Tuesday, May 31, 2011

Thoughts on PSST - and then some thoughts from Brad DeLong on it

There's something very appealing in Arnold Kling's posts on "PSST", which is basically just a rebranding of a conception of the market that we've had for a very long time. PSST is essentially just an extended division of labor, after all - specialization and exchange. Modern market economies are nothing if not groups of people specializing and exchanging.

That's Adam Smith, and so it's no wonder we all find something appealing about it.

The problem comes in when you start to say that because complex specialization and exchange is what the market economy is, it follows that when market economies break it must be because something was unsustainable about the nature of the specialized exchange. This is an odd route to take, whether it is via Arnold Kling (who talks in terms of major "regime changes" - displacement of agriculture, IT revolution, etc.) or via Hayek (artifically elongated capital structure gets revealed to be unsustainable). The romance and enticement of organic, complex systems can mislead people. When other complex, emergent systems like organisms or ecosystems get sick, we don't just assume that the complexity that used to work no longer works anymore. We assume something specific went wrong with it. So why don't we all assume that with markets? I'm not sure.

Brad DeLong isn't sure either. He writes: "When you ask believers in "recalculation" what pattern of production and trade proved to be unsustainable in 2007, they answer: "building so many houses." When you ask believers why the market economy has been unable to sort out this problem in three years, they answer with nothing--silence. When you say that OK, there were $300 billion of excess houses at the start of 2007 but now construction has been so depressed for so long that there are $1 trillion fewer of houses than trend and why isn't the 2007 pattern of production and trade sustainable again, they answer once again with nothing--silence. That annoys me."

Markets work. What a lot of the PSST talk amounts to is the suggestion that not only can markets spontaneously stop working, but they have a very hard time picking themselves up when they trip after going down an unsustainable path. Hayek is slightly better on this count than Kling - Hayek actually offers a mechanism: artificially low interest rates that allow an elongated capital structure to actually build up before it is revealed to be a house of cards. That's somewhat better than Kling's story which is basically the old technological unemployment case.

We need something better than that PSST story. We know why markets work so well. We know how markets correct themselves. So when markets break you have to have a pretty specific mechanism in mind not only for why they break so sharply, but also why (in this case, as in the 30s) they don't always seem to fix themselves. The mechanism I have in mind is that a popped bubble (which is psychological and institutional - and also well understood - not vague hand waving about "patterns of exchange breaking down") increases demand for safe and liquid assets, and exogenously increases savings rates for all actors at once. Traditional monetary accomodation is limited by a zero lower bound on the nominal interest rate. Interest rates that can't adjust to the negative-whatever-it-is-according-to-the-Taylor-Rule percent they need to be depress investment because the marginal efficiency of capital consistent with full employment is negative. Massive underutilization of resources means there is little demand-pull on inflation - which is one thing that could actually help the interest rate situation (or sticky wages - if you're the sort of person that worries about that).

That's the mechanism I have in mind. It seems to me to fit the facts better than any other mechanism I've ever been offered. It also tells me why we had a recovery in 1920-21, why we didn't for a while in 1929, why we had a reasonably quick recovery for most of the post-war period, and why Japan didn't have a quick recovery in the 1990s and why we aren't having one now. When you have an explanation that fits with a broad set of data like that, that's a useful theory.

Hayek's elongation of the capital structure sounds plausible to me, and certainly interesting. Jonathan Catalan makes it more plausible by regularly reminding us that ABCT can result in a general depression - it doesn't have to have the naive "internet Austrian" impact of only being concentrated in certain sectors. But it still leaves open the question: why are we still in this pit after three years? And for that matter - why was 1921 and 1981 different from 1929 and 2008? ABCT can't answer that very well. DeLong, Keynes, Friedman, Hicks, and Krugman can answer that question.

