Monday, March 16, 2015

Levine on Keynes

This piece be David Levine is a lot of fun. Paul Krugman refers to Levine as "incompetent" and "ignorant," which I think we can interpret as a strong positive signal.

80 comments:

  1. Levine's piece is equivalent to a modern physicist mocking the idea of the speed of light being constant by proudly stating that all speed is relative to a reference frame and therefore the speed of light can't be constant, because logic duh! Of course this is what happens if you forget or simply ignore (and or are ignorant of) everything that was hard learned by the important work of Einstein and others, even if it's slightly counter intuitive at first. Just like David Levine forgetting, ignoring or being ignorant of the hard work done nearly two centuries ago by Mill or even Malthus, let alone Keynes and Friedman, as Brad Delong shows: http://www.bradford-delong.com/2015/03/time-for-a-rant-why-oh-why-cannot-we-have-better-economists.html

    ReplyDelete
  2. Dude, is the behavior of light particles governed by logic the same way the behavior of humans is? It is a bad habit for people to use analogies from natural sciences to describe economic models. Quit, it will do you good!

    ReplyDelete
  3. As far as DeShort's rant, let's see, he is discussing the appropriate monetary response to fluctuations in the demand for money (he assumes of course that producers cannot simply lower their nominal prices but, being himself, he does not bother to be upfront about it). Seems to me, this is not exactly what your average Keynesian has in mind, and not what Levine is writing about.

    ReplyDelete
  4. The point isn't about how logically the behaviour of humans can be analyzed. The point is not that Keynesian economics is right in the way the speed of light is right, as Keynesian/Friedmanite economics could easily be wrong.

    It's that this is completely ignoring basically all Keynesian, pre-Keynesian, post-Keynesian and even monetarist arguments for concepts like deficient demand, classical economics only being a special case at full employment, how adding money to the mix changes the analysis fundamentally etc.. It's basically returning to dark age economics where we're starting right back at square one, as if we've forgotten the last 170 years of economics - nobody should be having to teach basic economics to a distinguished economist. Just like nobody should be teaching special and general relativity to a physicist.

    ReplyDelete
    Replies
    1. As Levine points out in the piece, he was well-schooled in Keynesian economics at UCLA, where he was an undergraduate. Then he went to MIT for graduate school, and graduated in 1981, at a time when MIT had more Keynesians than you could shake a stick at. Krugman of course also graduate from MIT, 5 years earlier. So Levine is well-schooled in Keynesianism. I'd venture to guess that his Keynesian credentials top yours. So accusing him of not knowing it is pretty silly. Levine is a top economist in the world, and he's trying to make sense out of Keynesian economics. What's deficient demand anyway? People talk about that as if it is self-evident, but I'm afraid not, which is part of what Levine is clear about.

      Delete
    2. Reading the Brad Delong post, it would seem the deficiency is an increase in demand for safe, liquid assets. This takes demand away from other goods, and combined with price/wage frictions causes a recession.
      So, what is the problem with this simple story? Levine's post doesn't directly comment on the monetary aspect of the problem, so it's not very useful in that respect. Is something wrong with the mechanics of hoarding leading to a recession? assumptions on frictions? or perhaps there's something else going on (e.g. something has caused people to save more all of a suddenly, perhaps that something is the root problem) and we are misdiagnosing the problem?

      Delete
    3. "...it would seem the deficiency is an increase in demand for safe, liquid assets."

      Nothing wrong with that story. DeLong and I agree on that. I don't think it has anything to do with what's in the General Theory though. I just call it a shortage of safe assets.

      Delete
    4. Arguments by authority never end well, Stephen. David Levine probably fell asleep during Samuelson and Solow's classes...

      Delete
    5. "Arguments by authority never end well..."

      Never? You would probably say Google maps is an authority, I think. Do they always lead you astray?

      "David Levine probably fell asleep..."

      You're intuiting this, or what?

      Delete
  5. "Levine is a top economist in the world...."

    Here we go again: "top" economists, "elite" schools, "scholarly" journals. Give it a rest. If you had a case to make you wouldn't resort to this kind of tosh. Keynes explained what he meant by deficient demand, without resorting to this kind of blather. Of course he made no secret of the fact that he was Marshall's pupil, that he knew F.Y. Edgeworth and discussed philosophy with the likes of Ramsey, Wittgenstein and Russell. But he didn't try to use that as a bulwark for his theories; these were not the sort of people to be persuaded by argument from authority.

    I've no doubt Levine had good teachers and obviously he has a good brain. Sadly there's nothing in that piece to suggest he has much of a clue about Keynes, or any of the many varieties of Keynesian economics. It's a pity. He surely could write something good if he really put his mind to it.

    ReplyDelete
    Replies
    1. "there's nothing in that piece to suggest he has much of a clue about Keynes, or any of the many varieties of Keynesian economics"

      I think you contradict your own point. If one has to spend much time contemplating "what is Keynesian economics", then the theory is so vague that its usefulness in terms of policy analysis is limited. And Levine makes this point quite well!

      And while argument from authority is indeed not effective, it is necessary when half of De Short's response is an ad hominem attack on Levine and what he does and does not know.

