Tuesday, July 3, 2012

Reply to Sumner's Reply

Scott Sumner replies here to my last post. Some comments:

1. I've discussed at length why I think QE is irrelevant under the current circumstances, most recently here. If the Fed could tax reserves, that would certainly matter, as Sumner points out, but that's not permitted in the United States, and so is not a practical option. Sumner has no special claim to be rooted in the "real world" here. I'm looking at the same real world he is. However, I think I'm more willing to think about the available theory. Apparently he doesn't think that's useful. The key problem under the current circumstances is that you can't just announce an arbitrary NGDP target and hit it with wishful thinking. The Fed needs some tools, and in spite of what Ben Bernanke says, it doesn't have them.

2. Sumner is correct about the change in views of central banks that occurred post-1970s. However, Volcker needed courage to change course. He faced a lot of opposition from within the Fed and without. I'm more cynical about the way central banks adopted New Keynesian economics later on. Rightly or wrongly, this wasn't stuff that was challenging what they were doing - more like an exercise in reverse engineering. How do we write down a framework that justifies the status quo?

3. Sumner says:
...RGDP fluctuations at cyclical frequencies are assumed to be suboptimal, and are assumed to generate large welfare losses—particularly when generated by huge NGDP shocks, as in the early 1930s.
My point in looking at seasonally adjusted nominal GDP was to point out that fluctuations in nominal GDP can't be intrinsically bad. I think we all recognize that seasonal variation in NGDP is something that policy need not be doing anything to eliminate. So how do we know that we want to eliminate this variation at business cycle frequencies? In contrast to what Sumner states, it is widely recognized that some of the business cycle variability in RGDP we observe is in fact not suboptimal. Most of what we spend our time discussing (or fighting about) is the nature and quantitative significance of the suboptimalities. Sumner seems to think (like old-fashioned quantity theorists), that there is a sufficient statistic for subomptimality - in this case NGDP. I don't see it.

4. Friedman rule: My conjecture for nominal interest rate smoothing I guess reflects a view that the intertemporal Friedman rule distortions could be more important than the sticky price/wage distortions. Further, it's not clear an optimal policy in a sticky price/wage model is going to imply a lot of variability in the nominal interest rate.

5.
I’m kind of perplexed as to why Williamson calls himself a “New Monetarist.”
Randy Wright talked me into it. It's as good a name as any. We like some things about Milton Friedman, but some things we don't like - 100% reserve requirement, his need to separate assets into money and not-money. We're a lot more interested in theory too.

49 comments:

  1. Hayek discussed just this issue and distinction in his 1928 article "Intertemporal Price Equilibrium and Movements in the Value of Money" which introduced dated goods to economics:

    "My point in looking at seasonally adjusted nominal GDP was to point out that fluctuations in nominal GDP can't be intrinsically bad. I think we all recognize that seasonal variation in NGDP is something that policy need not be doing anything to eliminate. "

    ReplyDelete
  2. In 19th century & early 20th century economics economists were well aware of season changes in output, credit demand & money deman, as this was a central fact of the agricultural economy.

    So they worked onmdistinguishing between these seasonal patterns and non-season booms & busts in credit, money & output.

    ReplyDelete
  3. Just a couple of questions here:

    1) If you don't believe QE (in any size) will increase the price level, do you think the Fed could (in theory) buy back all 15 trillion or so of US debt? At that point couldn't they sell it to the Treasury for cash without any inflationary impact? That doesn't seem realistic, but if it were, wouldn't it be a good thing to do?

    2) If you do believe the above action would result in increased inflation expectations, then doesn't that imply that QE would accomplish it's goals through the expectations channel?

    ReplyDelete
  4. So we turn the remainder of the outstanding federal government debt into reserve accounts at the Fed. Why do you think that makes a difference?

    ReplyDelete
    Replies
    1. I'm saying if your assertion is that it wouldn't make a difference, then that is a very strong argument in favor of large-scale QE. We could zero out the federal debt with no adverse inflationary consequences. Isn't that the implication of your model?

      Delete
    2. JSR, QE doesn't "zero out" the public debt. It changes the maturity of the debt, not the quantity.

      And from a tactical point of view, now would be a good time to sell more long term bonds, not fewer!

      Delete
    3. I agree with the first two sentences of Max's reply. The last sentence is a point that John Cochrane also made. For the US I think it doesn't matter, given how the private sector can intermediate government debt.