I would add to Brad DeLong's list of absurdities that nobody can provide a reasonable answer to this question of "regime uncertainty". The argument goes that we haven't recovered because investors are tremendously uncertain. But just as Brad notes "If you want to argue that there is a disruption of patterns of sustainable specialization and trade, you need to point to such a disruption right now that is large enough to produce an 8% shortfall in spending", I could also say that if you want to argue that regime uncertainty has kept us in the doldrums for three years you need to point to something people are uncertain about that is large enough to produce an 8% shortfull in spending. Health reform? That was the schtick originally - they said that "regime uncertainty" about health reform kept investors from investing. But now it's done and passed and there's no appreciable difference in the macroeconomic situation. They try and argue that the details and regulations are still being worked out for health reform, and that's true - but are they really convinced by that? Is that really sufficient to keep us in depression? Our health reform is modest by the standards of many of our peers, and they didn't seem to get caught in a depression because of their health care system. And are any businesses actually saying that's why they're not investing? Is there any sign at all that investors are worried about tax increases or health reform? No! There's no evidence of this whatsoever. Time after time when businesses are surveyed they cite demand as the problem - not concerns about government overreach or taxes. This whole "regime uncertainty" argument doesn't even pass the smell test in the first place, and on top of that there is no empirical evidence for it.

There's a lot of bad economics out there, despite the fact that we have a pretty decent sense of what's going on. Even if we have some disagreements about what to expect from the slightly less predictable fiscal policy and whether we've fully used up monetary policy options, there is still a broad agreement about the mechanism that is causing this depression. And as Brad has pointed out in the past, it's an agreement that unites Keynesians, monetarists, and many others - back to both Jean Baptiste Say and Malthus.

20 comments:

  1. I don't personally like Kling's recalculation story because it doesn't explain WHY everyone was hypnotized - it just asserts so, and then says "Aha! Now we need time to get back to reality!"

    I should hope the Mises/Hayek explanation is more plausible than that. Mises and Hayek give extraordinary detail into how it is that business cycles are created. Why are we still in this mess 3 years later? That probably has a lot to do with regime uncertainty, moral hazard, a continued policy of 0% interest rates, liquidity traps (which intertwines with regime uncertainty) and high tax rates.

    I don't believe ABCT has an answer for why this particular recession is still going with high unemployment, inflation appearing likely, etc. But it certainly can tell us how we got here.

    ReplyDelete
  2. DK: Until today I wasn't aware of this. My first impressions is that it does seem like it could be useful in trying to better grasp recalculation, but at the same time it seems like it could end up being a complex model. And when you said it's another case of tech unemployment(reminds me of rbct) are strictly referring to regime uncertainty? I'm not exactly familiar with PSST so there are some things still confusing me.

    Matt: A year ago I would have said regime uncertainty was playing into Mattheus because of the elections, bush tax cut expiration, and healthcare, but a years passed and the situation has hardly improved. I also don't see why you think low interest rates(implying raising them would solve the trick), liquidity traps, and [low] tax rates have to do with the situation.

    ReplyDelete
  3. No - regime uncertainty and PSST are two different things, and technological unemployment was a reference to PSST.

    I'm not entirely unsympathetic to technological unemployment arguments - new or old - and Arnold Kling doesn't provide the sort of naive technological unemployment arguments that assume increased efficiency does nothing to increase demand. We have decent evidence of skills biased technological change in recent decades, and throughout history it clearly had a displacing - as well as a growth - effect. I could believe that innovation seeking out greater efficiency exascerbates job-finding in recessions. I don't think there's a lot of evidence that it's a cause of recessions.

    ReplyDelete
  4. I agree with other posters that that ABCT mechanism is a better explanation than Kling's.

    Remember that not every Austrian is a Rothbardian. Probably most are from the monetary disequilibrium/fractional reserve free banking side. Anyway, I am.

    To us the recession was first caused by the ABCT mechanism. It was then exacerbated by the rise in demand for money that wasn't immediately served by the central banks, and exacerbated further by the regime uncertainty. That's our explanation of why the past 3 years have been such a mess.