      Delete
    2. Kevin, you don't seem to be getting to the substance of what he wrote. It's very easy to say something is bad and call it "blather." Why exactly? Why is it that I find it interesting and you don't? Why would you get so hot and bothered about it?

      Delete
    3. Kevin Donoghue: "Here we go again: "top" economists, "elite" schools, "scholarly" journals. Give it a rest. If you had a case to make you wouldn't resort to this kind of tosh...."

      The world is a complex place. None of us are well enough educated to understand anything but the most basic of analogies or argumentation in fields in which we do not specialize.

      No one would listen to Krugman or Delong if they did not have the credentials that you deride. They do not convince people because of their arguments.

      As I've never heard of you, I assume that you are not a practicing macroecoomist and from your post I infer that you are an ideologue.

      You don't understand the elucidated arguments for or against their views but you choose to support them because they argue for views that seem intuitively attractive to you.

      If Levine had conveyed much the same information but concluded that Keynesian thought is the solution to economic downturns, you'd be singing his praises.

      Delete
    4. Kevin, you don't seem to be getting to the substance of what he wrote. It's very easy to say something is bad and call it "blather." Why exactly? Why is it that I find it interesting and you don't? Why would you get so hot and bothered about it?

      Compare the discussion of involuntary unemployment in GT Chapter 2 with Levine's "explanation" of the concept. Do you actually assert that he's playing fair with his readers? Do you truly find it interesting, or does it please you merely because it annoys your foes? The idea that the GT is like an M. C. Escher drawing would be interesting, if Levine made any serious effort to show that this is so. All he actually gives us is a shaggy-dog story about a phone-maker who doesn't want a tatoo.

      Keynes's GT is available online. Anyone who cares can read the first few chapters and verify that Levine is making a dog's dinner of the argument. That disposes of "the substance of what he wrote" by showing that there's nothing at all to it.

      Why get hot and bothered about it? If Jonah Goldberg published something like that, nobody would imagine that it was a serious argument. Levine hopes to get away with it because he's a distinguished economist; so, when he claims to have read Keynes, students are likely to believe him. (Maybe he did; if so it was about as profitable as my attempts to read Heidegger.) As a result, they won't learn and in due course they will be the teachers of another generation of students every bit as ignorant as they are. That's how we get nonsense of the sort discussed by Simon Wren-Lewis at the link below.

      http://mainlymacro.blogspot.ie/2012/01/mistakes-and-ideology-in-macroeconomics.html

      Delete
    5. I'm not sure why you're imagining this is some kind of shady business. Everything is there up front to look at and discuss. You seem to be afraid of it. Why so insecure?

      Delete
    6. Shady business? I've never suggested that Levine is some kind of mafioso. And if I'm insecure, what of it? Does that excuse Levine's distortions? Keynes's concept of involuntary unemployment surely has its problems, as many economic concepts do. But it has nothing to do with phone-makers who decide they don't want tatoos:

      Men are involuntarily unemployed If, in the event of a small rise in the price of wage-goods relatively to the money-wage, both the aggregate supply of labour willing to work for the current money-wage and the aggregate demand for it at that wage would be greater than the existing volume of employment.

      If you think Levine's tale of the phone-maker, the burger flipper, the hairdresser and the tattoo artist highlights some problem with this definition, I'm all ears.

      Delete
    7. "Levine hopes to get away with it..."
      Certainly sounds like some kind of shady business to me.

      "Men are involuntarily unemployed If, in the event of a small rise in the price of wage-goods relatively to the money-wage, both the aggregate supply of labour willing to work for the current money-wage and the aggregate demand for it at that wage would be greater than the existing volume of employment."

      That's exactly what is going on. The indivisibility in the model is just a stand-in for sticky wages and prices - that's part of what makes it so simple. Suppose the phone guy is close to indifferent to producing a phone in exchange for a tattoo. But he doesn't want to produce, because the relative price of of tattoos in terms of phones is too high. If we think of everyone in this world as producing one unit of their good or service with one unit of labor, then the wage is one, in units of anything. Thus, if you drop the "wages" of everyone but the phone guy, in units of phones, by a small amount, then aggregate demand and aggregate supply of labor both increase. QED

      Delete
    8. Well score one for the optimistic kid: there had to be a pony in there somewhere. Thanks.

      Delete
  6. Dude, Krugman most assuredly did not refer to Levine as "incompetent."

    ReplyDelete
    Replies
    1. If you read very carefully, Steve, you will see that Krugman starts off talking about incompetence, compares that to ignorance and asserts that Levine's knowledge is a couple centuries out of date -- which is a charge of ignorance, I guess. So -- very technically -- Krugman didn't actually call him incompetent but he did everything but that...

      Delete
    2. Dudes, you don't have to read it all that carefully. Krug is clear that there is the D-K effect for the incompetent and that there should be a similar effect for the ignorant. Levine he clearly places with the latter. SW, you are no dummy -- but your anti-Krug mania has gotten to the point where it is interfering with your reading comprehension.

