Thursday, January 30, 2014

Question of the Day

What does this mean?
And not really an accident either. If you take secular stagnation seriously, as you should, then we have a chronic problem of too much saving chasing too few good investment opportunities, which means that you only feel prosperous when money thinks it has found more good places to go than it really has — and soon enough figures that out, with nasty effects.
Turn that into a coherent economic argument, and I'll give you a prize.

45 comments:

  1. It means that this Art Deco Inlay was put in upside-down:

    http://www.agilitynut.com/08/8/ordbank.jpg

    That is, "Spend it all before it's just...just...GONE!"

    ReplyDelete
  2. It means that someone should lay off the nargile!

    ReplyDelete
  3. Actually, if I may take a second shot, it is an attempt to conduct a partial (dis)equilibrium analysis with some arbitrage and irrational exuberance added.

    ReplyDelete
  4. PK seems to be saying that secular stagnation can be obscured for a while by an investment bubble. When the bubble bursts and the lack of good investment opportunities is thus revealed, the resulting recession and the sudden demand for safe assets may have very bad consequences for the economy.

    This is not PK's best writing and the argument may be wrong, but the quoted fragment does not seem incomprehensible, especially in context of the whole PK post. I guess if you do not believe in bubbles or believe they are always rational in some sense then the quotation would seem to you to be devoid of content. But that is a personal question.

    ReplyDelete
    Replies
    1. I'm not sure where the bubble comes into it. In the PK world, a bubble is inefficient, so when it's gone, you would think that would be efficient.

      Delete
    2. I don't think PK sees a bubble as inefficient. He sees the problem as a sort of long-run lack of demand, and so the economy is constantly looking to create bubbles to generate that demand.

      Think of bubbles in Farhi-Tirole for instance. Then combine with some sort of New Keynesian model?

      Delete
    3. Here's what Krugman says: "...what do we mean when we talk about bubbles? Basically, I’d argue, we mean that people are basing their decisions on beliefs about the future that are based on recent experience but can’t be fulfilled..."

      So he says it's driven by irrationality, and therefore inefficient (that's a little slippery of course, as it's now hard to define efficiency with irrational people in the world). Generally, among the irrational bubble people - Shiller for example - bubbles are bad. So having no bubbles must be better than having them. But if we have no bubbles we have no demand, so that's bad? Seems contradictory.

      Delete
    4. I am not sure about what you say about irrational bubbles being bad. It all depends on whether the economy is dynamically consistent. If we have secular stagnation...

      Delete
    5. I mean, dynamically efficient.

      Delete
  5. The last part seems fine. There are models whereby an investor may be overconfident about a noisy signal they receive. There could be agency problems between investors and money managers, etc.

    I don't see the link to the first part. Why would a 'secular stagnation' make these problems worse? I guess you could argue that the agency, overconfidence, etc. problems are the same, but the lack of productive investments amplify the effects.

    ReplyDelete
  6. hmm. I used your textbook when I was a primitive undergrad. Then did all the grown men stuff in graduate school, etc. Now I'm taking some time off in an investment bank. A sordid place with one advantage: one learns to distinguish reality based economics from the other, parallel universe-based one. I know you find the man's arguments vile and ridiculous, but the argument he makes in the comment is a stylised version of events, without needing to make a claim for monopoly over truth. If you go and see what the traders are doing, what news they respond to and why, you'll get the point. I'm saying this without any hint of malice. I read your recent QE paper and talked it over at the pub with former phd mates. I have to say, sir, slap yourself in the face, stop playing the piano with those equations, go out and see some stuff, then go back to the writing. Good research ideas come from looking at the world, not only from pouring over that overtly-inbred profession of ours. If not, do some econometrics, that's a)useful to others, since you make tools for them b) good for you, a bit like medidating.

    ReplyDelete
    Replies
    1. "...a stylised version of events, without needing to make a claim for monopoly over truth."