    The idea that "regime uncertainty" is opposed to demand explanations is wrong. What we are concerned about here is an increase in demand for safe assets that can't be reproduced in the short term, and in the increase in demand for money. I think that is what Keynesians are concerned about too when discussing the liquidity trap. I agree with Keynesians that this increase in demand for safe assets reduces the demand for risky assets. And, investment (that is the GDP flow "investment") is more involved with risky assets than with safe one. Hence, a fall in demand for those assets causes a fall in investment spending and in aggregate demand.

    But, I fail to see why we can't create more money. Monetizing existing assets is easy if the banking system is allowed to do it.

    ReplyDelete
  5. A year ago I would have said regime uncertainty was playing into Mattheus because of the elections, bush tax cut expiration, and healthcare, but a years passed and the situation has hardly improved.

    It's not so much regime uncertainty I suppose, but market uncertainty. Banks are fearful of opening lending channels because private purchasing power is still weak. But the uncertainty in the market is directly related to the dangerous policies the Bush and Obama administrations followed. General 'uncertainty' might have been a better word to use.

    I also don't see why you think low interest rates(implying raising them would solve the trick), liquidity traps, and [low] tax rates have to do with the situation.

    Low interest rates set by the Fed are anemic to recovery because it continually misleads investors into creating production processes different than what consumers desire. This sets us up for bubble after bubble.

    If the problem is too low rates, I don't want too high rates either. It's not a too low or to high question. It's a control question. A central authority in charge of the total supply of money cannot possibly set the interest rate in a way to rationally induce investment opportunities to reflect consumer desire. Only a free market in money production can do that. Raising rates would encourage a lot more savings, a lot more debt default, and a sharp reduction in the supply of money. I personally think these are good things - but raising rates isn't a panacea.

    The liquidity traps are a product of low interest rates (easy financing) and uncertainty in the market. Then there's the fact that these banks are essentially paid to sit on this money. So it's no wonder that we have a huge glut of money in these large banks, that are refusing to lend it out, and aren't suffering for their inaction.

    High tax rates discourage production and savings. It should be fairly obvious that high tax rates are antithetical to a quick recovery.

    ReplyDelete
  6. But, I fail to see why we can't create more money. Monetizing existing assets is easy if the banking system is allowed to do it.

    Of course it is. But do you understand the consequences of monetization?

    ReplyDelete
  7. DeLong is wrong. There have been plenty of arguments as to why the retrenchment has been so long and so deep. Now you can argue that these arguments are incorrect, but silence is not the right description to say the least.

    "But now it's done and passed and there's no appreciable difference in the macroeconomic situation. ... Is that really sufficient to keep us in depression?"

    Merely because some legislation was passed does not mean that it is "done." If it passes constitutional muster (and hopefully the 6th Cir. will overturn it in the near future) even then it will not be "done" - in the monstrous administrative state that we live in huge bills of social engineering like the health care law take years to get "done." In other words, there remains all manner of uncertainty with regards the law.

    Since when were we in a "depression?" And yes, the moronic health care legislation explains much of why we remain stuck where we are.

    "Our health reform is modest by the standards of many of our peers, and they didn't seem to get caught in a depression because of their health care system."

    No, they have permanent high levels of unemployment. Anyway, this is sort of apples and oranges; different nations react differently to legislation.

    As for what the surveys say:

    "Seventy-three percent of all owners said the current period was NOT a good time to expand. Of those, 69 percent cited the poor economy as the reason, but 18 percent blamed the political environment, unchanged from July.

    'If the poor political environment is top of mind for nearly 1/5th of those opposed to expanding, it is likely second on the list for most of the others,' said Dunkelberg."

    There is a significant amount of evidence that "regime uncertainty" plays a large enough role to be something of a concern (at the very least).

    http://www.nfib.com/press-media/press-media-item?cmsid=54603

    ReplyDelete
  8. Mattheus,

    I'd say I do understand what monetization is about.

    On of the main problems we had during the bust was an excess demand for money. That could have been solved by increasing the supply, and there was some increase which helped to some degree, but not enough. The interest rate is a red herring, as Scott Sumner has repeatedly explained.

