Sunday, July 21, 2013

New Keynesian Models and Monetary Policy

For people who want a basic explanation of where New Keynesian models came from, why central bankers find them attractive, and a critique, from a policymaker's point of view, this speech from a couple of months ago by Charles Plosser is helpful.

37 comments:

  1. Ah, Plosser, another guy who is deeply worried about inflation while unemployment is high.

    Surprising how many folks have not read their Fisher:

    "Should inflation expectations begin to fall, we might need to take action to defend our inflation goal, but at this point, I do not see inflation or deflation as a serious threat in the near term"
    http://www.bloomberg.com/news/2013-05-14/fed-s-plosser-says-slowing-inflation-not-a-concern-for-policy.html

    ReplyDelete
    Replies
    1. Could be he's changed his mind since May. Seems Bullard, for example, is more concerned with low inflation than Bernanke is.

      Two questions:

      1. Seems you think the unemployment rate determines the inflation rate. Where do you get that idea?
      2. I assume you mean Irving Fisher. You'll have to expand on that. Not sure what you mean.

      Delete
    2. Then you might wanna read "The Debt-Deflation Theory of Great Depressions".

      About the former question, sounds like you are implicitly denying that monetary policy can have any positive effects on output, like you are denying that the basic trade-off of monetary policy is output stabilization vs. keeping inflation (and inflation expectations) low.

      Delete
    3. It doesn't help to cite Irving Fisher. Give me an answer to the above two questions, in your own words. I'm not arguing for the short-run non-neutrality of money, just trying to understand what you're trying to say.

      Delete
    4. Yes it does. Right now we are in a balance sheet recession so Fisher '33 applies: we gotta worry about deflation / low levels of inflation as they have detrimental effects on deleveraging.
      This is basic stuff we learned in the thirties, when we are not in an ordinary recession but one which is caused by too much debt and when we are in addition to that in a liquidity trap we are "through the looking glass", fair is foul and foul is fair, we should not worry about what we usually worry about, the long run goals of keeping inflation, inflation expectations and government debt on low levels.

      Delete
    5. So, suppose we could have more inflation if we wanted it. You're arguing that more inflation would be good, as this would reduce the real value of private debts. A couple of problems with this. First, you're redistributing from lenders to borrowers. You might think that this is redistribution from rich to poor, but that's not clear. Plenty of rich people hold large mortgages. Second, monetary policy is a blunt tool for accomplishing what you want. Fiscal policy would work better, if your goal is debt forgiveness and redistribution from rich to poor.

      Delete
    6. The goal is not redistribution but ending the balance sheet recession via deleveraging.
      Companies do not invest when they are indebted, banks do not lend loans when they gotta bring their balance sheets in order, consumers do not spend when they carry a lot of debt around.

      Delete
    7. The way you put that, it seems debt must always be a bad thing. Maybe you think we should ban it altogether?

      Delete
    8. Besides denying Fisher '33 it looks like pathetic polemics is all you are able to do.

      There is nothing wrong with debt IN GENERAL as many economic agents can deal very well high level of debt. But there is a lot wrong with debt in a balance sheet recession as it keeps down consumption, investment and credit supply.

      I guess this is the last thing I will post on this blog as talking with an "economist" who denies the 101s of his discipline is as pointless as debating with people who think that the earth is flat, the difference being that the latter are harmless crackpots whereas the former, given the vital real-world importance of economics, are dangerous crackpots.

      Delete
    9. "I guess this is the last thing I will post on this blog"

      That's the best news I've gotten all week.

      Delete
    10. Too bad he/she is gone. I was going to ask for a definition of "balance sheet recession," to see where that would go. That's one piece of jargon I can do without.

      Delete
    11. But the creditors to which companies and consumers are in debt CAN invest and consume more. The question is, why don't they? Your balance sheet recession story does not answer this question. You seem to forget that the creation of debt requires two parties. Not everyone can be indebted at the same time!

