Believe those who are seeking the truth. Doubt those who find it. Andre Gide

Monday, May 12, 2014

Desperately Seeking Humility

For what it's worth, I think that Paul Krugman and Simon Wren-Lewis come down way too hard on Tony Yates

In a nutshell, Krugman and Wren-Lewis claim that the economics profession knows how to diagnose and treat post-Lehman-like recessions. The diagnosis is self-evident: "deficient demand" (not the same thing as a decline in demand.) The treatment is written down in the most basic Econ 101 IS-LM model: increase G (or, at least, do not decrease it). 

Evidently, we are to have great confidence in the IS-LM prescription because the model correctly predicted that a massive increase in government debt would not lead to soaring interest rates and that a massive increase in the (base) money supply would not be inflationary. 

Well, that's nice. But I don't read Yates as claiming otherwise. I read him as suggesting that there is still a lot we don't know about many things leading up to the financial crisis and the economic forces governing the (slow) recovery dynamic. And because of this state of affairs, politicians were largely free to pursue policies that fit more in line with their political instincts (when do they not do that, I wonder?). Well, I don't know. Sounds plausible. Certainly, I would not have called Yates out for his "self-destructive" attitude. For a similar "self-destructive" attitude, have a look here at what the physicist Freeman Dyson has to say about what people generally don't get about science: 
Q: "How can you tell if someone is a visionary or a crank?"
A: "You can't tell. The whole point of science is that most of it is uncertain. That's why science is exciting--because we don't know. Science is all about things we don't understand. The public, of course, imagines science is just a set of facts. But it's not. Science is a process of exploring, which is always partial. We explore, and we find out things that we understand. We find out things we thought we understood were wrong. That's how it makes progress."
Well, you might say, Wren-Lewis says we know (we know, damn it) that we had a demand deficit because nominal interest rates went to zero just about everywhere (well, except out along the yield curve).  And what about the fact that IS-LM made some correct "counterintuitive" predictions? 
Look, I do not think Yates is saying that this interpretation is wrong. He is saying (correctly, in my view) that it is not necessarily correct. That's a big difference, and it's an important one. 

First off, there are many off-the-shelf models that permit zero nominal interest rates without "deficient" demand. So just because we see zero nominal interest rates is not proof of "deficient" demand. Please be more careful! 

Second, yes, the basic IS-LM model correctly predicted no soaring interest rates and no soaring inflation in the face of fiscal and monetary stimulus. But you know what? There are plenty of models out there outside the IS-LM tradition that made the same prediction. The "new monetarist" models that Steve Williamson works with have this property. Even the simple overlapping generations model I used to interpret Japanese economic developments over a decade ago make the same prediction. Moreover, as I remarked in that paper back then, increasing G is not necessarily the right thing to do in a liquidity trap. Again, I'm not saying the contrary opinion is wrong. I'm saying that it's not obvious. 

And it is even less obvious when we stop to consider that all of these models (my own included) basically begin by assuming that everything in the economy is just peachy when suddenly, a bad shock happens. Goodness, even conditional on this obviously wrong approximation, shouldn't we at least be taking some time to figure out the true nature of the shock? Just because demand falls does not mean it is "deficient" in the way "deficient" is usually defined in this class of models. Demand could be falling for all sorts of reasons, and probably many reasons. Maybe demand should be falling, conditional on these other factors changing?

Maybe it's not a good idea to cut G during a severe recession (certainly, I have argued elsewhere that economists might at least agree to increase infrastructure spending with debt finance). But then, maybe it is (who can really know for sure?). Maybe you view my attitude as "destructive." I view it more in line with Freeman Dyson's answer above. 


  1. Maybe it's not a good idea to cut G during a severe recession.... But then, maybe it is (who can really know for sure?). Maybe you view my attitude as "destructive."

    I'd say "catatonic" would be a better description. Maybe it's not a good idea to execute a central banker every week while the recession lasts. But then again, maybe it is. How can we know?

    I'm not seriously proposing such a course of action, but given your radically agnostic approach I don't see how you can argue against it.

    1. Kevin, I like the label "radical agnostic." I should use it as the name of my blog!

      Alright, Kevin. Fair point. But now, let me tell you something I know for sure: it will definitely restore economic prosperity to execute a central banker every week a recession lasts. There--now do you feel better?

      There seems to be a demand for religion out there, and plenty of people willing to supply it.

      In any case, if you read my post carefully, you'll see that I am not so radically agnostic. I have in the past supported deficit-financed increase in G (infrastructure). The point of my post was not to advocate for policy paralysis, but for leading lights of our profession to show more humility. We don't really understand things as well as some of our high priests are preaching.

  2. Dyson:

    "We explore, and we find out things that we understand."

