Tuesday, February 11, 2014

Macroeconomics in the Blogosphere

A couple of months ago, I wrote a critical post about how the economics blogosphere dealt with a murky event - the recent turmoil at the Minneapolis Fed. Basically, bloggers got some basic facts wrong, made things up, and were willing to make strong inferences based on little evidence. In terms of basic reporting standards, it was a pretty poor performance.

Here's another example of poor blogging performance. In this case, the underlying facts are not murky. Kartik Athreya has published a book, Big Ideas in Macroeconomics, and the facts are plain. All you have to do is read the book to know what they are. Basically, Kartik has written a 400+ page book with the intention of explaining modern macroeconomics to the layperson - with no symbols or mathematics and only the occasional diagram.

The first mention of Kartik's book (other than my plug on January 10) seems to have been David Glasner's post on February 3. Though he gives Kartik some compliments, Glasner doesn't like the book. Kartik is in good company here, because apparently Glasner isn't too fond of modern macro in general. Though Glasner is confused, we can give him credit for politeness in this instance and, above all, he appears to have actually read the book.

After that, it gets more interesting. On February 10, a week after the Glasner post, there was a quick succession of blog posts, beginning with John Quiggin's. If you're familiar with Quiggin's book, Zombie Economics, you'll know that he hates modern macro with a passion. But, suffice to say, as I argued here, that Quiggin is poorly informed. We might say that Quiggin has high TFP - plenty of intellectual horsepower - but he's applied little labor input to understanding what modern macroeconomists do. Nevertheless he doesn't like us. Whatever.

In this instance, Quiggin's post appears to have more to do with reprising Zombie Economics than discussing Kartik's book. In his post, Quiggin claims to have the book in his hands, but I'm suspicious that he hasn't read it or, at best, he's spent little time delving into it. Why do I think that? Here's a quote from Quiggin's post:
The easiest way to see why the book is so striking is to list some topics that do not appear in the index (and are not discussed, or only mentioned in passing, in the text). These include: unemployment, inflation, recession, depression, business cycle, Phillips curve, NAIRU, Taylor Rule, money, monetary policy and fiscal policy.

By contrast, the book includes a lengthy treatment of such topics as Bayes-Nash equilibrium in game theory, intertemporal optimization of consumption and the theory of mechanism design.
Now compare that to what Glasner wrote:
The index contains not a single entry on the price level, inflation, deflation, money, interest, total output, employment or unemployment. Which is not to say that none of those concepts are ever mentioned or discussed, just that they are not treated, as they are in traditional macroeconomics books, as the principal objects of macroeconomic inquiry. The conduct of monetary or fiscal policy to achieve some explicit macroeconomic objective is never discussed. In contrast, there are repeated references to Walrasian equilibrium, the Arrow-Debreu-McKenzie model, the Radner model, Nash-equilibria, Pareto optimality, the first and second Welfare theorems. It’s a new world.

That's too similar to be a coincidence, so my best guess is that Quiggin got a bit of information (or none) from the book itself, and learned most of what he needed from Glasner's post. But note the difference between Glasner and Quiggin. Glasner says there are some things not in the index that he thought should have been there, and he says those things are actually discussed in the book. Quiggin claims those things are not discussed in the book. What's the truth? Well, there are extensive discussions about business cycles, unemployment, policy rules, monetary and fiscal policy, etc., in Kartik's book. There's certainly an extensive discussion of the theory, but of course that's necessary to get where the author wants to go, which is explaining how the theory is used as an input to policy, and to understand what we observe.

So apparently, Quiggin hasn't done his homework, but he's willing to make some bold conclusions:
The result is that there is almost zero intersection between Big Ideas in Macroeconomics and what I would think of as macroeconomics. It’s not so much that I think Athreya is wrong is that we are talking past each other. As Charles Goodhart said of DSGE, Athreya’s version of macro excludes everything in which I am interested.
So is this just "talking past each other." I don't think so. In my experience, Quiggin is not willing to engage. I took some trouble to write a defense of modern macro/critique of Zombie Economics, and Quiggin took note of that in some comments on a blog post of mine. Then he made excuses about being busy, and disappeared.

A typical criticism of modern macro is that it is too technical for a layperson to understand, and that macroeconomists make no effort at accessibility. Well, in this case, Kartik's goal is to be accessible. Certainly Glasner is not accusing him of being difficult to understand. Talking past each other? Baloney.

Shortly after Quiggin wrote his post, Noah Smith added his two cents' worth. He's certainly up front about how he researched his post:
I myself have not yet read the book, but David Glasner and John Quiggin have, so I'll be lazy and free-ride off of their effort.
So, Noah is freeriding on Quiggin and Glasner, but it seems likely that Quiggin knows no more about Kartik's book than what he read in Glasner's post. Further, though Glasner (to his credit) seems to have read the book, he's got some hostility toward modern macro - claims we're "methodologically arrogant," whatever that means - and his post glosses over some of the more important contributions to be found in Kartik's book.

So, having not read the book, but rather the Cliff's Notes version filtered through Glasner and Quiggin, Noah is ready to state the following:
Possibility 1: Athreya is trying to balance out the public discussion of macro. He knows that a lot of lay people and bloggers already talk about Keynes and Friedman, monetary policy and fiscal policy. His intent was to write a book about all the other macro stuff that the public doesn't know about - general equilibrium and game theory, incomplete markets and search frictions, and so on.

Possibility 2: Athreya just doesn't care that much about the stuff he leaves out. He's not interested in the history of economic hypotheses, only in the history of economic methodology. He cares vaguely about policy and about the economy itself, but only insofar as it's an interesting application of his beloved methodology. The ideas he thinks are "big" are ideas about how to make macro models, not about what assumptions go in the models or what conclusions come out of the models.