      Delete
    3. OK, I see what you mean. Strike the incompetence, but retain the ignorance. It's nasty anyway. My guess is Krugman never read Levine's piece, and just took DeLong's word for it. From my point of view, Levine has as at least as much status as an authority on Keynes and macroeconomics as Krugman does. For one thing, he wrote this paper:

      https://ideas.repec.org/a/bla/restud/v60y1993i4p865-88.html

      which is the better version of this:

      https://ideas.repec.org/a/oup/qjecon/v127y2012i3p1469-1513.html

      without the sticky prices.

      Delete
  7. The people in David Levine's story are using phones as a medium of exchange. So the "phone guy" is the Fed. And when the Fed stops producing money there's a recession.

    ReplyDelete
    Replies
    1. Seems like the best thing to do here is to engage. Otherwise the Keynesian types will always say that the non-Keynesian types are addressing the problems in their models rather than the problems in the actual economy.

      So the phone guy is the government. He can produce phones out of thin air. He trades the phone to a tattoo artist in exchange for a tattoo. (Tattoos are public goods, I guess.) Tattoo artist decides to hold the phone instead of trading it for a haircut (insufficient demand). Burger flipper and hairdresser now unemployed. What can the phone guy do? He could give another phone to the tattoo artist. Now the tattoo artist is willing to trade one of the phones for a haircut. Everyone is employed again. This is QE. There is no inflation because the tattoo artist still wants to hold that extra phone. There would be inflation if the tattoo guy decides he didn't want to hold the phone after all.

      Or the phone guy could use fiscal policy. Buy a haircut. Pay for it with a phone. Hairdresser can then use that phone to buy a burger. etc.

      Delete
    2. The model is just Kiyotaki and Wright (1989), but with 4 people instead of 3, and directed search. And we're looking at a commodity money equilibrium. I wouldn't get too involved in thinking about central banks and QE - the model is too simple for that, as we know from 26 years of work on this stuff.

      Delete
    3. OK, but no-one should expect Keynesians to be convinced by this example.It generates unemployment by a drop in the supply of phones rather than an increase in demand for phones. By not framing it as a demand shock, the example gives Keynesians another opportunity to point out the ignorance of economists who are too attached to their models to understand what the Keynesians are saying.

      Delete
    4. "OK, but no-one should expect Keynesians to be convinced by this example."

      That's not the audience for it. There is no example that would convince a diehard Keynesian of anything.

      "It generates unemployment by a drop in the supply of phones rather than an increase in demand for phones."

      That's easy to do, and a nicer example I think. Suppose the burger flipper has a demand shock - i.e. her utility from phones drops sufficiently that she won't accept a phone in exchange for a burger. Then we get the same result. Everyone is unemployed.

      Delete
    5. I should add that I think this has most, if not all, of the essential parts of the Keynesian narrative. Note in particular that the fact that the prices are fixed (everything's indivisible) is very important.

      Delete
    6. "Note in particular that the fact that the prices are fixed (everything's indivisible) is very important."

      Nope. New Keynesians might create an underemployment equilibrium in their models via their imperfect competition models (downward price/wage rigidity arises endogenously in a Blanchard/Mankiw type of model) but Keynes explicitly rejected the notion that more wage flexibility would get us out of the underemployment equilibrium via doing GE (lower wages lead to lower consumption as wage-earners spend large parts of their incomes, especially in a recessions which again leads to lower outpout; the labour market in a PE analysis works better but demand falls so the overall effect upon GDP is ambiguous).

      Of course it is, as you rightly point out, fairly irrelevant what Keynes thought (although it would not hurt if all the people who talked about Keynes would have actually read him). What this whole debate about price and wage stickiness shows is that the Old Keynesians were totally right: the causes of underemployment equilibria are multi-causal. It could be imperfect competition / price stickiness, it could be imperfect financial markets (Stiglitz and Greenwald wrote a bunch of great papers during the 90s about it), it could be coordination issues, etc.
      Now it would be great to create models in which we analyze all of these micro market failures together and actually see how they aggregate into the macro market failure ... but we are decades if not centuries from being able to do anything of this kind and for practical reasons it suffices to know that undermployment equilibria exist. And it suffices to know that monetary policy usually deals with them unless interest rates are close to zero, then fiscal policy is necessary.

      Delete
    7. I was agreeing with you until you got to the last paragraph. What I meant was that the fixed prices are important in the example, just as they are in a lot of what is called Keynesian economics. I agree that there are many frictions in the world that we need to think about to explain what is going on. I'm a little more positive about what we actually know. There has been important progress in thinking about credit market frictions, monetary economics, banking, information economics, I think. It puzzles me though, how you can say we know so little, but you're so sure about what is going on and how to fix it.

      Delete
    8. From a theoretical perspective I am no fan of Old Keynesian and mainstream New Keynesian models (obviously they work empirically) and I am all for microfounded macro. I' also claim that financial market imperfections are most likely the most important micro cause of the macro failure (I like to add though that there are aggregation issues and even without micro issues macro issues can arise; this is the whole point of the coordination failure literature).
      Everybody has his pet models but I like the abandoned research program by Stiglitz&Greenwald. Stiglitz is the natural expert on financial market imperfections due to asymmetric information and their models about credit market failures had an interesting reduced form, something similar to IS-LM.