      Delete
    4. Maybe I wasn't clear -- in my example:

      1) Fed buys all US debt (or close enough that it makes no difference)

      2) Treasury buys the debt from the Fed with printed money

      3) Treasury tears up the debt

      In what way does this not "zero out" the public debt? It has effectively taken the debt and monetized it.

      Delete
    5. Yes, you were clear. We get you. You don't get us. It's useful to think of the central bank and the fiscal authority as one consolidated entity. The fiscal authority determines net indebtedness of this consolidated entity to the private sector, and the central bank can conduct asset swaps that change the composition of consolidated-entity debt. The central bank cannot change net indebtedness of the consolidated entity. Currency and reserves are just part of this net indebtedness - they are liabilities of the ocnsolidated entity, which is why they appear as such on the Fed's balance sheet. "Monetize" the debt, and you don't change net indebtedness to the private sector.

      Delete
    6. But what you say only makes sense if public debt and base money are perfect substitutes, which they are not.

      Delete
    7. Thanks for that -- I think I understand your position better now. That raises other questions for me, if you have time to address them:

      It sounds like your position that reserves and treasuries are perfect substitutes is conditional on the fact that banks are currently holding excess reserves. Do you believe that once a recovery has sufficiently taken hold, these excess reserves will be used elsewhere and inflation will rise? If so, doesn't that still imply that increasing reserves now has an impact on expectations of future inflation?

      Delete
    8. "So we turn the remainder of the outstanding federal government debt into reserve accounts at the Fed. Why do you think that makes a difference?"

      and then turn them over to the Treasury to effectively extinguish the debt with base money. Extinguishing the national debt by replacing it with base money would be equivalent to a massive tax cut now (plus a tax increase later when inflation rears its head). Imagine for example that the Govt cut taxes to zero and bought everything with money. In the limit, that is massive QE.

      Only a "little" QE works because such a debt monetezation policy is not credible, and in fact only a little debt is getting monetized. The more QE the Fed does, the more revenue is raised through seigniorage. You can call that monetary policy, fiscal policy, or "Blue." It still stimulates the economy and raises AD.

      If you really think that QE does nothing, lets extinguish the debt and cut taxes now!!

      Delete
    9. "Only a "little" QE works because such a debt monetezation policy is not credible"

      i mean: Only a "little" QE has a little effect because such a debt monetezation policy is not credible (settling aside the IOER issue).

      Delete
    10. dwb,
      Substituting Treasury debt with Excess Reserves merely swaps one government liability with another. It reduces the tenor of government debt and shifts duration risk from the private sector to the public sector.

      Such an asset swap has no inflationary consequences; it just transfers risk from private agents to taxpayers.

      The concept of "base money" is less than useful in the presence of ER's.

      Delete
    11. "Substituting Treasury debt with Excess Reserves merely swaps one government liability with another."

      only if you think that both kinds of liabilities are identical. Turns out, most vendors do not accept treasury bills as final payment.

      I don't understand why you object to testing your theory. lets first try extinguishing 1/2 of the publicly held treasury debt (about 4 Tn) by "swapping" or substituting it *permanently* with base money.

      Delete
    12. ... and even though base money and currency is accounted for as a liability on the FRB balance sheet does not make it so. Currency never needs to be "repaid" in terms of (direct) taxes, so its more like equity than debt. Reserves are special because they can be converted to currency. So, no, QE is not swapping one liability for another, its changing the financing structure / composition of the balance sheet and therefore the effects depend in large part on credibility and perception of permanence.

      Delete
    13. "It sounds like your position that reserves and treasuries are perfect substitutes is conditional on the fact that banks are currently holding excess reserves."

      Yes, exactly.

      "Do you believe that once a recovery has sufficiently taken hold, these excess reserves will be used elsewhere and inflation will rise?"

      Yes, in the course of a recovery, other assets should become more attractive to banks and, if the interest rate on reserves is not increased, the only thing left to give is the price level, which must rise.

      "Do you believe that once a recovery has sufficiently taken hold, these excess reserves will be used elsewhere and inflation will rise?"

      Only to the extent, I think, that it extends the date at which the excess reserves go to zero and we return to the pre-financial-crisis monetary regime.