I don't know Athreya, but my intuition is that whether Possibility 1 is true or not, Possibility 2 is true - Athreya seems to be in love with modern macro methodology. If he were just trying to balance out the public discussion, he would probably have been more explicit about that goal. Also, the fact that he doesn't seem to talk much about evidence is telling. "Big ideas" apparently doesn't mean ideas that are successful in explaining the data, it means ideas that are neato and cool.
So, Noah's first big mistake is taking Quiggin at his word. Quiggin's claim that Kartik left out discussion of unemployment, policy, etc. is false. But of course Noah goes further. He doesn't know Kartik, as he points out, and he hasn't read his book, but he's going to use his intuition to look inside Kartik's head and tell us how he thinks. Basically, Noah is going to make up some stuff. Very un-neato and un-cool, Noah. On the disgusting side, actually.

So, we know that when this crew gets together, Paul Krugman can't be far behind. It's clear that Krugman didn't read the book either. He only read what Quiggin wrote. Further, we know Quiggin's post is more about Zombie Economics than about Kartik's book, and to the extent it deals with the book, falsely represents what is in it. So this has become a game of telephone, where whatever was stated in Kartik's book comes out in Krugman's blog as
John Quiggin looks at a recent book that purports to explain the big ideas in macroeconomics, but doesn’t contain any, well, macroeconomics.
And, well, that is one big falsehood.

It's now been more than four years since Krugman wrote How Did Economists Get It So Wrong? If you remember, these were the instructions Krugman gave us in 2009, in that piece:
So here’s what I think economists have to do. First, they have to face up to the inconvenient reality that financial markets fall far short of perfection, that they are subject to extraordinary delusions and the madness of crowds. Second, they have to admit — and this will be very hard for the people who giggled and whispered over Keynes — that Keynesian economics remains the best framework we have for making sense of recessions and depressions. Third, they’ll have to do their best to incorporate the realities of finance into macroeconomics.
So, let's go through the checklist.

1. Sure, financial markets are imperfect. This was something that we knew in 2009. By then, macroeconomics was replete with work on incomplete markets, financial contracts, banking, credit market frictions, etc. Krugman didn't have to tell us to think about those things. As to "delusion and madness of crowds," we have Shiller to tell us about delusion, and the "madness of crowds" is Roger Farmer's territory.

2. Is Keynesian economics the "best framework for making sense of recessions and depressions?" If by "Keynesian economics" we mean the version of Keynesianism that is practiced by current researchers, i.e. New Keynesian economics, and if making sense means that a baseline NK model with sticky prices and wages can explain all the macroeconomic phenomena we might be interested in, then the answer the profession has delivered is a resounding no. The interests of macroeconomists have shifted, naturally of course, to work on financial frictions of various kinds. In the work of freshly-minted PhDs in economics, there is distinctly less work on sticky prices than was the case before the financial crisis. I think a typical modern macroeconomist understands that there is a wide array of frictions that we want to think about when explaining what is going on in the world and understanding the role for economic policy. For some problems (perhaps all - we're really not sure) sticky prices and wages may be peripheral or totally unimportant. I don't get the same feeling of Keynesian religious fervor that used to exist among some people.

3. The realities of finance? I think we've made a lot of progress but, again, it wasn't like we were finance dummies before the financial crisis.

So, what effect has Paul Krugman had? Has Krugman changed how macroeconomics is done? Have we taken his messages to heart? Hardly. For your average working economist, what Paul Krugman thinks is irrelevant. But John Quiggin claims that Krugman and Brad DeLong are "setting the terms" of at least some elements of the debate on public policy. Here's my conjecture. I think it's likely that Kartik Athreya has had a larger effect on macroeconomic policy in the United States of America than Paul Krugman and Brad DeLong put together. Kartik is a Vice President at a Federal Reserve Bank, and his views affect those of the President of the Richmond Fed with regard to both monetary policy and banking regulation. Krugman and DeLong are journalists. Have their views helped to persuade anyone about anything that had any tangible effect on a policy decision? I'm not sure.

Kartik is a serious macroeconomist. As is typical in regional Federal Reserve Banks now, and at the Board of Governors, Kartik is expected to be engaged in research at the level of academic economists. Here's his CV, which includes work on credit, bankruptcy, default, public insurance - a lot of things that critics of modern macro seem to think we don't do. Part of Kartik's job is also to do policy, and at the Richmond Fed there is a lot to think about. In addition to monetary policy issues, the Richmond Fed district includes Charlotte, NC, where Bank of America is headquartered. So bank regulation is a big deal for the Richmond Fed economists. Kartik thus has a wealth of experience, which he has used to write a book which explains modern macro and how we do it. Pay special attention to Chapter 6, where he discusses how the available tools can and were used to make sense of the financial crisis. Big Ideas in Macroeconomics, is an excellent book, and I think John Quiggin, Noah Smith, and Paul Krugman should read it. You should too.

93 comments:

  1. Steve, take a proverbial Chill Pill.

    I do intend to read the book, and also, I recommended everyone read the book.

    My claim was that Athreya looooooves modern macro methods. Do you think my claim was wrong? If I didn't use Athreya as my example, I could have just as easily used you!

    I have read some of Athreya's research. Actually it wasn't what I expected - it was pretty simple models of credit markets.

    But anyway, calm down!

    ReplyDelete
    Replies
    1. The implication seems to be that my words are hot. I think my tone is actually cool and reasoned. You have to understand that you are discussing real people. These aren't some fictitious opponents. What you did was to comment on an individual's motivations and thought processes, without giving the poor guy a chance by actually reading his book. So, get it into your head that this is inappropriate, and try to do better next time.

      Delete
    2. First you trash his book, and now you're recommending that we all read it. Both without your actually having read the book. Are you now going to recommend that everyone run five miles because there's a chance that you might do it in the distant future?

      So which is it Noah? As they say, the hits keep on coming.

      Delete
    3. I run 3 1/2 miles every other day. I highly recommend it.

      Delete
    4. Well Steve, you said you were "disgusted". That sounds like a strong emotional reaction...