      So I agree with you on a lot concerning the theoretical stuff. But, as Mankiw pointed out long ago, macroeconomists are not just scientists but also engineers. The Fed doesn't just use a large DSGE models but puts a structural models around it, they are agnostic about theoretical purity. If you cannot microfound something you just put in in there. You do after all want something remotely realistic.
      And in terms of policy recommendations nothing, except for the expectation stuff, has changed since 80 years. You moderate the business cycle with monetary policy except during the rare event of a liquidity trap, then and only then you use fiscal policy.

      It is not rocket science and economists who pretend that we have no solutions for managing the business cycle are lieing

      Delete
    9. "abandoned research program by Stiglitz&Greenwald..."

      That's not abandoned at all. You could say there is some of that in a lot of the work on credit markets and monetary economics that we do.

      "And in terms of policy recommendations nothing, except for the expectation stuff, has changed since 80 years."

      I wasn't around 80 years ago. But I think the FOMC discussion, for example, is much more rooted in science and academic work than it was in 1935. Holy cow.

      Delete
  8. I think it's an interesting piece, but I'm not sure I follow his reasoning.

    He describes a situation where "Keynesian" policies wouldn't work. Which is fine. I don't think even fundamentalist Keynesians would suggest such policies as a solution to every problem. But he then appears to take this to mean that Keynesian policies do not work in any circumstances. Which seems a bit of a leap.

    I don't know why he concludes this, but maybe this is something he has proved elsewhere using "complicated maths". I suspect though that it is just his belief that the sort of problems Keynesian policies are directed at do not arise.

    ReplyDelete
    Replies
    1. So much has been written about Keynes that you can probably find most of what is in Levine's piece elsewhere. But it's an interesting example I think. As well, I think he's correct to point out that the focus on Keynes and the General Theory in particular is mostly a waste of time. That's just an interesting piece of intellectual history now. We have better ways of thinking about our economic problems, given the progress we've made since 1936.

      Delete
  9. An interesting debate. Particularly that the substance of what it means to be a Keynesian is still so unsettled.

    Like others, I have my own take on Keynes, some of which are:
    1) The world was mired in a deep depression at the time Keynes published his theory, hence this was mainly a reaction to that event.
    2) The immediate crisis was that millions of people were out of work and suffering greatly, thus this was the prime problem to solve in short order.
    3) The world had a savings glut, if one can count the idled factories, mines, and facilities, and the build up of raw and intermediate materials as savings.
    4) The mismatch in the economy was due to the expectation of continued increases in consumption, which drove producers to expand capacity, while the demand actually dropped due to a lack of a concurrent rise in incomes. I.e. not enough of the value of the created goods and services was being directed to the wages of the workers who would be consumers.
    5) That the government could borrow (deficit spend) to stimulate demand of the consumers by giving the unemployed jobs and other wealth transfers, which would alleviate their immediate suffering. Stimulate demand for goods and services which were going unconsumed or not produced by existing means of production. Which would stimulate hiring in those factories and enterprises to meet the rising demand. Which would draw people off of the government support programs and jobs. Which would increase tax receipts of the government that would be used to pay back the borrowed money.

    Thoughts?

    ReplyDelete
    Replies
    1. I had a comment below, i don't know about your points 3-5, but I agree with 1-2, and because 1-2, some revisionists like Murray Rothbard defended the view that Keynes was not original, before he even started his "theory" he already knew he wanted fiscal spending to be the solution. His 'theory" was a collection of ideas already wandering around for many years or decades before. All he did was to put them together and call it a "general theory". The idea the rentiers have wealth that can be taxed and distributed to the benefit of society at large is very compelling, as we can see in this very comment session. It does not matter to keynesians in the past or now that observed fiscal multipliers are virtually ZERO. Keynes theory does not resist hard facts, period. As John Cochrane and others put it simply, keynesians of today are 80 years behind in terms of ideas, and at least 40 years after these very ideas failed in practice.

      Delete
    2. "All he did was to put them together and call it a "general theory"."

      Read the book. The title is a reference to Einstein. In physics Newtonian mechanics is the appropriate model most of the times, unless you are e.g. close to the velocity of light. Same in econ, the classical model is appropriate most of the times, unless you are in a recession.

      "It does not matter to keynesians in the past or now that observed fiscal multipliers are virtually ZERO."
      Nope. The last IMF paper I read on the issue estimated a multiplier above 1.

      "As John Cochrane and others put it simply, keynesians of today are 80 years behind in terms of ideas, and at least 40 years after these very ideas failed in practice."
      There are indeed some folks who are 80 years behind. It's the folks who deny that demand exists.

      Delete
    3. I read the book, i suggest you read chapter 24 and tell me if that does not look like the Venezuelan government. Most papers I read about fiscal multipliers, and actually found something, also show that most significant effects go away after you control for other exogenous shocks, most notably monetary policy. I don't deny AD can be managed, I just don't think "liquidity preference" explains AD shortfalls, and more importantly, that fical spending can solve it. If you care to read other authors of the time you will see that people were already concerned about inequality and pushing for fiscal spending, before keynes wrote about it, so, his ideas were far from original, they actually justified actions that a lot o people wanted done anyway. This hardly constitutes a "general theory" a name that his ideas don't deserve.

