Two hawks and a Kotcherlakota

I see a lot of people referring to the three hawks who dissented from the Fed decision to promise two years of near-zero rates.  And yet can we really assume they are all inflation hawks?  Perhaps Narayana Kocherlakota dissented because he thought low rates were contractionary.  After all, didn’t Kocherlakota make this statement last August?

If the FOMC hews too closely to conventional thinking, it might be inclined to keep its target rate low. That kind of reaction would simply re-enforce the deflationary expectations and lead to many years of deflation.

And didn’t mild-mannered Nick Rowe respond as follows?

That speech by Narayana Kocherlakota is really disturbing. This guy is a top macroeconomist, and he totally f***s it up. I mean totally. It wasn’t just misspeaking, because he is quite clear the second time he makes the mistake. If the natural rate of interest rises exogenously, and the Fed doesn’t raise the nominal rate in response, the result will be….DEflation! And he’s a Fed President (so presumably this guy has some sort of power over monetary policy?).

You guys in the US are so scr***d. (And maybe we up here are too, since you are so big, even though we’ve got flexible exchange rates).

He went straight from a math undergrad into a PhD. I bet that’s the problem. He missed Intro Economics. (And he has the nerve to cr*p on Intro Economics too).

These are the people who will determine the fate of the world economy over the next few years.  As Arnold Kling likes to say; “Have a nice day.”

PS.  In the comment section feel free to explore the avian world for suitable symbols of Kocherlakotian monetary theory.

PPS.  I couldn’t find the Nick Rowe link, but Paul Krugman found a similar quotation by Nick.



28 Responses to “Two hawks and a Kotcherlakota”

  1. Gravatar of Morgan Warstler Morgan Warstler
    10. August 2011 at 11:29


    “Let me turn next to the issue of the Fed’s balance sheet. The Federal Reserve has 2.3 trillion dollars of assets””over 2.5 times what it owned in September 2008. Over 2 trillion dollars of those assets are in Treasuries, debt issued by Fannie Mae and Freddie Mac, or mortgage-backed securities issued by Fannie Mae and Freddie Mac. These MBSs are not toxic assets in any sense of the word””they are fully backed by the U.S. government, and so the Federal Reserve faces no credit risk in holding them. But the MBSs do have another kind of risk called prepayment risk. If long-term interest rates are low, then many people prepay the mortgages in the MBS. The owners of the MBS””in this case, the Fed””get a large coupon payment and the MBS’s principal falls. However, if long-term interest rates are high, then few people make these prepayments.

    This kind of fluctuation in prepayments is at the heart of the FOMC’s new policy action in August. Long-term interest rates declined surprisingly fast in the past three months. But the fall in long-term rates meant that more people were prepaying their mortgages, and the Fed’s MBS principal balances were falling. In this sense, the Fed’s holdings of long-term assets were shrinking, leaving a larger share of the long-term risk in the economy in the hands of the private sector. This extra risk in private hands could force up the risk premia on long-term bonds and be a drag on the real economy. The FOMC decided to arrest the decline in its holdings of long-term assets by re-investing the principal payments from the MBSs into long-term Treasuries.
    The FOMC’s decision has had a larger impact on financial markets than I would have anticipated. My own interpretation is that the FOMC action led investors to believe that the economic situation in the United States was worse than they, the investors, had imagined. In my view, this reaction is unwarranted. The FOMC’s decisions were largely predicated on publicly available data about real GDP, its various components, unemployment, and inflation. I would say that there is no new information about the current state of the economy to be learned from the FOMC’s actions or its statement.”

    I don’t know, this sounds like a guy who is convinced things will be fine if rates are allowed to appreciate and suffering is no longer put off.

  2. Gravatar of Lars Christensen Lars Christensen
    10. August 2011 at 11:33

    Did Nick really write that??? haha…

  3. Gravatar of John Paul Lewicke John Paul Lewicke
    10. August 2011 at 11:45

    Scott, Nick Rowe actually wrote that in your own comments section. See .

  4. Gravatar of George Selgin George Selgin
    10. August 2011 at 11:57

    Nick has got his Wicksell down right, far as I can tell; which means that K. (too hard to spell it) is indeed…well, choose your own phrase. The concern ought to be, as I indicated in a comment to a previous blog, that natural rates eventually rise from their now unusually low levels, and that the Fed’s “peg” therefore ends up resulting in an inflationary “cumulative process.” If the “Hawks” who dissented dissented on account of this fear, they were in my opinion perfectly right in doing so.

  5. Gravatar of Lars Christensen Lars Christensen
    10. August 2011 at 12:11

    George, what policy response would you suggest right now? Free Banking is the right answer, but you are not allowed to say that;-)

    If NGDP targeting – how would you suggest it should be implemented?

  6. Gravatar of amv amv
    10. August 2011 at 13:52

    @ Selgin

    Wicksell’s model is based on flexible prices and backward-looking agents (adaptive expectations). His cumulative process is very much a result of the latter assumption, that is, it is not robust for the RE framework that K. implicitly assumes. K. also talks about long run yields; he correctly employs the neutrality postulate (as Hayek did, right?). If we follow Wicksell and assume flexible prices, rational expectations suggests neutrality along the entire yield curve. Then, the Fisher eq. given, higher nominal yields suggest higher inflation expectations irrespective of terms to maturity. Even if frictions account for short run non-neutrality, as allowed for by K. and NK models, central banks essentially control nominal values in the medium and long run.