    ReplyDelete
  9. re: "The interest rate is a red herring, as Scott Sumner has repeatedly explained."

    I think it's better to say clearly that the interest rate is a red herring when it comes to diagnosing whether monetary policy is tight or loose. This is what Sumner says, and I imagine what you are saying too. Monetary policy is tight right now, is the point.

    However - the interest rate is not just some funny little price that confuses people about the tightness or looseness of monetary policy. It isn't a red herring at all when you have to pay or when you earn interest, and when you have to compare it to other rates of return.

    ReplyDelete
  10. You know I'm an ABCT guy, but let me play devil's advocate for a moment with PSST.

    Stuart Kauffman points out that downturns follow a power law distribution. He also points out that waves of creative destruction do so as well. If one could correlate the two, that would be a powerful argument for PSST. It is not entirely unlikely that a new technology could come along that, while creating incredible opportunities for great wealth, also destroys great swaths of already-existing industry. Such a disruption would of course end up being a recession. The time needed to retrain all those people would delay recovery, particularly if there was a domino effect to the collapse. All of these are distinctly possible.

    (BTW, the organism metaphor doesn't work -- the economy is more like an ecosystem, which can be utterly transformed by its own internal dynamics, or by external forces.)

    Now, consider this possibility. It may be closer to the truth that PSST is a relatively slow-moving process most of the time. The extinctions of firms take place over time, people have time to adjust, etc. However, a shock to the system may trigger sudden, large avalanches. Enter ABCT.

    Thus, ABCT may explain the bubble leading to the bust, while PSST explains in part the longevity. I would in fact include such things as regime uncertainty. Health insurance reform is pretty spread out, so the uncertainty is spread out (also, I suspect the Europeans weren't so foolish as to introduce their reforms during a recession, either -- that matters as to the economic response to such a reform). The Cash for Clunkers was a disaster, and is now resulting in the highest prices for used cars ever -- something which harms the poor the most, who are now less able to move around, including getting to jobs (does this contribute even a small amount to unemployment? perhaps). Extending unemployment benefits is known to delay people finding employment, and there have been several extensions. And a lot of people are freaked out about the federal government's spending and are unsure what it means for them, including whether it means higher taxes are coming soon. In other words, it's not just one thing, but a series of actions by our federal government that keeps people guessing as to what to do. People need to know that the rules are fairly stable, and they have no idea what the rules may be tomorrow with the Obama government.

    These same factors were present during the Great Depression as well, and contributed to its extension. There was a major shift from an agrarian to an industrial economy, and that was massively disrupting. If a new species invades an established ecosystem, there can be widespread species destruction until the ecosystem adjusts to the new species' presence. New species can and will emerge in response to the presence of the invasive species, etc. in the same way as occurs in an economy. The initial round of destruction looks horrific, but it's simply part of the evolutionary process. If the species emerges naturally, we think little of it; if we introduce the species ourselves (equivalent to artificially low interest rates creating a bubble), we are deeply concerned about the destruction we wrought. (Sadly, this is a limited metaphor only because we don't seem to be at all concerned about the fact we have created the bubble through artificially low interest rates, and thus keep introducing new species to fight the old one, resulting in more disruption, etc.)

    ReplyDelete
  11. One thing I'm still a little unclear on is exactly what "P" it is that is being disrupted. The idea isn't non-sensical, it just seems like a "this could work" response to the question without any really good reason for accepting it.

    If you look what's happening in financial markets, with interest rates, on balance sheets, etc. - you do see evidence for the sort of mechanism that DeLong has been describing. It's not so much that I'm absolutely ruling PSST out as it is that I'm not sure if I should be convinced.