      Delete
  10. Steve, this is hoisted from the comments and it should really be the banner of this blog:
    If one has to spend much time (80 years and counting) contemplating "what is Keynesian economics", then the theory is so vague that its usefulness in terms of policy analysis is limited.

    There are good micro-founded models of "Keynesian" ideas like Roger Farmer, Peter Diamond, John Bryant, Angeletos et al. Why do these dinosaurs like Krugman and de Long keep wanting to regress to ISLM ?

    ReplyDelete
    Replies
    1. "Why do these dinosaurs like Krugman and de Long keep wanting to regress to ISLM ?"

      You could write a book about that.

      Delete
    2. "If one has to spend much time (80 years and counting) contemplating "what is Keynesian economics", then the theory is so vague that its usefulness in terms of policy analysis is limited."

      You clearly have not studied history, philosophy, political science, or such subjects where you have to read substantial books. People are still debating what Marx said 150 years later, Plato 2000 years later. It is because there is a lot to ponder in these books. I will find the quote, it has appeared here on blogs before, but in a direct reference to Keynes, Lucas has been circumspect and remarkably modest about what has been achieved in Macro-economics since the General Theory, unlike a few of the more gung-ho members of the Rational Expectations school. For sure, he is someone that understands the General Theory relatively well, you could say the same about people like Farmer.

      Delete
    3. I think the point is that, if Keynes were alive, he wouldn't understand what Farmer and Woodford are doing.

      Delete
    4. "Why do these dinosaurs like Krugman and de Long keep wanting to regress to ISLM ?"

      First, IS-LM models a short-run general equilibrium of the bonds, market and output market so it is the natural starting point for any short-run macro. It is the equivalent of not deriving a supply and demand function in micro via solving the objective functions of the firm and the household each and every time you wanna analyze what happens when there is a demand and supply shock.

      Second, IS-LM has recently performed empirically as well as the more complex models (Woodford's paper on the liquidity trap is the obvious benchmark). It has correctly predicted that in a liquidity trap massive expansionary monetary policy does not lead to increasing inflation and fiscal policy does not lead to rising interest rates.

      So I fail to see why using a simple model that has performed empirically virtually perfectly should be viewed as regress. Of course I know where you are coming from, you mainly care about internal and not external consistency... which is simply unscientific.

      Delete
    5. John D., you are back with a vengeance I see.

      "People are still debating what Marx said 150 years later, Plato 2000 years later"

      Yes, and that is the difference between humanities and sciences. Though, there are people who argue that even humanities should reform. Since you claim you like to read here is a nice essay by John Brockman:

      https://edge.org/conversation/the-new-humanists

      Delete
    6. "So I fail to see why using a simple model that has performed empirically virtually perfectly should be viewed as regress. Of course I know where you are coming from, you mainly care about internal and not external consistency... which is simply unscientific."

      Obviously you haven't thought seriously about this. You're just repeating what Krugman has told you. If what you say were true, then we would all be using IS/LM in our research and in our policy discussions. But we're not.

      Delete
    7. ISLM has performed well empirically. But then so has the theory that people don't float off the Earth into space because the devil is pulling them with all his might into the pit of hell. Its externally consistent, so let's go with it.

      Delete
    8. "you clearly have not studied blah blah blah." John D, please put on your tweed jacket, retire to the common room at Cambridge with your pipe and sherry and take the 2000 years or whatever you need to ponderously ponder Plato. Meantime let the scientists get on with the job of building models. When you have time to take a break from Das Kapital or whatever floats your boat, have a look at Angeletos, Acemoglu, Carvalho to see what a real macro model looks like.

      Delete
    9. "If what you say were true ..."

      Once again, when we are in a liquidity trap (the IS curve is to the right of LM) expansionary monetary policy does not increase output and thus not inflation (the part from inflation is not in IS-LM so you need to add AD-AS here). Expansionary fiscal policy does not lead to crowding out so there is no risk of a public finance crisis (the European issue is obviously different as monetary policy is made for several countries).

      So tell me in what way IS-LM is wrong. And tell me in what way whatever model you used (if you used any) to make your predictions a few years ago that high inflation is around the corner was right.

      Delete
    10. correction: IS is to the left of LM

      Delete
    11. What you wrote is based on a crucial but false assumption. That the IS-LM model is the only model that predicts that expansionary fiscal policy has an effect on output. This is not true of course. Many other models predict the exact same. Even real business cycle models (which some self-described Keynesians passionately dislike) predict that expansionary fiscal policy affects output and employment, and in some of them like Greenwood and Huffman (1991) business-cycle stabilization using fiscal policy is in fact welfare-improving. So the IS-LM is not the only model that makes such a prediction.

      The problem with IS-LM is not its predictions but its mechanics, how it arrives at those predictions. And the mechanics matter for two reasons: 1) Because the goal of policy-makers should be to enhance social welfare, not to maximize employment/output. Otherwise, the best policy would be to institute labor camps and enslave people to work. 2) Because there may be other policies that achieve the same outcome but in a more efficient way. For example, if the problem is a shortage of safe assets, a temporary tax cut financed by issuing short-term government bonds is potentially a more effective policy than increasing government spending, though the latter would have a similar impact.