    And then people cast doubt on that understanding because of political considerations like what happened with Galileo.

  3. Kevin Donoghue, above, sums the attitude up well: better to act in ignorance than freeze when confronted with uncertainty. I think what this means is that it "feels better" to act, not that it necessarily is better.

    To be fair, many claim the risk of inaction outweighs the risk of action. I personally would "feel better" if those making that claim pondered the unintended consequences of their proposals.

  4. The advice "don't just do something, stand there!" is accurate. I think that the main lesson of macroeconomics is guarded agnosticism. If anyone says "I know what to do!" take great care. When the data can fit many models that would suggest different policies, and there's no empirical way to say which is the correct model, we have to be guided by theory. In other words, which model is most theoretically consistent with how real humans behave? We also have to keep in mind the lessons of public choice (Tullock, Buchanan, North, etc.) and understand that even if the prescription is to raise G, not all G is created equally and incentives matter.

  5. What a bunch of baloney. Krugman is not saying thinks are simple. He IS saying that given a choice of policy prescriptions about what to do to recover from the recession some of them were not good enough AND some of them were absolutely the wrong thing to do, hence it was easy to see that:

    1. The stimulus was too small.
    2. The turn to deficit reduction was disastrous.
    3. Householders needed help and HAMP, HARP etc were just not good enough.
    4. Unemployment benefits needed to be extended - even in 2014!
    5. The long term unemployed need an action plan.

    All of this was easy to analyze yet for some reason the powers that be decided it would not be done - I wonder why?

    1. Pinkybum,

      When PK says we can appeal to Econ 101 to diagnose and fix problems, he is saying that things are simple.

      And in terms of your list, I do not think it is so easy. The stimulus was too small? How big should it have been, and how long should it have lasted, and what would have been the short and long-term effects? Do you know? Maybe we had the right quantity but with the expenditure not directed to its best uses? What does IS-LM have to say about that? Was the economy just humming along beautifully prior to the recession? Most models assume yes. Is that reasonable? Might policy prescriptions depend on the state of the economy prior to the recession? What is the optimal policy response if the current policy in place led to the recession? I could go on and on. There are lot of unanswered questions. I certainly respect people having a point of view, like the one you have above. BUT, I do not respect the statement that this is *obvious* or *easy.* It is not. Not for me, at least. Thanks.

    2. I agree with you that the stimulus could have been spent more wisely. A large proportion was used for tax cuts which do not have a very large multiplier. We know they are not that effective because growth through the 2000s stayed about the same until the recession despite there being the biggest across the board tax cuts in our history. However, we do know what the output gap was in 2009 Q2 GDP was $14357B potential GDP was $15499 (according to FRED) so we were down by $1.1T.

      If all the stimulus had been spent on getting people back to work then maybe $700B was enough but I think it would have fallen short. The economy before the recession obviously had many structural flaws most notably the housing boom and wage stagnation (for the 99 percent) however, it cannot be argued that the GDP necessary to maintain full employment should have been lower. In fact unless you are going to cut everybody's wages (note the wage stagnation mentioned before) getting back to full employment required the government to spend, fix the unregulated financial industry and restore tax levels to at least Clinton levels. There should probably also be higher tax levels on the richer amongst us and taxation of interest income just like earned income from labor.

    3. Over at Simon Wren-Lewis's blog, DA writes, "Other "fundamental" forces that could depress demand include people being afraid or uncertain of the future evolution of fiscal policy (taxes that punish the accumulation of physical and human capital)." So he presumably thinks that your "higher tax levels on the richer amongst us and taxation of interest income just like earned income from labor" would be punitive and depressing. On the other hand, if we actually came out and instituted higher tax levels on the rich, on interest, even (gasp) on capital, then maybe DA's fearful capitalists could get over their fear of being punished and could get back to investing their hard-won money.

  6. David, nice post. I like the humility angle! You talk about different models explaining what we observed and about science. I'm not a scientist nor an economist, but I enjoy reading macro themed blogs and trying to understand them. Recently I found one that offers explanations for many of the things we've observed recently and over the past 80 years or so in various countries, and this blog takes a very unique approach which examines how the market transmits information. I'm super curious as to what more established economists such as yourself think of this approach. The author (Jason Smith) offers some formulas which reduce to more familiar economic results under certain conditions (such as the QTM). For example he has functions for the price level (P) and long and short term interest rates, and he has an explanation for why different countries react differently to increases in their monetary base (MB) including an analysis of the US over different time frames (pre-WWII, 1970s, and present):

    rl = R(NGDP, M0)
    rs = R(NGDP, MB)
    P = P(NGDP, M0)

    Where M0 is currency in circulation.