      Delete
    12. Exactly. And you have to worry about default. Bankruptcy laws, and the law governing foreclosure on mortgage debt provide some kind of coarse contingency to what would otherwise be non-contingent debt contracts. That is, the legal structure governing debt is providing some insurance against bad states of the world for debtors. If you're saying that the central bank needs to step in to provide some extra inflation whenever we think the default rates on on private nominal debt might become high, we're saying that somehow the central bank can provide better insurance than we're getting through the legal system. And of course, as in all matters regarding insurance, there's moral hazard to consider. Nothing obvious about this problem at all.

      Delete
    13. Even before Stiglitz&Weis we have known that financial markets are imperfect via mere observation, there are financial intermediaries and the interest rates of lenders are smaller than those of borrowers.

      When we are no longer in the Arrow-Debreu world, when there are credit rationing and liquidity constraints, high levels of debts have an impact if there is a collapse of the price of an agent's assets or a reduction of cash-flow (be it because of house price decline and unemployment in the case of household or output demand decline in the case of firms). Once again, debt matters because of imformation-incentive problems on financial markets.

      Not that you need all this theorizing, just taking a look at what economics agents actually do right now suffices to not even understand but just plainly see that their deleveraging reduces investment, consumption and credit supply.

      The implications of taking deleveraging seriously are not necessarily "loose money", that you should inflate debt away.
      The only implication is what Fisher said back in the days, that too low levels of inflation or deflation makes deleveraging too hard.
      So it is asymmetric, it is not about the potential chances of high inflation (I disagree with progressive economists who think that 4% or 5% inflation would be a good thing) but the risks of low inflation / deflation.

      Delete
    14. I agree with about 80% of that. It could be that the costs and benefits of inflation (anticipated and unanticipated) have more to do with debt contracts denominated in nominal terms, and the costs of default than the things people usually think about (sticky prices and wages, intertemporal distortions). It would be interesting to sort this out properly. Currently, I don't think we have much of an idea whether 2% inflation forever is a good idea (relative to 0% or 10%, for example), whether the optimal inflation rate should vary over time in response to aggregate shocks, etc.

      Delete
    15. "Not that you need all this theorizing, just taking a look at what economics agents actually do right now suffices to not even understand but just plainly see that their deleveraging reduces investment, consumption and credit supply"


      Well, suppose there is an increase in the cost of financial intermediation that financial institutions pass on to borrowers, causing firms to cut down on investment. My guess (although I need a model to confirm) is that with the right labor market frictions the outcome would be somewhat similar to what we observe today. Yet the resulting deleveraging would be an outcome of the crisis, not the cause. And the solution would be not to assist the deleveraging with higher inflation, but to address the increase, if possible. I am not saying that this is truly what is going on (though it might be). My point is that correlation does not imply causation! Many theories can produce results qualitatively similar to what we observe for at least a few economic variables. This is why theories need to be evaluated based on their quantitative predictions, based on how well they can mimic the behavior of an array of economic variables.

      Delete
    16. That was part of the 20% I have a problem with. You in fact do need "all this theorizing." Nothing is obvious from staring at the data.

      Delete
    17. Of course we need theory. But if you are unsure which model might best explain a real world phenomenom it might help to simply take a look at what economic agents are doing and ask them why they are doing it. In other words, get out of the ivory tower for a short time.

      I am not advocating anything à la Binder's "On Sticky Prices: Academic Theories Meet the Real World", actually making papers out of stupid interviews. But casual contact with the real world would at least prevent the economics profession from coming up with nonsense like "an increase in the cost of financial intermediation leads to deleveraging". This is only moderately more sane than Mulligan's "foodstamps cause unemployment" story.
      Of course long-run costs of intermediation should have an impact upon the average long-run level of debt ... but what we are dealing right now with is a short-run demand-side and not a long-run supply-side problem.

      Of course what you guys are doing is to deny that we have a rough understanding of what is going on right now and claim that everything is a mystery and that there could be dozens of explanations precisely in order to disavow the actual answers people on the side where the short-run importance of demand and liquidity constraints are not denied.
      Which is fine if, to use Mankiw's terms, you just wanna be a scientist. But if you wanna be an engineer you better come up with a story that gets the rough picture right. At least I prefer to be roughly right than precisely wrong.