      The IS-LM model can't tell you any of that because it lacks microfoundations so its impossible to perform welfare analysis to figure out if the benefit of combating the recession is worth the cost, is static so it doesn't take into account how people's expectations affect the efficacy of the policy, there are only two assets (currency and bonds), there is no asymmetric information so the lending and borrowing rate are the same, the list goes on. It has nothing to say about why the Ted spread increased after the collapse of Lehman and how that affected the credit channel even as the safe interest rate was falling. It has nothing to say about the effectiveness or not of so-called "operation twist" by the Fed. I can keep going, but I hope you get the picture.

      Delete
    12. "When you have time to take a break from Das Kapital or whatever floats your boat, have a look at Angeletos, Acemoglu, Carvalho to see what a real macro model looks like."

      I certainly hope they will not be discussing this boring garbage - or rational expectations models with sticky prices in 80 years. I will take a look, but I suspect that when it comes to the crunch this stuff is irrelevant, lacking in any any real world historical and analytical depth that is drawn from observations (quantitative and qualitative) that are gathered by people who are out on the field looking at what actually goes on and is useless. I suspect it is mostly artificial construct and fiction. Who is John D?

      Delete
    13. "Why do these dinosaurs like Krugman and de Long keep wanting to regress to ISLM ?"

      You could write a book about that."

      I guess his reasoning basically is that all models are going to be totally inadequate depictions of how the real world target system actually works. So if one simple wrong model gives exactly the same result as the complex wrong one, you go with the simple one. Efforts to more properly replicate real world target systems with a model lead to more complex models, or make existing models more complicated, and as one commentator has said earlier in relation to your comment on search models, end up as "nonsense on stilts," In the end you are actually distracted from properly looking at how a real world target system actually works which is to be found by concentrating on everything "knowable" - looking at as much evidence as you can, even if this does not lead to a neat, tidy or even systematic story.

      Delete
    14. "For example, if the problem is a shortage of safe assets, a temporary tax cut financed by issuing short-term government bonds is potentially a more effective policy than increasing government spending, though the latter would have a similar impact."

      First of all, I don't disagree that IS-LM is basic model which lacks many features. But I claim that it is a reduced form of many more complex models just like ordinary supply and demand curved are a reduced form of a household/Firm optimization problem. If you read about increased milk demand in China you can solve the complex micro model in half an hour or you can just quickly sketch a demand and supply curve and be fairly sure that the demand shock will lead to an increased price and an increased quantity.
      Of course a Woodford paper on the liquidity trap is theoretically superior to sketching IS-LM but if one does not care about economics as science but economics as engineering IS-LM suffices and all the New Keynesian literature does actually more harm than good (the bonds and money market are far clearer in IS-LM than in Woodfordian stuff).

      Now about shortage of safe assets, right now is not the cause of the problem but rather a symptom/ intermediate step:

      When we have a supply-side recession (I guess most recessions in this century will be supply-side due to resource shortages) demand side management is obviously poison.

      When we have a normal recession we let monetary policy do the trick

      When we have a balance sheet recession and a liquidity trap, i.e. interest rates are close to zero and a lot of private player gotta deleverage because an asset bubble burst, conventional monetary policy becomes ineffective (the CB can swap its safe asset money with risky assets of private players). Private Players deleverage, savings increase and we arrive at the situation you describe, a shortage of safe assets. QE as well as government spending of any kind do increase the average quality of the risk pool of private players so they should become more willing to tolerate risk. But this is just a side effect, the main goal is to jumpstart demand. You cannot do this via a tax reduction as private players save part of that very tax reduction so you gotta temporarily increase government spending.

      Please note that I would argue against deficit spending if we were not in a liquidity trap. There might be some half-wit Post-Keynesians (Richard Koo has provided great insights about balance sheet recessions but his tirades against the use of monetary policy are braindead) who think that fiscal policy should always be used and there might some some half-wit Neoclassicals who think that deficit spending should never ever be used ... but all the non-extremist economists who are less guided by politics and ideology and more by technical stuff first take a look at what kind of recession we are in before they provide solutions.

      Delete
    15. "In the end you are actually distracted from properly looking at how a real world target system actually works which is to be found by concentrating on everything "knowable" - looking at as much evidence as you can, even if this does not lead to a neat, tidy or even systematic story."

      Totally agree. Now of course we should nut shun complex models ... but what is the use of complex DSGE without finance/money in them? Sure, AFTER the crisis everybody tacks finance upon DSGE but the point is that the New Keynesian focus upon imperfect competition + menu costs (resp. Calvo pricing) has been utterly wrong.
      It is no coincidence that it were people like Roubini, i.e. guys who did NOT have a "unified theory" which is totally disconnected with the real world, who predicted the financial crisis.

      Delete
    16. The problem with IS-LM is not its predictions but its mechanics, how it arrives at those predictions. And the mechanics matter for two reasons: 1) Because the goal of policy-makers should be to enhance social welfare, not to maximize employment/output.