      Delete
    14. Ok, thanks. It sounds like your position leaves room for the possibility that QE may cause increased inflation in the future, although you think the extent of that increase would be small. I just have a few last questions to complete the logical chain:

      Do you believe that the extent of the future inflation increase would scale with the size of the QE? i.e., could the Fed make up for the relatively small traction on future prices by performing QE at a very large scale (trillions)? And do you believe that increased expectations of future inflation would affect the price level today?

      Delete
    15. @Stephen Williamson who said: ""Monetize" the debt, and you don't change net indebtedness to the private sector." -- this is clear error. Monetizing the debt means from what I've seen not using the existing moneys supply but literally printing money, electronically and physically. As such, in the long term it is inflationary.

      Delete
  5. "The key problem under the current circumstances is that you can't just announce an arbitrary NGDP target and hit it with wishful thinking. The Fed needs some tools, and in spite of what Ben Bernanke says, it doesn't have them."

    The Fed can't hit an arbitrary target, true, but the Fed can inflate if it wants merely by announcing a (level) target.

    That's because, given that inflation expectations can change spontaneously, the Fed can promise inflation after the zero bound isn't binding, and this promise will generate immediate inflation. It doesn't matter that the Fed can't lower the interest rate, nor does it matter whether QE works.

    ReplyDelete
    Replies
    1. The Fed can only make promises about things it can actually control.

      Delete
    2. I agree. And the Fed can promise to set the bank rate at an inflationary level when that is possible. And that promise will actually generate immediate inflation given, as I said, the assumption that inflation expectations can "randomly" change on their own (so no policy instrument is required to force a change).

      Delete
  6. I mistakenly posted this in yesterday's post so I'm importing it here:

    Incidentally just in case you aren't aware of it. Lars Christensen has an answer to your answer to Sumner's answer...

    Think I said that right

    http://marketmonetarist.com/2012/07/04/the-fed-can-hit-any-ngdp-target/

    To sum up he's "perplexed" why you're a Monetarist.

    "Williamson claims that he does not agree with everything Friedman said, but I wonder what Friedman said he agrees with. If you don’t believe that NGDP is determined by the central bank then it makes absolutely no sense to call yourself a monetarist."

    ReplyDelete
    Replies
    1. I could call myself a two-headed frog. Who says I'm not?

      Delete
    2. Can I have one of your heads then? You don't seem to need both to write good papers, so share the wealth a little.

      Delete
    3. We are in Alice in Wonderland now. One can define whatever terms they want to mean whatever they want. Last I saw that being done was by patent lawyers.

      Delete
  7. Replies
    1. The commenters on that page are worse than EJMR.

      Delete
  8. The key problem under the current circumstances is that you can't just announce an arbitrary NGDP target and hit it with wishful thinking. The Fed needs some tools, and in spite of what Ben Bernanke says, it doesn't have them.

    This seems to me to be a very, very, VERY strange comment. The Fed is a central bank. As such, as long as there is some asset somewhere in the world that they don't yet own, they can print money and buy it.

    NGDPLT works as long as the Fed is credible, and there is nothing preventing the Fed from being credible.

    ReplyDelete
    Replies
    1. "As such, as long as there is some asset somewhere in the world that they don't yet own, they can print money and buy it."

      That doesn't mean anything happens as a result.

      Delete
  9. I have a basic question about the interest rate on reserves (IROR).

    Unless I misunderstand some aspect of the operation, when IROR increases, the growth rate of base money increases (all else equal).

    One might normally expect an increase in the growth rate of base money to be inflationary/stimulative. Yet an increase of IROR is normally said to have the effect of a monetary tightening (and thus to be contractionary).

    What am I missing?

    ReplyDelete
    Replies
    1. 1. What a change in the IROR accomplishes depends on what kind of monetary regime you are in. If it's a channel system, in which excess reserves are essentially zero, this just tightens the channel, and reduces the cost of holding required reserves (if there is a reserve requirement). There are economic effects, but it's not a big deal in terms of monetary control. In a floor system (what we are in now - excess reserves in the system always), a change in the IROR is a big deal, as the IROR is essentially determining all short rates. If you think of an increase in short rates as contractionary, then indeed an increase in IROR is contractionary.

      Delete
    2. Thanks Stephen, that’s helpful. I’m not yet convinced that the IROR determines all short rates under a floor system.

      Let’s say we’re in a floor system (as described in your June 4 post). Under such a system, a bank presumably would never lend a reserve balance to another bank at an interest rate below the IROR. I think we’re in agreement here.