      Delete
    5. Interesting. Maybe it's hard to separate the intellect from the animal in us all. Saying "calm down" is a bit off though. It suggests the person has lost it. I'm not sitting here with a red face shouting at your photograph.

      Delete
    6. I'm not sitting here with a red face shouting at your photograph.

      Just for the record, if you ever do do that, I want a recording of it.

      Delete
    7. You know the guy who plays Moriarty in the recent BBC version of Sherlock Holmes? I think he could set me off.

      Delete
  2. I took some trouble to write a defense of modern macro/critique of Zombie Economics, and Quiggin took note of that in some comments on a blog post of mine. Then he made excuses about being busy, and disappeared.

    That's not how I remember it. first you wrote a rather dismissive blogpost, which John Quiggin responded to on the Crooked Timber blog. Paul Krugman noted that you were a bit full of yourself. You then wrote a more thoughtful critique of the book, which John Quiggin again responded to on Crooked Timber. You remarked (in comments here) that he was engaged in philosophy rather than economics and left it at that.

    ReplyDelete
    Replies
    1. For anyone who cares, here are the links confirming Donogue's account http://crookedtimber.org/2011/08/04/not-even-wrong-is-not-praise/
      http://crookedtimber.org/2011/09/24/the-poverty-of-rationality/

      Delete
    2. Excellent. Welcome back. Hmm. Doesn't seem to me that you ever successfully responded.

      Delete
    3. Why I am I not surprised by this?

      Delete
    4. I looked at some of the 3-year-old stuff. I'll concede that "running away" doesn't quite capture where we left off. But, you were being dodgy for sure. You didn't address the main criticisms.

      But, I'm all ears now. Talk to me. I'll repeat my comment from below:

      So let's get down to what's actually bothering you. You have some concerns, which I'd say are income and wealth distribution, the role of the government, and the state of the labor market. I'm concerned with those things too - what might be bad about the state of the world and how we can make it better. All this modeling we do is for that purpose. As I wrote before (or something like this), modern macro was not developed to bludgeon the poor and the unemployed.

      Delete
    5. Still waiting here. Just thought of this: Zombie Economics was in part a criticism of modern macro. It says that economics - macro in particular - contains some bad ideas. Then, according to your argument, the events of 2008 made obvious that those ideas are bad. Well, in Chapter 6 of Kartik's book, he goes to some effort to make the case that all of the modern macro he went through in Chapters 1-5 can be brought to bear in making sense of the financial crisis. So, I would think you would be very interested in that. But you're dismissive. You claim you're not interested. I would call that a failure to engage, but maybe I'm wrong.

      Delete
  3. Then there was Kartik's previous intervention that made him famous ("macroeconomics is hard and does not belong to blogs"), or maybe infamous as the Ph.D. from Iowa for technocratic rule.

    ReplyDelete
    Replies
    1. That's how some people may have mischaracterized what he wrote. If you read the actual document, I think you'll find a carefully measured critique of one aspect of what people (DeLong and Krugman in particular) do in blogs. They sometimes write as if the answers to our economic problems are incredibly obvious. Kartik's claim, is that we have to be more careful about it. That's not exactly arguing for "technocratic rule." Maybe you think all science is a big conspiracy? And why exactly is the PhD from Iowa a relevant consideration?

      Delete
    2. Sometimes the answers are incredibly obvious. Arguing for expansionary fiscal policy while we are at the ZLB is not something only Krugman and DeLong and all those other nasty Old Keynesians who never too Lucas seriously do, it is what MAINSTREAM MACRO models suggest.

      What you are doing is either defend Chicago morons like Cochrane who adopt the treasury view and thus repeat 80 year old fallacies or obfuscate the issue by saying , just like Kartik, that macro is too complicated and that there are no clear answers and so on.

      Delete
    3. OK, if the answers are incredibly obvious, tell me the following.

      1. "While we are at the ZLB..." Why are we there?
      2. It seems the central bank could at any time choose to be at the ZLB. So does this mean that if would be a good idea, under any conditions, for the central bank to choose to be at the ZLB, and for government spending to increase?
      3. What exactly are the effects of government spending?
      4. Is there a multiplier?
      5. Through what mechanism do we get the multiplier?
      6. How large is the multiplier?
      7. It's clear there is a basic inefficiency involved, otherwise you would not be calling for government action. What is the inefficiency, exactly, and how does the increase in government spending correct the problem?
      8. Does it matter what goods and services the government spends on? Explain why.
      9. How long should this increase in government spending last? What determines that?
      9. How should the government spending be financed? Why?
      10. What would be the long-run determinants of the level of government spending? How should we integrate these long-run considerations with short run fiscal policy?
      11. Is the level of the government debt a concern, or does it matter? Why?
      12. Should the potential for default on government debt make a difference to the calculation? Why?

      I could go on, but work on that to start with. I want precise answers.

      Delete
  4. I went to the Amazon page for the book.

    Unfortunately, I didn't find any big ideas in the book summary. Too much to ask?

    http://www.amazon.com/Big-Ideas-Macroeconomics-Nontechnical-View/dp/0262019736

    ReplyDelete
    Replies
    1. The ideas aren't amenable to sound bites, I think. You have to read the book to understand what the title means.

      Delete
    2. To be clear, I actually bought the book and look forward to delving in. The title, though, seems misleading. If you can't say what an idea is, it may be helpful, it may be interesting, but it is not "big". I'm exempting quantum physics from that characterization. Everything else, it seems appropriate.

      Delete
    3. I guess you could think of the book as an answer to a question: "What are the important ideas in macroeconomics?" Kartik thinks about that for a while, starts writing, and before he knows it he has 400 pages of stuff. Then, what do you call the thing? "Important Ideas in Economics" doesn't seem too compelling. "Big Ideas in Economics" is more catchy. It makes you want to read it.

      Delete
  5. I've read about half of the book. And I commented on it back on January 29: http://robertvienneau.blogspot.com/2014/01/economics-too-hard-for-kartik-athreya.html.