      Absolutely not. And this one of many reasons we need a political science education. It is up to the elected population in a democracy to determine an elected government's macro-economic policy agenda. It is not to up to you or policy authorities to decide what "maximises social welfare". The elected government's mandate may be to increase employment and in fact in many countries, that is precisely what it is at this very moment. To show how to get that the ISLM does that as effectively as any other model, and more simply. If you want to bring in discussions about social welfare, you are going to have to make assumptions that could involve the aggregation of individual utility functions (assuming anyway we know what those actually are), even when it is not being asked to be done, and you are unnecessarily making the problem more complicated, not clearing it up.

      Delete
    17. Once we re-liberate Macro from Micro, understanding that they are as distinct as Psychology and Sociology, and thereby not get tangled up in aggregation issues and Say's Law type fallacies - we can start returning to progress and relevance in Macro-economic theory.

      Delete
    18. "Once we re-liberate Macro from Micro..."

      Who is the "we," and what do you have planned? Does this involve the guillotine and a reign of terror, or what?

      Delete
    19. A few people will lose their heads, but at the end of it will be the emancipation and democratisation of the discipline.

      Delete
    20. Interesting democracy you have there.

      Delete
    21. And I'm wondering, once you kill all the heretics, what happens in your lovely democracy when another heretic comes along who wants to change things?

      Delete
  11. "David K. Levine, who apparently cannot even conceive of the possibility of a general deficiency of demand — which puts him a couple of centuries out of date."

    Krugman is spot on. Neoclassical "economists" like Levine or Williamson are demand deniers and thus 80+ years behind. Or if one argues from a microfounded macro perspective, demand deniers are in denial of the very basics of their discipline.
    The single, obvious reason for that is obviously political-ideological.

    ReplyDelete
  12. I think that most people today ignore that Keynes ideas were not original, they were being recycled in the 30's, because of the depression. According to Rothbard, all Keynes did was to take those ideas and put it in a somewhat "organized" and somewhat "obscure" way, so that people would spend a lot of effort trying to find out "what he really meant". But, if fact, Keynes already knew what he wanted before he even started, and that was find a way to tax the wealthy and justify spending. There is no value judgement here, I am not saying this was good or bad. I am just saying the Keynes "theory" did not come out of the hard facts, it is just something people wanted to do anyway.

    ReplyDelete
  13. Levine: "You say that we think that the "central problem of depression-prevention has been solved." Has it
    not? Are you forecasting that this recession will turn in to a depression?"

    This Levine guy is indeed a total hack, totally unaware that Krugman is quoting one of his own camp, Lucas.

    ReplyDelete
  14. I think Gertrude Stein had ISLM in mind when she said "There is no there, there".
    And how can we call Krugman and de Long macroeconomists ? Do they present research at macro conferences ? Do they publish in JPE. AER ? I don't think so. They are the equivalent of creation scientists. The world has moved on and they cant or wont keep up.

    ReplyDelete
    Replies
    1. Krugman published a macro paper in the QJE a couple of years ago. What have you done?

      Delete
  15. From DKL: "The thing is that the Keynesian prescription - spend more and don't worry about the bills coming due - sounds a bit too good to be true, almost like a perpetual motion machine."

    Come on, one could stop reading there. Keynesians believes that an increase in AD will increase output when unemployment is above Nairu, not under any conditions. By the same logic, if u < Nairu, one should decrease AD.

    For DKL, it is all about supply. If you give a phone to somebody you have to take it from somebody else. You are always on the frontier of the PPF. He does not even consider you could be within it for whatever reason (sticky prices frictions, animal spirits a la Farmer, whatever).

    I am not saying that keynesians are right, just saying that DKL fails to even engage with their argument. And supply-side people are left to explain how slumps such as the Great Recession could even occur. All of a sudden output goes down: what we were able to produce yesterday, we are not even close to be able to produce today. The PPF suddenly shifted inward. Did they bomb our cities?

    ReplyDelete
    Replies
    1. "Keynesians believes that an increase in AD will increase output when unemployment is above Nairu, not under any conditions. By the same logic, if u < Nairu, one should decrease AD."

      So you have a belief, and you say some words. Levine writes down an explicit little model to try to give those words some content. There is indeed something we can recognize as "full employment," there is a multiplier. I'm not sure how you're separating demand from supply. Levine could have introduced his "shock" at any stage in the chain, and the whole thing would break down. For example, as I told another commenter, if the last person in the chain suffers a "demand shock," i.e. they don't like their consumption good, to the point where it's not worthwhile to trade it for a phone, we get the same result. Everyone's unemployed.

      Delete
  16. The example chosen is telling. Unemployment arises because of a preference shock.* So obviously, it is just an unfortunate event. After all, unemployment is not a goal of itself as there are always some who are voluntarily unemployed (actually in DKL’s case, out of the labor force but never mind). With this set-up, obviously there is no free-lunch. If you want to make the fate of the “involuntarily unemployed” (I love the quotation marks here, clearly these are fictional people) better, then you have to stick it to someone else.