      Another prize for explaining that one.

      Delete
    2. That line seemed perfectly understandable. It's one person's summary of events and what caused them. Not a fully worked out rigorous peer-reviewed theory, or by any means the final word, but good enough for a blog.

      Delete
    3. Same troll, different day!

      Delete
    4. "...but good enough for a blog..."

      As I've stated before, it looks like we need higher standards for economics blogging.

      Delete
    5. If we did, you would arguably be out of the economics blogging business, although you might be able to eke out a living in the Unjustified Snark market segment.

      Delete
  7. I *think* it means: "The natural rate of interest is below the growth rate of GDP, so the economy is dynamically inefficient, so that a Ponzi scheme can, in principle, be sustainable, like in Samuelson 58, but we don't know *which* of the many possible Ponzi schemes will survive."

    ReplyDelete
    Replies
    1. "The natural rate of interest..."

      Define it.

      Delete
    2. In this context (only): "what the equilibrium (real) rate of interest would be if there were no Ponzi scheme".

      Delete
    3. For example: what the equilibrium interest rate would be in Samuelson 58 if each generation thought that the next generation would not accept its "money".

      Delete
    4. OLG model. Agents produce when young but cannot produce when old. The only storable goods are intrinsically worthless shells. It is costly to collect shells from the beach and store them. There is a large variety of different coloured shells. Each period there is a 10% probability of a sunspot. When there is a sunspot, the old shells become worthless, and a new colour of shell becomes valuable.

      I *think* that model could be made to work, but I'm not sure.

      Delete
    5. Yep. If there's a red sunspot, the young spend part of their time picking up all the red shells from the beach, and the rest of their time producing the consumption good. (The old are too decrepit to either produce the consumption good or pick shells).

      Next period, if there is no sunspot, the old sell their red shells to the next generation of young in return for consumption. Nobody picks up green shells, until there is a green sunspot, because the costs of picking and storing green shells are too high relative to the low probability of a green sunspot, because there are lots of different colours of shells and sunspots.

      Delete
    6. Nick,

      Actually, that's not bad. Jumping from one bubble to the next. You should tell Krugman, and see what he says.

      Delete
    7. I was thinking that the allocation is inefficient, alright, but what about the savings and investment part of the story? If time spent on the beach picking up shells is measured as investment, we have too much investment, not too little. Even better, extend the model a bit, call the shells houses (with different colors) and suppose people get a utility flow from them.

      Delete
    8. Thanks Steve! But yes, on second thoughts, I'm not sure if that is the sort of thing Paul had in mind. My agents (I think) are rational. I think (not sure) Paul is talking about irrational beliefs?

      Good point about picking up shells being measured as investment.

      It's not so much houses (which depreciate) but land. Why can't useful land replace shells? Dunno.

      Delete
    9. What exactly is a sunspot? (I know what a seashell is).

      Delete
    10. JP: it's a device some economists use to jump the economy from one equilibrium to another when there are multiple equilibria. It is some random event that everyone observes (and it is common knowledge that everyone observes it) but which has no intrinsic relevance.

      The Diamond Dybvig model of bank runs is a sunspot model. When a sunspot happens, there is a run on the bank.

      (Originally, in Jevons, sunspots were very different, because they do in fact affect the climate, and so were a real exogenous shock to supply of corn.)

      Delete
  8. Stephen, In response to your earlier (6:12) response to my comment:
    1. Thanks for it
    2. Krugman might agree that "after the bubble is gone that would be efficient", in the sense of being a more "genuine" (I know, define genuine) equilibrium. But that equilibrium might reflect the sad revealed reality that we are in a secular stagnation. The veil would be lifted from the sad reality. Anyway, I would bet a dollar (my personal maximum) that his deluded money was a reference to the bubbles pre GFC. Also, that is consistent with his writin just after the Summers talk late last year.

    ReplyDelete
    Replies
    1. "...in a secular stagnation. "

      Which is what, exactly?