    The book is too boring to be accessible. I think it would be more interesting if it had some account of the personalities that created these ideas. I do not how to do that and shorten it.

    I think one minor change could have helped: break it up into more chapters or have a more comprehensive table of contents (with subsections listed) so the overall argument is easier to see.

    ReplyDelete
    Replies
    1. To make it accessible to a crackpot like you Kartik would have needed to write it in crayon and use only addition.

      Delete
  6. I should have mentioned that Williamson should mention that he's in the acknowledgements.

    ReplyDelete
    Replies
    1. Excellent. You indeed have read the book. I'm in the index too.

      Delete
  7. Just finished reading the book. While I did think it could have been more engaging, overall I really enjoyed it. As someone who is sympathetic to (some of) the critiques of modern macro it really annoys me to see people dismissing the book before reading it and mainly because of something the author wrote in the past rather than the substance of the book itself.

    ReplyDelete
  8. "Notice that Athreya makes no distinction between a reduction in consumption in which people shift into long-term real or financial assets and one in which people shift into holding cash. The two cases are hardly identical, but Athreya has nothing to say about the demand for money and its role in macroeconomics."

    I seriously doubt that the entire book ignores money and is pre-Friedmanian, pre-Keynesian and pre-Wicksellian (and thus simply not macro but just overblown micro).

    But I think we have to seriously wonder whether starting with a perfect world, Arrow-Debreu GE, and then adding frictions (basic DSGE does not even incorporate one friction, monopolistic competition, it uses the short-cut of Calvo pricing; so much about rigorous microfoundation and no ad hockery) should be the only approach in macro.

    Of course we should do general equilibrium analysis and in general I like the mainstream approach. But when something is too complex to include in DSGE we might not wanna wait for decades but in the meantime make due with a more stylized model or partial equilibrium analysis.

    Let's also not forget that Arrow-Debreu is anything but a robust model. Leijonhufvud has shown back in the days that without a magical Walrasian acutioneer trade at disequilibrium prices can lead to a collapse of demand, imperfect competition and non-convexities lead to problems and Stiglitz and Greenwald have shown that when there are incentive issues due to asymmetric information (which is basically always the case) the two welfare theorems do not hold anymore.

    I am all for GE and analyzing market failures but I am also opposed to "one model to rule them all" (and as I tried to indicate above, there have been many potentially fruitful research paths in the sixties, seventies and eighties which have died out because of this). More Model diversity should be in the economic spirit of competition.

    ReplyDelete
    Replies
    1. "I am all for GE and analyzing market failures but I am also opposed to "one model to rule them all""

      I'm in complete agreement.

      Delete
  9. "If by "Keynesian economics" we mean the version of Keynesianism that is practiced by current researchers, i.e. New Keynesian economics, and if making sense means that a baseline NK model with sticky prices and wages can explain all the macroeconomic phenomena we might be interested in, then the answer the profession has delivered is a resounding no."

    I totally agree with this. Without wanting to sound too much like a Stiglitz fanboy, in the eighties and nineties he wrote a bunch of papers in which he dissed the New Keynesian literature (wages fell during the Great Recession so how can wage rigidity be the source of the problem?) and focused on financial imperfections to explain macro phenomena. Kind prophetic.

    About the term New Keynesian, as the Woodfordian price rigidity explanation is precisely the position against which Keynes argued in the thirties the term is at least from a historical perspective false.
    And while I am anything but a fan of the New Classical revolution in macro, the claim that price/wage stickiness was a cheap trick to re-Keynes-ify RBC (then again basic DSGE, sans ZLB and sans any frictions besides imperfect competition, is more monetarist in its predictions) is not totally ludicrous.

    ReplyDelete
    Replies
    1. "in the eighties and nineties he wrote a bunch of papers in which he dissed the New Keynesian literature (wages fell during the Great Recession so how can wage rigidity be the source of the problem?) and focused on financial imperfections to explain macro phenomena."

      Yes, Stiglitz said something like that fairly recently - sticky prices and wages give you the standard types of distortions you get from taxation and, he argued, the welfare effects can't be that big, nor could they give you a mechanism to produce the magnitude of fluctuations we actually see. Then he said something about how it puzzled him that models used at central banks didn't have banks in them. So, good for him.

      On Woodfordian economics, yes, the typical model is just baseline RBC with a twist. I wouldn't call it a cheap trick, though. There's some work to get there.

      Delete
  10. I’m really glad that you wrote this post. I almost never write or comment on blog posts mostly because I don’t think I have the depth of expertise to write something intelligible and interesting to the general public. I also don’t particularly care for being the subject of personal attacks – and the comments on Kartik (note that I say “Kartik” and not “Kartik’s book”) show that my fear is justified. I have no problem with vehement arguments about content. I do have a problem when people argue about content without knowing it. And I have a big problem when arguments attack the person and not their ideas.
    “Caring vaguely about policy” wouldn’t have got Kartik very far at the Fed where economists are expected to contribute to actual policy and where people can and do get fired for not doing their job.
    If Kartik is in love with anything, it’s economics. He is passionate about communicating economic ideas to non-economists and continues to be willing to put himself out there and engage with the general public despite the inevitable blogosphere backlash. He’s a braver man than I!

    ReplyDelete
  11. I absolutely LOVE Athreya's book. Actually, I think of this book as a short "Modern History of Economic Thought" book. Would I have done a couple of things differently or worded differently, or presented differently? Perhaps, but that is true with any book.
    Athreya has done a really magnificent job. It is not easy to summarize years and years of modern economic thought into non-technical pages. It is not easy to summarize "why modern economists think what they think" into a non-technical version of it.
    I am with Stephen here. For anybody interested in modern econ, read the book.

