country. But what determines human capital? (i) raw ability; (ii) individual effort; (iii) educational opportunities. In more highly developed countries, the educational opportunities will be better than in poor countries, which makes individual effort matter even less relative to "where you were born." So your last sentence seems to be correct, and for more reasons than what it is in the previous paragraph.

  28. My comment above that confused you was regarding the misspelling of Malawi. Other than that small note, interesting read all around.

  29. 'flat-out' is so common an expression for really busy in Australian english that neither John nor I would remember, or until now, I did not even know that it may not be in American English

  30. Those who long for pre-1970 macroeconomics should bne careful for what they wish for.

    When Friedman set out his model in a 1970 JPE conference, it turned out to be a generalised IS-lM model.

    As for his general framework, remember Brad de Long in The Triumph of Monetarism? Journal of Economic Perspectives, Vol. 14, No. 1 (Winter, 2000), pp. 83-94 at 84:

    “All five of the planks of the New Keynesian research program listed above had much of their development inside the 20th century monetarist tradition, and all are associated with the name of Milton Friedman.

    It is hard to find prominent Keynesian analysts in the 1950s, 1960s, or early 1970s who gave these five planks as much prominence in their work as Milton Friedman did in his.”

  31. DeLong's 2000 article views the macro world in terms of two camps, which is not quite correct. The New Keynesians think of what they do as a synthesis of everything that came before it - Old Monetarism, Old Keynesianism, neoclassical growth theory (Prescott et al.). They would like to view it as a consensus, but everyone is not on board, I think. Further, I'm not sure the New Keynesians have Friedman's skepticism about stabilization policy. Most of them seem quite eager to fine tune.

  32. Actually "flat out" means "going as fast as possible (already)". I didn't realise it was idiomatically Australian.

  33. I see Jim Rose already made this point

  34. John,

    An American might say "she's flat-out wrong," or talk about driving his Farrari "flat out." Urban dictionary says the extended Australian slang is "flat out like a lizard drinking," which I like a lot.

  35. This is the stuff from which Hayek starts, and the significance of the collapse of "shadow money" in the post-boom bust is the topic of some important papers by Credit-Suisse economists, who quote Hayek on the topic.

    Stephen writes,

    "Now, where would a real estate bubble come from? MBS were used exten-sively in financial market asset trading prior to the financial crisis, and small quantities of MBS could potentially support a huge amount of financial ex-change, as one MBS can potentially be used many times as collateral, through the process of rehypothecation (see Gorton 2010). Since MBS had value in ex-change, there was an MBS bubble, and this fed back into the market in mortgage originations. Because the MBS sold at a high price reflecting their high liquidity premium, competing mortgage originators were willing to grant good terms on mortgage loans, many borrowers could finance real estate purchases who would not otherwise be able to do so, and prices of real estate shot up. Of course, the underlying mortgages, particularly in the subprime market, did not generate the promised payoffs, there were a series of fatal incentive problems in the chain of financial transactions that created some MBS, and the MBS ultimately were no longer perceived as safe, liquid assets. A huge quantity of financial exchange went away, and the bubble “burst,” so to speak. Now, nowhere in my story did I invoke irrationality, nor did I violate any basic principle of arbitrage pricing, though certainly there are frictions that play a key role in the story. Asset bubbles, as monetary economists understand them (see Williamson and Wright 2010, 2011) arise because of information frictions -limitations on recordkeeping and information flows. As well, the basic financial market incentive problems associated with the financial crisis were moral haz-ard and adverse selection, which are the key private information frictions that economists know a lot about."

  36. Quiggin says "By contrast, I normally use ‘rational’ to refer to the kind of behavior found in the simplest form of the DSGE models: farsighted, and purely egoistic, agents maximizing the expected utility of stochastic consumption streams over time."

    Geoff Brennan did battle with Quiggin on this narrow definition of rationality as egoism in ‘Rational Actor Theory in Politics: A Critical Review of John Quiggin’ by GEOFFREY BRENNAN and JONATHAN PINCUS in Economic Record, Volume 63, Issue 1, pages 22–32, March 1987!

    They too note that rationality says nothing about the content of preferences.

    The distinction between rationality and egoism is about the difference between the structure of preferences and the content of preferences. The abstract of their paper is:

    “John Quiggin's paper attacks public-choice theory. among other things, for its use of the assumption of ‘rational egoism’. The object of our response is twofold.

    First, to distinguish egoism from rationality, and to indicate that rationality postulates, when faithfully applied, provide reasons for believing that political behaviour and market behaviour will be systematically different, and specifically that the former will be less egoistic than the latter.

    Second, to indicate that comparative static propositions in public-choice theory (and in economics more generally) can be sustained on rather weaker behavioural assumptions than homo economicus embodies, and that consequently some of the public-choice orthodoxy would survive any attack on the egoism assumption.”

    The egoism assumption is an irritant, the assumption of far sightenedness must be backed up by references. Perfect foresight is miles away from rational expectations.

    Lucas worked out a theory of Phillips-type correlations that result not from slow wage or price adjustments, but from individuals’ mistaken belief about current macroeconomic conditions, misperceptions that arise because individuals have incomplete information concerning the state of the economy.

    This information deficit is the exact opposite of far sightedness. less than perfect information is the key to Lucas’s model.

  37. the last post relates to Quiggin's reply at crooked timber

  38. Stephen--

    I've given "life" to Quiggin's Zombies using