      But isn’t this just a special feature of the interbank market for reserve balances? What implications does it have for other lending rates, or for the price level?

      For example, suppose the IROR is 5% under a floor system. Suppose also that banks can issue deposit liabilities bearing 0% interest. It seems entirely possible under this scenario that banks might be willing to lend to some category of borrowers at, say, 3%. In other words, far below the IROR. In that case, the IROR isn’t a “floor” on short rates generally. I’m assuming that the supply of reserve balances is finite and determined by the central bank.

      If this is right, then it’s not clear to me how an increase in the IROR is contractionary, even under a floor system. On the contrary, it seems to expand the supply of base money and might therefore be inflationary/expansionary.

      Delete
  10. Let me get make sure I understand this correctly - you believe that QE is ineffective because the Fed is substituting one liability (cash) for another (Treasuries), in effect, issuing dollars and retiring/nettting debt. I agree that the two are substitutes, but not perfect substitutes, as dollars circulate with higher velocity than debt, and affect the prices of a different set of assets (debt drives asset prices, dollars drive consumption prices, though obviously there is leakage, as higher asset prices tend to release more dollars out of "storage", or increase velocity).

    However, if I understand the monetarists correctly, this kind of misses the point of level targeting, which is to increase or decrease the aggregate money supply (Fed liabilities, private liabilities, plus other claims) in such a way to maintain the level target. NGDP is really just another monetary aggregate, so NGDP targeting is managing the growth of a monetary aggregate. Monetarists argue that the Fed has unlimited power to create money and expand the money supply (don't ask me about the legal constraints, because I don't really know what they are and I'm pretty sure they could be changed in a pinch). Monetarists appear a little inconsistent with respect to QE - either seeing it as incoherent or seeing it as a signal for expansionary monetary policy. As a practicioner, I can say that the markets interpret it as the latter.

    I tend to view QE in the substition form, except as directed release of excess reserves held at the Fed that otherwise have no velocity.

    ReplyDelete
    Replies
    1. "you believe that QE is ineffective because the Fed is substituting one liability (cash) for another (Treasuries), in effect, issuing dollars and retiring/nettting debt."

      No, it's substituting reserves for Treasuries. There's no change in anything that changes the terms on which people want to hold currency, so think of the stock of currency as unchanged. Intraday exchange involving reserves can be supported with a very small quantity of reserves, so most of the reserves sitting in the accounts of financial institutions with the Fed just sit - all day and overnight. They're not much different from T-bills, though in fact we could argue that T-bills are more liquid, and this is reflected in a T-bill rate (3-month) that is currently 0.07%, while the interest rate on reserves is 0.25%. But this also extends to exchanges of reserves for long Treasuries, as I've argued elsewhere. If you can't change anything by exchanging reserves for Treasuries, you can't get to your level nominal income target.

      Delete
    2. The "money base" concept seems unnecessarily confusing. It implies since all reserves are part of the "base", they are also automatically "money". I think this is why many people -- even economists? -- have a hard time seeing reserves as s.t. assets.

      Delete
  11. Stephen, your QE claims so intrigued and frustrated me that I've recently spent over 40 concentrated, uninterrupted, serious study hours on Wallace 1981 AER. And it's extremely hard for me to make that kind of time (as opposed to little breaks here and there for blogging). I'm still working on it, but I've gone through it very carefully, line by line, equation by equation. I'll eventually have a lot to say about it.

    What I would like to ask now is, you've said that if the government adds, or exchanges for, currency, that can make a difference, because currency is something the private sector cannot create; only the government can create this unique asset. But isn't that also true for 30 year treasury bonds? No private sector party can create something that's that close to this, because no private sector party can make a 30 year claim very close to that risk free.

    Also regarding Wallace, AER, 1981, I think there are some significant errors in the paper:

    1) On page 268, second paragraph, the expression K(t)x/(t+1) + xY(t+1) should be K(t)x(t+1) + Y(t+1).

    2) In equation 2, the two utility functions should be partial derivatives of the utility functions with respect to consumption in states i and j respectively.

    Do you agree?

    ReplyDelete
  12. Wallace's paper is just an example of a particular neutrality theorem. It's very strong, and doesn't really have much to do with QE in the current circumstances. You'll learn something from the paper, but it's not the key to what I'm talking about.

    If you think there are errors in Neil's paper, you should check with him.