    ReplyDelete
  12. "What you did was to comment on an individual's motivations and thought processes, without giving the poor guy a chance"

    Pot, meet kettle. I hadn't read Glasner's review until it was linked in comments to my posts. The fact that we both made the same point reflects the fact that this point is glaringly obvious. The main issues in macro (as most people understand it) aren't mentioned in the index and are only discussed *briefly* (note that you misrepresented me here) in the main text. I don't have the book to hand right now, but having checked carefully when I reviewed it I recall a brief mention, and dismissal, of involuntary employment, and nothing on the Phillips curve. Taylor rules, or fiscal policy, at least in the sense I understand the term. Perhaps you like to give some page references to the extensive discussions I missed.

    ReplyDelete
    Replies
    1. Unemployment: See Section 5.11 beginning on pg. 277 (Search Model) That continues through pg. 286. The discussion of "Search Models and Voluntary versus Involuntary Unemployment continues for two pages 283-284.

      Fiscal policy: Here the index is helpful. Try: (i) goods, public; (ii) government, in economic modeling; (iii) taxation; (iv) time inconsistency.

      Taylor rules and Phillips curve: In Chapter 6 (excellent discussion of recent events, by the way) pages 361-363, there's a discussion of DSGE, with particular mention of the "new-Keynesian "trinity." That's IS curve, Taylor rule, and Phillips curve. He doesn't use exactly those words, but there it is.

      I'm certainly willing to concede that modern macro may look different to you from what you were taught at ANU as an undergraduate. It certainly looks different from what I learned in 1975. But I'm puzzled. You're well-known as a decision theorist, so my guess would be that someone like you would look at Kartik's book, see familiar ground, and welcome the change in macro. Great, it's just normal economics!

      Delete
    2. I take snippets of 2-3 pages in a 400 page book as "mentioned in passing". YMMV

      And by "fiscal policy" I don't in this context mean "public finance", I mean stimulus vs austerity, the multiplier, crowding out and all those other old-fashioned things I learned about in the 1970s.

      Delete
    3. John, every one of the questions you ask about is tackled by macro. But they are all done with variants of the growth model, or the OG model, or the incomplete-markets model. One has to have some sense of these models in order to make contact with analysis of any of the topics you care in a meaningful way. The distributional impact of austerity, which you and many others are clearly, and understandably, concerned with, are precisely what the latter two model classes allow you talk about have allowed us to tackle. Please, try to moderate your views towards the very thing that is bringing your concerns to the table that never before could have been.

      Delete
    4. Now we're just quibbling about what's enough of this, and what's enough of that. You shouldn't want to see the "old-fashioned" things. You already know that.

      So let's get down to what's actually bothering you. You have some concerns, which I'd say are income and wealth distribution, the role of the government, and the state of the labor market. I'm concerned with those things too - what might be bad about the state of the world and how we can make it better. All this modeling we do is for that purpose. As I wrote before (or something like this), modern macro was not developed to bludgeon the poor and the unemployed.

      Delete
    5. John Quiggin is simply not very good at economics. Face it, John, you're intellectually incapable of discussing macro; stick to regret theory and 1982.

      Delete
    6. Shows what a newbie you are anon. Regret theory is Loomes and Sugden's work.

      Delete
    7. Yes, he did anticipated utility, which is within the broader class of non-expected utility theory. Hard to see what why someone who wrote his 1982 paper would get bent out of shape about modern macro. We live to be surprised.

      Delete
    8. There is nothing to be surprised here. Modern macro has unlearned lessons that were known decades ago.

      Delete
    9. I do not see the contradiction. If you believe that expected utility theory has major problems associated with it then modern macro must just seem ridiculous to you.

      Delete
    10. Non-expected utility theorists wanted to modify the existing theory to explain data that expected utility theory could not. Part of the modern macro research program is a drive to develop and refine the theory to try to explain features of the data that existing theory has trouble with. Seems like a similar enterprise, don't you think?

      Delete
    11. Yes, I do see what you mean. However, non-expected utility theory arose from the fact that experimental evidence indicates that people do not behave in the ways expected utility theory requires them to. Yet, the vast majority of modern macro takes expected utility theory as given in its analysis and then claims to be able to explain important things with it. I would expect somebody who doubts the validity of expected utility theory to find that approach somewhat strange.

      Delete
    12. Apparently the poster above is unaware of the idea of "approximation". Expected utility may not be "true", but it might be approximately true. And for some questions we know the answers don't change when we abandon it (see Tallarini's JME 2000 paper, for example).

      And welcome Irineu de Carvalho Filho, new crackpot! If you work hard you'll join the ranks of Robert Vienneau, Greg Ransom, John D, and other luminaries!

      Delete
    13. Interesting, are you suggesting that this "approximation" invalidates any concerns somebody may have about the use of expected utility theory? That sounds like a fantastic result.

      Delete
    14. The fellow who calls other posters crackpot is obviously unaware that "approximately true" can imply huge welfare deviations.

      Yellen and Akerlof wrote a couple of papers in the eighties on near-rationality in which they used the envelope theorem to illustrate that not optimizing constantly can makes sense as the region near the optimum is "flat" and as there might be some transaction costs when you reoptimize. While changes in the objective function are second-order there can be first-order welfare changes.

      Delete
    15. There is a debate, of course, about whether the evidence actually says that expected utility does not fit the relevant data. If it's experimental data, we may not trust the experiments, for example. But suppose we thought that expected utility theory was wrong, in that it is inconsistent with some features of the data. It may be the case that for macro phenomena, expected utility is fine. Practically, we can't explain everything, and we just want our models to be "right" in terms of the purposes for which they were designed.

      Delete
    16. Experimental data is not perfect by any means but it cannot be ignored either. And it does by and large suggest that people do not adhere to the axioms of expected utility theory.
      I agree that if that does not matter for macro phenomena then there is no issue. I'm not sure why that should be the case though.

      Delete
    17. People have certainly thought about this. Here's one paper about how non-expected utility theory can be used to think about asset pricing:

      http://www.stanford.edu/~schneidr/review20.pdf

      There are others too.

      It's well known that baseline asset pricing models do not fit some features of the data - price volatility, equity premia, etc. One avenue that's been pursued is that there may elements of preferences that aren't captured well. So, people have thought about non-expected utility, habit persistence, and other stuff.