      Delete
  9. This is a ridiculous post as Krugman's arguments about secular stagnation are crystal clear (unless one quotes him out of context) and as secular stagnation is anything but a new idea. Keynes thought that in the near future the Wicksellian natural rate of interest (i.e. the interest rate at which there would be full employment) will be low/negative for a long time ... and thankfully his prediction was wrong. Recently some folks, not just Larry Summers, have played around with the idea. I like to point out a good post by Fata (http://fatasmihov.blogspot.de/2013/11/saving-glut-or-investment-dearth.html) as it contains some stylized facts.

    While I am anything but fully convinced of the secular stagnation hypothesis I appreciate that a "it is all a short run problem - so deficit spending!" guy like Krugman considers that it might actually, at least to some degree, be long run problem.

    ReplyDelete
    Replies
    1. I think the idea is in Alvin Hansen's work, and not in Keynes. I could be wrong, but I think Hansen just used words to describe what it is, and the idea seems to focus on demographics and territorial acquistion. So, now Krugman is discussing the idea, but he seems to be talking about something different. I'm trying to figure out what it is.

      Delete
    2. Stephen

      I think your quest to understand what he is saying will ultimately be futile. But you probably knew that already, given that you promised a million dollar prize.

      The problem with trying to understand PK is that he uses a lot of fancy terms, none of which can be measured. In particular, this means that there is no *rule* or *definition* that he can provide that will tell you what each term means in terms of the data. What is "secular stagnation"? When are we in a "bubble"? Ex post rationalisations are not useful. This is why Fama doesn't take Shiller's protestations of "bubbles" seriously.

      It seems that only a select few like Shiller, PK, etc, can look at current data, and diagnose things exactly. Mind you, they won't (can't?) give you a policy rule to follow, precisely because there are so many undefined terms and things that any sentient being cannot measure.

      This also points towards why SW (and the profession at large) likes models with microfoundations. All terms are well defined.

      Delete
  10. Land and demographics doesn't sound like secular stagnation to me. This is basic stuff (unlike you I do not dislike Krugman but I'd say it is clear that he mostly writes about fairly basic stuff), secular stagnation simply means that the savings curve shifts outwards while the investment curve shifts inwards.

    So you have a simple "theory" that matches three stylized facts: low interest rates / low, probably negative natural rate of interest + the increase of total saving and the decline of total investment in the first world.
    And every "secular stagnationist" neglects to look at development countries where investment/GDP has slightly increased during the last years.

    I don't wanna imply that it is totally useless to speculate about the lack of large technological breakthroughs in the near future or the decline of the rate of technological progress (here you could directly test the hypothesis but then again I think that if you do not just measure the Solow residual but use something more elaborate there might be ambiguous results) but it is fairly blind guesswork. And the other stuff behind this "theory" is just ordinary macro: looking at aggregate data like current accounts and capital movements over the worlds.

    ReplyDelete
    Replies
    1. "So you have a simple "theory" that matches three stylized facts: low interest rates / low, probably negative natural rate of interest + the increase of total saving and the decline of total investment in the first world."

      1. A low "natural rate of interest" is not a fact, as you need a theory to define "natural rate."
      2. In the world as a whole, savings=investment.

      Delete
    2. "In the world as a whole, savings=investment."

      Right, and it seems that, for this reason, PK is adding a qualifier and saying: saving > good investment. But what the heck does "good" mean?

      Delete
    3. I said "low, probably negative natural rate of interest" and never claimed that I am sure about whether the natural rate of interest is negative. I also already said what the natural rate of interest is: "the interest rate at which there would be full employment". How come that a guy without a PhD knows this basic stuff while a professor does not? Willful ignorance?

      What Krugman has in mind is a picture in which S and I intersect at an interest rate of zero with multiple equilibria. Economically this means that a large chunk of savings flows into money which is not invested aka liquidity trap.