    I am really not moved on the regime uncertainty argument. You seem to expect the general population to react to Obama in the same way that you react to Obama. A large portion of the population sees an end to regime uncertainty with the start of the Obama administration. You're listing things that many people see as stablizing, rational decisions or minor annoyances that haven't had all that much effect, despite how ill advised they were (i.e. Cash for Clunkers). You're essentially trying to demonstrate regime uncertainty by listing things that you're uncertain about without providing any reason to think the preponderance of people agree with you. And once again I will stress - when you survey businesses they do not cite these problems as their concern, and as Gary demonstrates the NFIB has been bending over backwards to explain away their own data because it doesn't say what they want it to say. Might many people rank taxes and regulation second as a concern? Sure. But there is no evidence of a sharp increase in it (which you would still expect if it was a major factor), while there is evidence of an increase in concerns about demand conditions.

    The creative destruction you describe is obviously part of an emergent market order. This is not the question. The question is - does it explain recessions? The shift from an agrarian to an industrial economy was gradual and there was nothing about a monetary or financial crisis that would have sped that up (farmers and businesses alike are faced with credit constraints and if anything the mechanization of agriculture would be delayed by a lack of credit, not accelerated!). Certainly creative destruction happens and certainly markets adapt to the consequences. This is the whole point of my critique of PSST. As I said above, "markets work". If you're going to try to explain discontinuous breaks in the market order it seems odd to cite natural market adjustments as a cause.

    ReplyDelete
  12. Daniel,

    Sumner and the Quasi-Monetarists say that the interest rate doesn't describe the stance of monetary policy. That's right but there's more to it than that.

    In my opinion the demand for money is the crucial issue that turns real shocks (such as from ABCT) into longer lasting recessions. The problem is that if there are sticky prices and an excess demand for money then prices and output will fall. The interest rate doesn't have to be at any particular level for money supply and demand to meet. If the interest rate is zero that doesn't matter.

    ReplyDelete
  13. "And once again I will stress - when you survey businesses they do not cite these problems as their concern, and as Gary demonstrates the NFIB has been bending over backwards to explain away their own data because it doesn't say what they want it to say."

    Actually, fully 1/5th of those surveyed by the NFIB (does this include potential business owners - if start ups are leery of, well, starting up, then that is also an issue) do cite regime uncertainty as their primary concern. 1/5th is nothing really to sneeze at. As for the NFIB bending over backwards, well, that looks like a rather lame argument.

    In your effort to completely dismiss the concept of regime uncertainty you come off sounding rather unconvincing.

    ReplyDelete
  14. I haven't "completely dismissed" the concept of regime uncertainty, Gary. Please try and read my comments before putting words in my mouth like this. I have doubts about how substantial a role it plays. Obviously I'm not going to, nor am I trying to "completely dismiss" it, no matter how many times you repeat that to yourself.

    ReplyDelete
  15. Daniel interesting stuff but can I complain about the rhetoric? For example, when DeLong says, he receives "silence," and how he's annoyed by it, I wonder what he means? He actually asked Arnold Kling--who posts about 8 times a day about what he's reading and what he had for breakfast and how he noticed PSST in his Fruit Loops--and Kling ignored an opportunity to talk more about his theory on the EconLog blog? No way. I think DeLong is just making that up.

    And then when you say there is no evidence "whatsoever" that businesses are worried about taxes and health reform, what do you mean by that? E.g. the Congressional testimony I just participated in, both the trucking company owner and the lady who was some kind of head of a small business council said these things. So were they both lying, because you know there is no evidence "whatsoever" about this fear?

    Anyway, a mild complaint. Kind of like when people say "nothing could be further from the truth," when what they really mean is, "I disagree with that."

    ReplyDelete
  16. You are welcome to complain about the rhetoric, but compared to other things DeLong has said that particularly case seems awfully mild!

    No - of course that doesn't imply that they were both lying, Bob. Well, put it this way - if you read the sentence in the context of the paragraph it doesn't imply they were lying, although perhaps I should construct sentences more carefully so that they don't have to be read in the context of the paragraph. All I was saying is that there is no evidence whatsoever that this is a major cause of the downturn. Certainly some people are uncertain about government policy. From my perspective, I would note that there are a lot of businesses out there who are concerned that Congress won't raise the debt ceiling and they're concerned there won't be additional stimulus. I'm not quite sure I have the evidence to say that that fear and uncertainty is what's keeping us in a recession, though - so I don't say that. But certainly I could furnish anecdotes about people worrying about that that true but irrelevant as evidence of the wider proposition.