      Delete
  13. Question for all those urging me to buy Athreya's book: is there anything in it that I won't find in “Microeconomic Theory” by Mas-Colell, Whinston and Green? I bought the latter for €5 when the local bookshop went bust.

    If there isn't, then I'd guess that in some jurisdictions Athreya could be open to legal proceedings for passing off Micro as Macro. If he thinks they are the same thing, he needs to make that case.

    ReplyDelete
    Replies
    1. Kevin,
      With all due respect, your question is silly. No, you do not need to buy the book. You can just read 40 years of paper of economic theory, plus the theory textbooks that put them all together, and have the same outcome. Just do that, and you will be fine.
      Athreya's intention was to summarize of why do modern (macro)economists think the way they think. And yes, he goes back to basics, to the Walrasian Equil, because that is where it all started (at least, one of the starts). He goes on to justify the Walrasian Equil, because there is a huge, repeat huge, literature showing that even when you introduce changes, distortions, etc, you somehow bounce back to the WE.
      Then he has ample discussions on equity and efficiency, and on modern growth models.
      Do you need the book? As I said, no, you do not need it. Read the literature and papers of the last 40 or 50 years, and perhaps this helps you.

      Delete
  14. I haven't quite finished "Big Ideas," but I'll hazard a question nevertheless.

    Kartik argues that “in any thick market, especially any involving purely private goods and services, outcomes *will* be well described as Walrasian” (emphasis in the original, p. 67). This seems puzzling because the prices for books offered on Amazon.com, which meets most of Kartik’s criteria for a Walrasian clearinghouse, don’t look at all Walrasian. I checked the prices offered for a new copy of “Big Ideas in Economics,” including standard shipping, on Amazon. There were 25 offers, ranging from $23.99 to $75.81, with 10 offers at $50 or more. [I bought the book at the Amazon Prime price of $35.11 (with free 2-day shipping) and am pleased with the purchase.]

    This kind of price dispersion, which is not uncommon, for an identical good with the same shipping cost doesn’t seem consistent with prices in a Walrasian competitive equilibrium. I should add that, on this particular day, there was weak negative correlation between “percent positive ratings” and offer price, i.e., sellers on Amazon with a high percentage of positive ratings tended to offer "Big Ideas" at lower prices. Is this pattern of prices “well described as Walrasian”? This isn't a rhetorical question; I'm willing to be persuaded, but need to see the argument.

    ReplyDelete
    Replies
    1. I see the relevant issue as being the one of "at what prices do the trades take place?" And here, I would strongly suspect that almost everyone that bought the book paid roughly $35 for it. So that's the "walrasian" prediction. As for why some sellers post other prices, we can't be sure, but from the perspective of both buyers and sellers, the question is: if I post $35, "will I sell a copy?" / "can I buy a copy?" And I think the answer is roughly "yes."

      Delete
    2. I guess the question is what we might mean by a "thick market." Kartik does discuss search in his book, which is an approach that's been used to think about price dispersion. The remarkable thing is that you might think that dispersion goes down with internet trading etc. But that does not seem to have happened.

      Delete
    3. Anon, you write, "I would strongly suspect that almost everyone that bought the book paid roughly $35 for it. So that's the 'walrasian' prediction."

      Are you saying that $35 is the "walrasian prediction" because you believe that's what everybody who bought the book paid for it? What would you say if it turned out that some buyers actually paid $25, while some paid $50?

      Leaving our speculations about actual sales prices aside, why would someone pay $35 for "Big Ideas" when the book was offered for $24 by two sellers who have 99% positive ratings? And why would 19 sellers offer prices ranging from $40 to $76, when sellers can go to Amazon.com and see for themselves that there are highly rated sellers offering the book for $35?

      Wouldn't you agree that a much narrower range of prices on Amazon would be more consistent with Walrasian prices?

      Delete
    4. Hi Stephen,

      Unless you define "thick markets" as those in which there's little price dispersion, it's hard to imagine a thicker market than Amazon, well, it's hard for me at least.

      Delete
    5. The prices you are discussing are not transactions prices, right?

      Delete
    6. If you go onto Amazon, you'll see the prices I'm referring to. I don't have access to the actual transaction prices, though I can say that I've never bought a book on Amazon at a "low" offered price and had my "purchase offer" rejected.

      Delete
    7. I don't think anyone is questioning the idea that the low priced items are sold (which is what you are suggesting in your last comment), the comment from the previous anonymous poster is that the high priced item is not selling. Sure you can post a price of $50 to sell a paper clip, but you are never going to sell the paper clip so that is not a relevant "transaction price." The transaction price is the $0.05 that the paper clip typically sells for.

      The fact that there can be some dispersion in the sales price at the low end, some get sold for $25 and others for $35, can be explained by a simple price posting story. The seller posting the $25 item earns a small profit but sells the item very quickly. The seller posting the $35 item earns a larger profit but it takes much longer to sell the item. In equilibrium, when taking into account the speed of the transaction, the seller should be indifferent between posting the low price and selling quickly versus posting the high price and waiting a long time to sell.

      A third, but also likely explanation, is just that you post a price today at the market price, say $50, and tomorrow the item exogenously becomes a lot more popular, so the market price rises to say $75. You could take the time to go back every day and adjust your price, but that is costly and you run a shop that sells thousands of items on amazon so you choose not to. Instead you just leave the price of $50 and it sells right away. You could reverse the story for the price going down and have the item never sell (or take a very long time to sell).

      These are not static or steady state prices you are seeing on amazon, so that can explain most of the price spread.

      Delete
    8. Greg,

      I don't think this qualifies as an identical product. The book is the same, but the service is not. Some of the cheaper sellers have a longer shipping time than Amazon. All of them charge extra for shipping, which brings the price at the same level or higher than Amazon's, which offers free shipping if you also buy another item (thus bringing the total purchase to $35 or more). Some book sellers advertise themselves as "independent" so you are buying non just the book but the satisfaction that you helped a mom and pop shop. Others may not care so much about selling through Amazon because most of their sales are through a physical location, so they just post a price and hope that someone foolish enough may take it. It's complicated!