      As I said, Krugman is doing basic stuff. Now you can disagree with it (like I partially do) or do something less basic and more elaborate. But to pretend to not understand is slightly pathetic ... and if you do not feign to not understand it but actually do not understand it this is embarrassing.

      Seriously, if tenured folks do not understand the 101s of a liquidity trap while we are actually in one for years we truly do live in a dark dge of macro.

      Delete
    4. The other anonymousJanuary 31, 2014 at 3:34 PM

      "What Krugman has in mind is a picture in which S and I intersect at an interest rate of zero with multiple equilibria"

      Then it is still true that S = I (with sticky prices because output adjusts). Here is Krugman:
      "too much saving chasing too few good investment opportunities".
      Does this sound to you like S = I? Are you sure you understand what he is saying?

      Delete
  11. A possible model of secular stagnation:
    http://informationtransfereconomics.blogspot.com/2013/11/secular-stagnation-and-eu.html

    Concept: Diminishing marginal utility of adding money to the economy to capture AD and translate it into NGDP. Based on information theory.

    ReplyDelete
  12. PK is just saying what Delong wrote last month: the required real yield on private investments low enough to induce enough investment to balance savings at full employment is too low a yield to properly discourage bubbles and overleverage.

    ReplyDelete
    Replies
    1. So, the real rate should be low, but it shouldn't be low, and the only solution is more G?

      Delete
    2. Even if the secular stagnation issue hypothesis is wrong it is obvious that there are ample of public investments which lead to a return that is far above the financing cost (low yields on government bonds plus the efficiency losses due to distortionary taxation), e.g. infrastructure.

      But the political problem is neoliberalism aka lemon socialism, i.e. private companies use the state to create returns: cost-plus contracts during the Iraq War, more explicit subsidies of all kinds, private prisons and school and so on. Rent seeking gone crazy. By the way, this is not a US-only issue, it starts to happen on the other side of the big pond as well.

      So in my opinion and contrary to Krugman and other Old Keynesians the problem is not that G is too small. The problem is that G includes too little rational (return > costs) public investment.

      So no matter whether you approach the subject from my rent-seeking angle, the secular stagnation or e.g. the climate change angle (if we had a worldwide Pigouvian tax on CO2 there would be many more new investment opportunities), it is clear that there is a problem of too little investment (in the first world). But Krugman errs in two respects: this is a long-run issue and it can be solved not merely via more public investment.

      Delete
    3. Careful when you go outside, your tinfoil hat might singe your hair.

      Delete
  13. If most countries preferred to save more, by imposing austerity, isn't world saving higher than world investment? Why should I=S for the whole world?

    ReplyDelete
    Replies
    1. It is an identity. It must hold by definition. For example, for a closed economy (e.g. the world economy if there is not interplanetary trade) saving is defined as the total income (Y) not spent by households on consumption (C) or by the government on various goods and services (G). In math: S = Y - C - G. Also, world income (world GDP) is measured as the sum of private consumption expenditure, private expenditure on investment goods, and expenditure on goods and services by households. In math: Y = C + I +G. If you solve for investment you get: I = Y - C - G. Notice that the right-hand side is the same as the right-hand side in the saving equation. Therefore, it must be true that I = S. There are different ways in which this equality can be maintained. 1) If saving increases the real interest rate will decline so investment will rise by an equal amount. 2) Suppose this cannot happen. Then prices may decline so that even though nominal spending is lower, in real terms nothing will change. 3) If this is not possible either, then the increase in saving will result in a decrease in production and therefore income (Y). Even though people want to save more, they cannot because their income has declined. In fact, total saving may actually fall (the so-called paradox of thrift, popularized by Keynes). So, you see, for the world as a whole it should always be the case that S = I.

      Delete
  14. CA,

    Thank you very much for the explanation. As a beginner economics student I was wondering that excess saving relative to investment was a possibility.

    ReplyDelete