    ReplyDelete
  17. You completely dismissed it:

    "Is there any sign at all that investors are worried about tax increases or health reform? No! There's no evidence of this whatsoever. Time after time when businesses are surveyed they cite demand as the problem - not concerns about government overreach or taxes. This whole 'regime uncertainty' argument doesn't even pass the smell test in the first place, and on top of that there is no empirical evidence for it."

    Anyway, now you moved onto it not being a "major cause." Should that be considered a clarification?

    Anyway, there is evidence that for 1/5th of the businesses out there regime uncertainty is their number one conscious concern. So it is at the very least a contributing factor.

    ReplyDelete
  18. There is a difference between the general public and business owners. THhe general public might tend to like Obama's policies, but the business owners who are on the front line of dealing with them are in fact uncertain as to what is going to happen, and are responding accordingly. One has to look at it that way, and dissociate oneself from whether or not one likes the policies. Look at it from another's point of view.

    Like I said, with PSST, I was trying to play devil's advocate. I agree with you on the general gradualness of such changes. However, what if PSST were combined with ABCT, as I suggested? The ABCT situation would shorten the time scale that usually spreads out PSST, concentrating the effects. Or it may just be that some ABCT effects are worse than others -- the tech bubble hit a few millionaires and billionaires; the housing bubble hit people across the board, from investors to home owners to builders and their workers. It may be that PSST comes into effect precisely because one now has to train all these people who were misallocated human resources.

    Just some thoughts. I think it's important to think through all the elements one can, consider all possibilities.

    ReplyDelete
  19. re: "There is a difference between the general public and business owners."

    To a certain extent, yes. But the general public makes investment decisions too even if they aren't managing a business.

    That still doesn't change the fact that you're projecting your view of Obama onto the investor/business community.

    re: "One has to look at it that way, and dissociate oneself from whether or not one likes the policies. Look at it from another's point of view."

    Precisely my point, Troy.

    re: "However, what if PSST were combined with ABCT, as I suggested?"

    Maybe, but this would require that the adjustments under consideration are associated with capital structure elongation, and this does not seem to be the sort of thing that Arnold Kling is typically refering to.

    I have a new post up on PSST, in case you're interested.

    ReplyDelete
  20. "When you ask believers in 'recalculation' what pattern of production and trade proved to be unsustainable in 2007, they answer: "building so many houses." When you ask believers why the market economy has been unable to sort out this problem in three years, they answer with nothing--silence. When you say that OK, there were $300 billion of excess houses at the start of 2007 but now construction has been so depressed for so long that there are $1 trillion fewer of houses than trend and why isn't the 2007 pattern of production and trade sustainable again, they answer once again with nothing--silence. That annoys me."

    Please. How can recalculation possibly take place without a market-based interest rate, the most important price of them all? DeLong's beloved fiat-based super low rates will continue to impede economic calculation and thus "recalculation". There is no silence on this issue. Low rates have been catastrophic and will continue to be. As Hayek said in a 1975 speech:

    “The primary cause of the appearance of extensive unemployment, however, is a deviation of the actual structure of prices and wages from its equilibrium structure. Remember, please: that is the crucial concept. The point I want to make is that this equilibrium structure of prices is something which we cannot know beforehand because the only way to discover it is to give the market free play; by definition, therefore, the divergence of actual prices from the equilibrium structure is something that can never be statistically measured."

    We cannot and will not know the equilibrium interest rate because it will not be allowed to come into existence. No one can conceivably know what houses are actually worth because the possibility of any informed long-term economic calculation has been completely distorted by low interest rates with regime uncertainty thrown in for some extra bad effects.

    ReplyDelete

All anonymous comments will be deleted. Consistent pseudonyms are fine.