      Delete
    9. Isn't this just Stigler 1961? Sure, transaction costs of online shopping are smaller than those of non-online shopping, it takes more time to walk from shop A to shop B than to click from the site of shop A to the site of shop B.
      But it still takes time. Plus what CA said, a book isn't just a book, some retailers are faster, better at packing and so on and we should expect this to be priced in.

      I do not wanna deny Greg's point that trade at disequilibrium prices can occur. But I think that the output market features less market imperfections than labour and capital markets (at least the market of final consumer items, firm-to-firm markets often lack instantaneous payment so there is implicit credit and all the credit market imperfections apply).

      Delete
    10. CA, you write, "I don't think this qualifies as an identical product. The book is the same, but the service is not." Leaving Amazon Prime (2-day free shipping) aside, all of the offers to sell "Big Ideas" included a $3.99 standard shipping charge. Some sellers probably get their stuff out faster than others, but this "quality" may be reflected in the "% positive ratings," and, on the day I checked, sellers with a high percentage of positive ratings tended to sell at lower, not higher, prices.

      Anon (12:08 AM), I wasn't comparing online transaction costs with non-online transaction costs, but your point raises an interesting question. Suppose search costs are much lower for online markets like Amazon. What effect do these lower search costs have on the relative dispersion of prices on Amazon vs. prices on non-online markets? Also, given your theory, would you expect the price dispersion on Amazon to grow, shrink, or remain the same over, say, the next six months?

      Anon (4:08PM), you suggest that some sellers may post prices well above the minimum posted price, say $35 vs. $25, because they're willing to wait for the $35 sale. However, this "simple price posting story" only works if the prices for "Big Ideas" on Amazon are expected to rise rather than fall. Now, prices could rise, of course, but as used copies and remainders become available, prices could also fall.

      Delete
    11. You are hopeless.

      Delete
    12. @Greg: I don't have any elaborate theory, just Sitgler 1961, so I cannot tell how the price dispersion changes.
      All I know is that price dispersion for online shopping products can exist and persist for simple transaction cost reasons. Clicking takes time and sooner or later the marginal benefit of an expected lower price in a new online shop you discover is smaller than the shadow price of leisure/time and then you stop.

      If the point you actually wanna make is that the Walrasian auctioneer is a fairy tale and that trade at disequilibrium prices can exist I totally agree with you. But I don't think that your example is illuminating as there is too little data: we do not know whether trade is going on at the posted prices.

      Delete
  15. I'm not sure I'd read Quiggin as representing that he had carefully read the book. He says he's "just received" the book and he "just got a review copy" and he refers to a blurb on the back. Then he notes some things that are not in the index and goes on to describe not the book, but how he views Athreya's "world" which reads to me like a rehash of Quiggin's book, as you suggested (although I've not read his book).

    But perhaps he should have been more clear.

    ReplyDelete
    Replies
    1. I think it's fair to say he misrepresented the book, and also gave it short shrift, as you say. There are also some more subtle derogatory things. He mentions Kartik's piece on Bloggers (it's actually pretty good I think), and links to what could charitably called "disagreement" with that, and he never provides a link to Kartik's book. He also states that there is nothing in the book that he could find interesting. I find it hard to believe that none of Quiggin's work in economic theory bears any relation to what's in the book.

      Delete
  16. I am reading the book now. The adjective I would use to describe it is 'careful'. Athreya doesn't skip steps. To those who are skeptical or like solid foundations, this is a good thing.

    I am only a hundred pages in so far, but it strikes me as a discussion of how the important tools in modern macro were developed. This is understandable, as we need to hae the tools before we can use them.

    ReplyDelete
  17. "Krugman and DeLong are journalists. Have their views helped to persuade anyone about anything that had any tangible effect on a policy decision? I'm not sure."

    The President is fairly influential on these matters, and I would bet he reads the NY Times Op Ed Page.

    ReplyDelete
    Replies
    1. Here's a question for you. I've been doing this for almost 4 years now. Who actually reads these blog posts? Does it have any influence on actual policy decisions? Have I budged anything, or am a just hitting my head against the wall?

      Delete
    2. I should hire myself out as a book rep.

      Delete
  18. "The adjective I would use to describe it is 'careful'."

    I can understand this confusion. I am reading chapter 4, which discusses aggregation.

    Athreya discusses aggregating individuals into a representative agents. He distinguishes between an agent who is representative for positive work and for normative work. I do not recall him referencing Alan Kirman, but nevertheless, I found the discussion of interest.

    Then he talks about aggregating commodities, and the discussion is all about consumer goods. I do not see anything about aggregating capital goods or about aggregating production functions. If you can just ignore more than a half-century of literature demonstrating the invalidity of your approach, your reader might come away with a false impression.

    ReplyDelete
    Replies
    1. The Cambridge controversy. Haven't heard about that in a while.

      Delete
    2. That is a problem of course. If one ignores the Cambridge controversy, one may end up believing that RBC models are micro-founded, instead of ad hoc; or that measured TFP captures something akin to productivity shocks.

      Delete
    3. That's of course the big problem with TFP-based theories. What is it? This could be something going on at the firm level, or related to the allocation of factors of production across firms. What's the production function about? Surely a firm is a group of individuals, some heterogeneous machines, and some contractual arrangements and organization that bind them all together. And what are the boundaries of a "firm?" The deeper you go, the more interesting (and complicated) it gets. But you have to start somewhere. If it's too complicated, we can't learn anything from it.

      Delete
    4. This comment has been removed by the author.

      Delete
    5. Stephen, you write, "The Cambridge controversy. Haven't heard about that in a while." I can understand why you wouldn't have, but the history of the controversy is interesting, especially given some claims made on behalf of "scientific" economics.