    Here's a few links (starting w/ a comment in which he mentions those three equations above):

    And some analysis about the use of expectations in economics:

    Regarding an explanation for 2008, I recently asked Jason about that:

    1. Tom, I took a quick look at some of the links you refer to above. Let me read through them and ponder for a while--there's a lot of material!

    2. Here's one of Jason's that elicited multiple responses from Sumner, Sadowski, and Nunes in the comments section:

  7. Yates is right. We shouldn't rule out that JFK was shot by space aliens.

    1. Ah, Kaelberg...did that make you feel better?

  8. I don't remember much intellectual humility or radical agnosticism from the economic profession pre-crisis. Rather, it told the rest of us that there was no alternative, that its prescriptions were economic rationalism, that we know what works and free markets work, that we lived in a republic of the central bankers. No profession except perhaps medicine could match pre-crisis economics for swaggering, pompous, Prussian arrogance. And then you did it again with Reinhart-Rogoff, expansionary contraction, and internal devaluation, all of which were offered as doctrines from which no dissent was acceptable, and you pratfalled yet again. The whining now only highlights the strutting then.

    "Oh, we can't really ever know anything, and therefore, how can anyone really say we were wrong". Give me a break.

    1. Alex, you don't remember it because it lives beneath the surface, among the dedicated group of academic economists who toil away in relative obscurity.

      And I did not mean to suggest anything close to your parting sentence. I am having a hard time not thinking of you as a pathetic, ignorant fool. Oops, I guess I shouldn't have said that. Especially to an anonymous poster. Sorry.

    2. Remind me how I'm anonymous, clown.

    3. There is a charge the beneath the surface crowd cannot escape and that is that the public face of the profession was bought and paid for and no one said a word about the lack of ethics and conflicts of interest. The worst offenders are at Chicago, starting with Cochrane

  9. My amateur interpretaion is that we've spent so many generations increasing "G" that it is all that matters. That alone scares people. Its the type of realization that creates doomsday preppers who would not have been created otherwise.

    Good post david.

    1. Just read on another site a data point that backs up too much "G". Ukraine just issued bonds today, apparently, backed by a US entity, who in turn is only backed by american taxpayers. Now image being a stuggling middle class american and learning again that you are the only collateral for even more debt obligations to build another country in our image. Given the prevailing fear that the us can't handle its current set if commitments and convinced this ukraine bond deal is just more evidence of a hopelessly sacked us do you respond? You already know your effort and personal assets are all that support the issuance...and that you are already struggling. So, you give up and go on gov assistance or prep for a fight when they come for you.

  10. Again, a very informative essay, albeit a touch too much of a radical agnostic.

    Too many people with skin the game (and without incentive caused bias) have made compelling cases that lots was wrong before Lehman. Clay Christensen is at the forefront to say it was simply all lack of demand

    I am working on a paper on bottlenecks in the labor market of loan officers when the know how of a bank's existing loan officers becomes irrelevant due to changes in the market where it has historically lent that alone raises questions about problems well before 2007.

    Further, thanks to Roger Farmer and others we have a much clearer picture on animal spirits and an understanding that, due to limited information, neoclassical perfect equilibrium can never be attained.

  11. Respectfully, David, do you impose the same humility imperative upon those who've predicted accelerating inflation over the last five years?

    1. Jonas, yes, of course.

      Having said that, I'm not sure who these economists were that have allegedly been predicting accelerating inflation. I mean, mainstream types. I know there is Meltzer. I think Ferguson did, but he retracted. A few Austrians, maybe? No one at the Fed I talk to ever thought that. And when they did, they usually couched it the way Krugman did, as a *conditional* statement "if bank reserves enter circulation..."

      By the way, while I'm on this, what do you think of PK claiming that he "predicted" low inflation? He did no such thing. He always made conditional statements. He was being humble in the past. But he now seems to speaking as if he made an unconditional forecast.

      Finally, here is PK on Romer's textbook:

      "Romer’s notes still imply that a protracted liquidity trap should lead to accelerating deflation, which doesn’t seem to happen; I think most of us have turned to downward nominal wage rigidity as an explanation. In any case, this is more or less the state of the practical art, and I’m delighted to learn that he’s put it together."

      So, you see, the model that PK hangs his hat on actually predicts an accelerating deflation. But I guess that's no reason to toss that model. It's only an opportunity to crow how the inflationistas were wrong?

    2. I will never understand how you guys claim there has been absolutely no inflation since the crisis.

    3. Lles, who's claiming absolutely no inflation? They're claiming low inflation (like 1% to 1.5%... below the 2% target). If you don't trust the gov then look at MIT's billion prices project. It tracks the gov's numbers pretty well.