      Delete
    18. "what you guys are doing is to deny that we have a rough understanding of what is going on right now and claim that everything is a mystery and that there could be dozens of explanations"

      For some things I observe, I think I have a coherent theory, or set of theories, that is roughly consistent with those observations. In other cases not. Sometimes I look at what other people claim is their "rough understanding" of what is going on, and they seem convincing. Sometimes I think what they're saying is nonsense, and that those people are remarkably overconfident, deluded, liars, or some convex combination.

      So, how do you know you're close to being "right" and are not "precisely wrong?" A good engineer understands uncertainty, I think.

      Delete
    19. The problem is that the practice of economics is more similar to practicing medicine than engineering. And one of the most powerful principles in practicing medicine is the "primum non nocere" (first do no harm). Would you trust a doctor who might have a rough understanding of your illness stick an IV with some experimental concoction in you? If not, then you should resist the temptation to follow policy advocates whose stories are murky, simply because no better stories are available. The balance-sheet recession theory, when fully specified as in Krugman and Eggertsson, has so many holes that it is hard for me to take it seriously. For example, maybe you do, but I do not see net creditors going on a spending spree in response to lower interest rates, as thy do in K&E. How does this square with the story?

      Delete
    20. I do not know Krugman and Eggertson but I know my Stiglitz, credit markets are rationed so one has to obviously distinguish between a lower federal funds rate and lower loan rates. If the former falls while the latter does not it is clear that banks are still engaged in deleveraging.
      But as we are in a liquidity trap central banks are not using the federal funds rate as main tool, they are rather buying up assets which include ones with longer maturities.
      Of course here a problem appears, if banks can get rid of their crappy assets like this via QE shouldn't they have started to expand credit long ago?
      If you are a monetarist they should indeed but if you are a Keynesian this is hardly surprising: aggregate demand is low so investment and thus credit demand is low.

      This is hardly a murky story, it is basic stuff which all of us should have learned in our undergraduate studies and it is becoming tiresome to debate with people who deny econ 101.

      About your do no harm analogy, this is utter nonsense and a pathetic rationalization of an ideological and not a scientific position upon dealing with recessions.
      Every school of economics, be it monetarism or Keynesianism, accepts for example that monetary policy should be used as business cycle smoothing tool.
      What we usually debate upon are the details of monetary policy and the effectiveness of fiscal policy.

      So if you claim that public institutions should do NOTHING in a balance sheet recession, i.e. the most rate and worst kind of recession, you are not doing economics but are just guided by some laissez-faire libertarian ideology or whatever.

      Delete
    21. So many words, so few correct ideas.

      Delete
    22. OK OK, here are some questions.

      1) Are you saying that real interest rates on longer-term loans have not fallen (albeit not as much as the Fed funds rate)? If so, you should take a look at the data. You will find out that you are wrong, in which case households and firms that are not constrained by excessive debt should have increased spending while debt-constrained households have been deleveraging. Once again, I may be wrong, but I do not see such big dichotomy in the behavior of different types of consumers and firms. This raises questions regarding the validity of any version of a balance sheet recession story.

      2) Do you understand the difference between the Krugman balance-sheet recession story, which primarily concerns the demand for credit, and Stiglitz's story which primarily concerns the supply of credit? Do you understand that their policy implications are different?

      3) Do you understand that monetary policy is a mixed bag? That there is a difference between conventional policies on which there is consensus (e.g. lowering the Fed Funds rate), and policies like QE that are untested and not well understood?

      4) Are you unaware that the effectiveness of fiscal policy is also still in question, especially depending on what form it takes (e.g. fiscal transfers vs infrastructural investment)?

      5) Are you really incapable of understanding the difference between being skeptical of unconventional, untested, and disputable policies and being a laissez-faire libertarian?

      6) How can I deny ECON 101 when I happen to be teaching it? Could it instead be that ECON 101 is so basic that a background of ECON 101 alone is inadequate to understand the complexities of economic policy?

      7) You are correct on one thing though, the term "laissez faire" was indeed coined in the 18th century by a medical doctor turned economist, Francois Quesnay, perhaps inspired by the "first do no harm" principle. So have you contacted yet the American Medical Association to inform them that their practices are "not scientific"?