      Summing up the Cambridge Capital debate, Paul Samuelson was gracious enough to admit that his side - i.e., the side defending the neoclassical aggregate production function - had lost the debate on logical-mathematical grounds. A single technology could be the cost-minimizing alternative at a high rate of interest, lose this standing at a somewhat lower rate of interest, and regain it at an even lower rate of interest.

      There was a second round of work focusing on the conditions under which an aggregate production function is a meaningful concept. The most significant conclusion of this research is that “the conditions under which a well-behaved aggregate production function can be derived from micro production functions are so stringent that it is difficult to believe that actual economies satisfy them. Therefore, aggregate production functions do not have a sound theoretical foundation" (Felipe and Fisher, 2003).

      "O well, it’s just an approximation, and besides an aggregate Cobb-Douglas production function fits the historical data pretty well. Full speed ahead!"

      Actually, it turns out that Cobb-Douglas only “worked” because the labor share had been roughly constant over a long period. The “good fit” arises from the manipulation of an accounting identity, not from the presence of a Cobb-Douglas production function!

      Back to Felipe and Fisher: “Solow’s residual (total factor productivity growth), the Holy Grail of the neoclassical growth model, is nothing but a weighted average of the growth rates of the wage and profit rates . . . where the weights are the factor shares.” Although this relationship was unveiled by Jorgenson and Griliches (1967) long ago, economists continue to rely on aggregate production functions, apparently without realizing that this “dual measure of productivity” is “a tautology because it follows from an identity and has no relationship to the notion of aggregate production function, which does not exist."

      So, when one of the profession’s high priests praises the neoclassical production function as a construct that “deals with well-defined aggregate inputs and outputs,” and that “a final beauty is that it is based on a lot of theory” (Prescott 1998), you really have to wonder whether Dr. Prescott’s mind has made contact with these criticisms, or, more broadly, whether economists who continue to use aggregate production functions realize that “the issue at stake is a whole way of thinking, central to which is the (macro) neoclassical approach to distribution” (Felipe and Fisher).

      Delete
    6. Citing two crackpots to justify other crackpot ideas isn't going to win the argument.

      Delete
  19. We cannot learn much from an enterprise that starts wrong from the outset.

    That is not to say that we can't learn anything - the enterprise is not 100% useless...

    But unless one wants to look like a fool, one should be very cautious before taking any result based on 'measured TFP' seriously, and more so at business cycle frequencies and at a high level of aggregation.

    ReplyDelete
    Replies
    1. Your criticism and reference to TFP echoes much of what has been written by Richard Nelson. The alternative he proposes is moving to computational models where aggregate output is computed as the result of decisions made by many heterogeneous firms that follow feedback rules (See for example Nelson and Winter, 1982). However, while such an approach addresses many of your concerns, we have yet to see major breakthroughs in our understanding of economic phenomena coming from this line of research. Of course, it is true that this research program has been relatively small.

      In any case, as I have done some empirical work concerning measured TFP (http://onlinelibrary.wiley.com/doi/10.1111/rode.12013/abstract) I would argue that the measure, as bad as it is, takes us a long way in understanding differences in productive activity across space and time. See also Hall and Jones (http://www.stanford.edu/~chadj/pon400.pdf), Alexopoulos (https://www.aeaweb.org/articles.php?doi=10.1257/aer.101.4.1144), Zachariadis (http://ideas.repec.org/a/cje/issued/v36y2003i3p566-586.html), Jorgenson and Stiroh (http://www.aeaweb.org/articles.php?doi=10.1257/jep.22.1.3) and others.

      Delete
    2. CA,

      I do not doubt that there are cross-country differences in productivity, as convincingly shown by Hall and Jones.

      I actually believe that mechanisms such as those documented by Parente and Prescott explain a good share of cross-country income differences.

      But that does not mean that I consider 'measured TFP' is an acceptable measure of productivity (it is not); it does not mean that I should take seriously any study that uses 'measured TFP' to explain business cycle fluctuations and concludes that those are driven by productivity shocks (as it is not clear to me that measured TFP are productivity shocks to begin with!).

      One of the other comments cited the paper by Filipe and Fisher on the aggregation of production functions. I strongly recommend that paper. It is an essential read.

      Delete
    3. I don't think TFP is a measure of technology. If this is what you are saying then I agree. But not a measure of productivity? Then what does it measure? I Hall and Jones that is precisely how TFP is treated: variations in output that cannot be accounted for by variations in inputs.

      Delete
  20. "Then there was Kartik's previous intervention that made him famous ("macroeconomics is hard and does not belong to blogs")..."

    The most relevant comment of all.

    Athreya loudly announced that he was a huge A-hole a while back , and people are supposed to forget ? The very bloggers and blog-readers he trashed ?

    He deserves all he gets , and then some.

    ReplyDelete
    Replies
    1. All I can say is it's plain what you are announcing.

      Delete
    2. "Athreya loudly announced that he was a huge A-hole a while back , and people are supposed to forget ? The very bloggers and blog-readers he trashed ?"

      They deserved worse, and I told Kartik that at the time. It seems you are one of them, and so I'll say to you what should have been said earlier; you do not know what you are talking about, so you should shut up.

      Delete
  21. Nice review of John Quiggin's book here:

    http://www.goodreads.com/review/show/198058024?book_show_action=true&page=1

    ReplyDelete
  22. This comment has been removed by the author.

    ReplyDelete
  23. "I'm not sitting here with a red face shouting at your photograph."

    It's interesting you say that Steve, as it may be a misperception but if so it's one I've had too. Surely you shout just a little bit when reading Krugman?

    ReplyDelete
  24. I also think in all honesty you're taking it a bit far in calling Krugman 'a mere journalist.' He isn't only that if Arethya is a fine macroeconomist, Krugman is no slouch either-he did win a nobel prize.

    It's things like that that make me think you have Krugman Derangment Syndrome-by being totally uncharitable to him.

    ReplyDelete