    DKL then discusses coordination failures and notes that monetary policy does not work either. No discussion is made of fiscal policy. Meanwhile, in the real world, when the Great Recession came, states slashed their operating budgets and laid off workers, engaging in pro-cyclical fiscal policy. I take it from DKL that this had no negative effect on output. After all, slashing budgets freed resources that should have shown up elsewhere. Yes, some may be worse off and others better off, but we were on the frontier before and we remained on the frontier later. That this may aggravated the coordination failure does not seem to be even a logical possibility according to DKL. It is kind of ruled out a priori.

    *A side note here. Usually if you try to explain economic events with something strange happening to preferences, you tend to be mocked. And that's one of the usual objections levied against the so-called behavioral people. If you change the utility function enough, the argument goes, you can explain almost anything. But here we are supposed to take seriously that one of the best toy model to explain unemployment is based on preferences. Interesting.

    ReplyDelete
  17. p.s.: I should have posted my post below here as it was written as a response to your comment. I wrongly posted it as a separate comment.

    ReplyDelete
    Replies
    1. "Unemployment arises because of a preference shock."

      Keynesians are always talking about demand shocks. How would you model that?

      Delete
    2. AD could shift to the left because households have a negative wealth shock and so they cut back on spending. Sufi and Mian find evidence of that. Heathcote and Perri have a recent paper (Wealth and Volatility) with sunspot equilibria. Or credit from banks may go down because there is higher uncertainty about the possibility of loans being repaid.

      The way DKL sets it up, it is clear that there is not going to be a free lunch because the phone guy now likes leisure more than before. Yes, DKL calls the phone guy stupid, but those are just his preferences which just happened to change. As economists, what else can we say?

      The whole DKL’s post has quite a few of these gems such as “So what if the reason the phone guy stopped producing phones and buying tattoos was because he wanted to spend his time creating the "next great thing" that will lead to world peace and prosperity in a few years time?” And indeed states like Nevada and Michigan have become tech hubs since 2007, right? Another nice one is thinking of fiscal policy as forcing people to work (or sending them to the front) rather than investing in the not so healthy U.S. infrastructure.

      Again, my point is not that keynesians are right, just that DKL does not even start engaging with them. Which is similar to Lucas laughing at the idea of multipliers. There are no conditions under which fiscal policy can ameliorate the cycle. This is held as a sacred belief. I think this has more to do, rather than with economic theory, with the view that some economists have of the role of government. And with the view that these economists have of other economists who are more favorable to fiscal policy. It is more a sociological than a scholarly divide. But this is just my speculation and would require further arguments.

      Delete
    3. "...because households have a negative wealth shock..."

      Which is what, exactly? Why does wealth go down?

      "Heathcote and Perri have a recent paper (Wealth and Volatility) with sunspot equilibria..."

      Yes, we've known about sunspot equilibria for some time. Levine mentions multiple equilibria, but seems dismissive that this provides a role for policy. Currently Keynesians don't seem interested in the idea. Why not?

      Delete
  18. http://mainlymacro.blogspot.com/2015/03/is-walrasian-auctioneer-microfounded.html

    Well, according to Simon Wren-Lewis, New Keynesians are doing much better than RBC people in 80's because RBC price taking isn't microfounded.

    ReplyDelete
    Replies
    1. Well, Wren-Lewis missed the point, didn't he? There actually is no auctioneer in a competitive equilibrium model. The Walrasian auctioneer is just part of a story someone thought up about how you could actually get to an equilibrium. In a competitive equilibrium, everyone is optimizing, and markets clear. That's all there is to it. Equilibrium is pretty fundamental to all of economics, and if you want to throw that out, we would have to all go find some other line of work.

      A basic NK model makes a small departure from Prescott's RBC model. NK has monopolistic competitors in goods markets. They indeed set prices, and for some of them it can be infinitely costly to do so. But otherwise all the markets are competitive, including the bond market. So I guess Wren-Lewis would think of this as a model with a Walrasian auctioneer and a Calvo fairy.

      Delete
  19. I am somehow confused here: you say that you agree with Delong that shortage of safe assets causes recessions. By walras's law framework (that Delong mentioned), that means deficient demand for goods and other assets. In such a case, government printing money can work to solve some problems, say Brad Delong and Nick Rowe. Given your blog posts, I do see that you argue that purchasing government bonds may actually be worsening the problem and so on, but what Levine seems to argue here doesn't seem to go that well with what I understand as your economics understanding.

    After all, what he criticizes is not just Keynes's General Theory (that's not even a real issue anyway) - Levine criticizes Keynesian economics practiced - fiscal stimulus, etc. So if your praise of Levine's criticism is about criticism of General Theory, OK. But I don't get why you would support Levine's criticism.

    ReplyDelete
    Replies
    1. Levine wrote down a little example, nothing more. If we want to think about asset shortages, we need to do more work. That's what some of my research is about. Though definitely you don't solve the problem by "printing money." Actually there are cases where, if "printing money" means a standard swap of reserves for government bonds, that's a bad thing to do.

      The point of Levine's example, as I would summarize it, is that Keynesian economics isn't so simple, if you're serious about it. It's hard to see why it leads to the simple fiscal policy solutions some people talk about.

      Delete