      8) Don't you think it would be better for someone with limited knowledge on the subject they are debating to avoid pompous expressions and characterizations like "utter nonsense" and "pathetic"? I understand that anonymity largely protects you from ridicule, but still.

      Delete
    23. How can somebody who claims to teach economics not understand how QE works? The CB buys assets with longer maturity and assets which public players do not want anymore. It is a simple swap of risky for riskless assets and thus assists deleveraging.
      I don't even have PhD and have to teach a guy who claims to be a professor such basic stuff? Seriously?

      As you have no read your Fisher, Koo and Keynes the Econ101 you claim teach is obviously some neoclassical RBCish "food stamps caused the great depression" nonsense (I am all for classical economics in the long-run but obviously it is unable to explain short-run phenomena like involuntary unemployment).

      Claiming that we should not conduct massive monetary and fiscal policy with two-digit unemployment rates in many countries and with extremely low inflation and extremely high government bond prices (Europe being a currency union with a DeGrauwian multiple equilibria situation differs) because these employment-enhancing measures could actually do harm (Like what, inflation around the corner, exploding public debt or what?) is more than just ignorant and cynical. It is utterly repugnant.

      Delete
    24. "I don't even have PhD and have to teach a guy who claims to be a professor such basic stuff?"

      Clearly, to borrow a line from Jon Stewart, you live in Bullshit mountain (apparently the tea-baggers are not the only ones there). When you decide to come down, we can talk again. In the meantime, on whether we understand how QE works, you may want to read Steve's latest post, as well as some earlier ones on the topic. For your convenience, here are some links:
      http://newmonetarism.blogspot.com/2013/03/ben-bernanke-and-term-structure-of.html
      http://newmonetarism.blogspot.com/2013/03/the-balance-sheet-and-feds-future.html

      Oh, and once again, I am not defending classical, neoclassical, new-classical, or other-classical economics. How could I when in my latest paper under review I criticize the classical view of the labor market? I understand fighting a straw-man is easier, but still. Finally, the scientific world has moved beyond the classicists vs. Keynesians debate. We have newer and better theories. You should move on too!

      Delete
    25. Previous Anonymous,

      Just go away. Your first-grade understanding of economics does not contribute anything to society. As soon as you learn that, you'll be happier and so will we.

      Signed,

      The entire economics profession

      Delete
  2. Professor Williamson,

    I do not agree at most of what you have said last several years. But I respect you and have taken your opinions seriously. But not Mr Plosser (or R. Lucas or E. prescott). Our world is better off by ignoring them.

    a reader from Asia

    ReplyDelete
    Replies
    1. I don't agree with everything that Plosser (or Lucas or Prescott, for that matter) says or writes. But Plosser is one of the best people on the FOMC, in my opinion, particularly in bringing what he knows about economics to bear on the policy problem. At 72 (Prescott), and 75 (Lucas), I still learn something whenever I see either of those guys. They are both serious scientists, and fine human beings, and it's your loss if you ignore them.

      Delete
    2. Dear Reader from Asia,

      No one agrees with anything you have ever said, and everyone is right to do so.

      Signed,

      The entire economics profession

      Delete
  3. It is mind boggling that Plosser sees a consensus among economists that only unanticipated monetary shocks matter for output. How can anyone that holds such beliefs deserve to be taken seriously?

    ReplyDelete
    Replies
    1. Are you saying that there is no consensus on this point, or that the point is wrong? Can you cite any scientific work supporting your point?

      Delete
    2. Read the speech. He's saying that was the consensus 40 years ago. Then he goes on...

      Delete
    3. Anomino appears incapable of reading, or perhaps it is simply reading comprehension that eludes him.

      Delete
  4. The speech is from March 2012, not March 2013. But it's still great!

    ReplyDelete
    Replies
    1. Sometimes it's hard to tell a 2 from a 3, don't you think?

      Delete
    2. If you drink enough beer, it's even tough to tell a 2 from a 10!

      Delete
    3. Standing on your head, there's no problem with 8, but if you're drunk on the floor, and looking at it sideways, it's something else altogether.

      Delete