    ReplyDelete
    Replies
    1. Stephen,

      Really those are good questions to be answered, to educate us; they're important to know – What are then the main papers you base your QE ineffectiveness conclusion on, so we can read them, and if a currency swap is different because only the government can create currency, then why is it not the same with 30 year T-bonds, which are pretty unique to the government too.

      Delete
  13. "Wallace's paper is just an example of a particular neutrality theorem. It's very strong, and doesn't really have much to do with QE in the current circumstances."

    Oy, now he tells me, after I've used up my discretionary time for the next year! Ok, what are, then, the main papers you use to base your claim on that QE can't be effective, and that Fed buying of long-term treasury bonds, or even gold, won't move their price?

    And, you didn't answer my question; if, as you've asserted before, a swap for currency can make a difference, because currency is a unique asset that the private sector cannot create, then why can't a swap involving 30 year treasury bonds make a difference, as 30 year treasury bonds are a unique asset that the private sector cannot create. No private sector party can create an asset that comes that close to that level of risk-free'ness over a period as long as 30 years.

    ReplyDelete
    Replies
    1. "No private sector party can create an asset that comes that close to that level of risk-free'ness."

      Not sure I would go that far. After all, U.S. Treasuries are no longer rated AAA. Besides, I'm not sure the relative safety of the asset being swapped affects anything other than its price. I think you misunderstand the "uniqueness" concept. An asset convertible into currency is unique. All other assets are just that: assets.

      Delete
    2. Aren't you be worried that the debate over QE has ignored the vastly-important role for positional externalities?

      Delete
    3. Yes, I know I'm ridiculous to think that positional externalities are important for human utiltiy. Let's see, last time I quoted Nobel Prize winner Gary Becker, who's also ridiculous:

      "Traditionally, utility functions were assumed to depend on only the absolute level of current consumption, with no reference to past and peer levels. However, a large body of work now shows that our tastes are strongly influenced by our personal histories and social environment (e.g., Robert H. Frank 1985; Abel 1990; Becker 1996; and Brock and Durlauf 2001). Indeed, the explanatory power of economic models is greatly enhanced when including simple forms of personal and social capital in the utility function. In this way, otherwise puzzling phenomena-such as social influences on price, persistent habitual behavior, and neighborhood segregation-can be better accounted for."

      At: http://ideas.repec.org/a/aea/aecrev/v97y2007i2p487-491.html

      This time I think I'll add University of Michigan economist Miles Kimball. You may remember the name. Stephen said of him,

      "Miles Kimball may have been the first New Monetarist (actually, if you read the paper, he may have been the first New-Keynesian; he's basically outlining the basic NK model in 1995). Fortunately for us, Miles wants to blog, which is guaranteed to increase the average quality of discourse in the medium."

      At: http://newmonetarism.blogspot.com/2012/06/more-on-unconventional-open-market.html

      Here's what he had to say:

      "Interpersonal comparisons in the utility function are extremely interesting and indeed do provide the single strongest argument for high levels of taxation."

      At: http://blog.supplysideliberal.com/post/25205221766/avoiding-fiscal-armageddon#comment-560791474

      Delete
    4. "Yes, I know I'm ridiculous to think that positional externalities are important for human utiltiy."

      At least you're finally admitting it.

      Delete
  14. Dude:

    I'm a Randy Wright groupie too!

    ReplyDelete
  15. Great quote: "New Keynesianism...How do we write down a framework that justifies the status quo?"

    Love it.

    I make a similar point, less eloquently, in my post about Keynesianism more generally here:

    http://www.insofisma.com/wp2/the-long-run-is-here-keynes-is-dead/

    (The Long Run is Here, Keynes is Dead).

    ReplyDelete
  16. Stephen Williamson,

    If QE is as ineffective as you claim, why do asset markets seem to respond so strongly to announcements that would seem to raise QE expectations? Do you believe asset markets are responding as I claim?

    ReplyDelete
  17. "If QE is as ineffective as you claim, why do asset markets seem to respond so strongly to announcements that would seem to raise QE expectations?"

    Per Steven Williamson, asset markets respond but because QE has an expectational component, ergo it does not work (or something along these lines).

    ReplyDelete
    Replies
    1. So, markets have been wrong multiple times now when expecting QE to have some stimulative effect? At what point do we begin to consider markets stupid? And if markets are this stupid, why aren't those claiming QE doesn't work reporting making a killing in the market as a result?

      Delete