    In short, it makes little sense to compare Wicksell and K. without talking about the different views on expectations. Rational expectations preclude Wicksell’s cumulative processes.

  7. Gravatar of Wonks Anonymous Wonks Anonymous
    10. August 2011 at 14:02

    I asked Kocherlakota fan Steve Williamson about the quasi-monetarists. His response:
    “On quasi-monetarists: They have some similar ideas about the role of assets in financial trade, though they use a different language. However, it seems to me that nominal GDP targeting has to operationally be a specific class of Taylor rule, i.e. they are not proposing anything fundamentally different from what a New Keynesian would think about.”

  8. Gravatar of Martin Martin
    10. August 2011 at 14:22

    @amv, a silly question perhaps, but if the Fed ‘sets’ the rate, doesn’t this then reflect the expectation of the Fed of higher inflation in the future? And doesn’t the argument for higher rates then reduce to pulling up the economy by its own bootstraps?

    It seems that in K.’s world hyperinflation is caused by setting the rates at 1 * 10^6.

  9. Gravatar of George Selgin George Selgin
    10. August 2011 at 15:27

    “Rational expectations preclude Wicksell’s cumulative processes.”

    I don’t by it for a second. The gist of Wicksell’s argument, that persistent r<r* means persistent upward movement in P*, doesn't depend on adaptive expectations.

    For finite M expansion, its true, under RE there can be no "Wicksell effect," because credit demand and supply rise in unison: the increase raises P at once, without lowering r (or without lowering it below r*). But that's not what we have in this case. Here the Fed doesn't target M and thereby hope to lower r despite r*. It targets r itself, letting M grow "as much as necessary" to keep the target. Wicksell's point was that "as much as necessary" might mean "without limit." Under RE, the "without limit" possibility becomes all the more likely, for there's no finite rate of M expansion that can keep r from rising to meet r*.

    The argument can also be stated in terms of the instability of the particular RE equilibrium K. has in mind. But rather then make it that way I refer to another blog citing relevant works. I also recommend Allin Cottrell’s excellent paper, in which he observes how the issue in question “illustrates in a vivid manner the potential dangers of an over-literal application of the rational-expectations hypothesis.”

  10. Gravatar of George Selgin George Selgin
    10. August 2011 at 15:30

    Sorry: “by” should be “buy,” and the last link is to Peter Howitt’s paper, referred to in the aforementioned blog.

  11. Gravatar of John hall John hall
    10. August 2011 at 15:33

    The Fed statement says:
    “Voting against the action were: Richard W. Fisher, Narayana Kocherlakota, and Charles I. Plosser, who would have preferred to continue to describe economic conditions as likely to warrant exceptionally low levels for the federal funds rate for an extended period.”

    I originally read that as putting him in the hawk camp, but I suppose I see your point that “economic conditions” is actually rather vague and he could have some closet views that don’t make sense.

  12. Gravatar of Scott Sumner Scott Sumner
    10. August 2011 at 16:53

    Morgan, Even a broken clock . . .

    Lars, Yes, it was Nick.

    Thanks John.

    George, OK, but that’s not why the hawks dissented. They favor tighter monetary policy than what we have. They’d oppose QE3 as well. But your point is well taken.

    I agree with your comment about RE.

    Wonks, NKs favor interest rate targeting, quasi-monetarists don’t.

    John, You are probably right–this post was half serious, half tongue in cheek.

  13. Gravatar of TGGP TGGP
    10. August 2011 at 21:09

    A newer post from Williamson:
    “it seems the outcome the Fed would hope for is one where inflation increases, the interest rate on reserves increases commensurately, and the Fed proceeds to sell off assets so as to normalize the state of its balance sheet, with zero exess reserves. Committing to IROR = 0.25 for two years risks two outcomes that seem equally bad (if we believe that 2% inflation is optimal). One is the too-high-inflation outcome: […] The other is too-low-inflation: People anticipate low inflation, the reserves look more desirable, and low inflation is self-fulfilling. […] The second scenario was a concern of Jim Bullard, and Narayana Kocherlakota. I think both are possibilities, i.e. there are multiple equilibria.”

    One would think that simple inflation-targeting would be a way to reconcile the Fed “doves” with this side of Kocherlakota.

  14. Gravatar of amv amv
    10. August 2011 at 22:53

    George, I still disagree. you talk about persistent spreads of real rates. Of course, if such a spread occurs persistently, then your prediction follows. My point is that if you take Wicksells flexible prices and add RE instead of adaptive inflation, then you eliminate all persistemcy that could account for the banking systems ability to push real rates below the natural rate. Given the assumptions, central banks control nominal rates only. Assume a simple model made of a Fisher equation and a vertical Phillips curve. Such a model is unstable if agents cannot learn that simple model . Why should this be the case? Further, the link you provide just reinforces my point of view: if you accept modern macro, K.s case follows. If you dont accept modern macro, it is a different matter.

    Scott, I believed that you will side with me. After all, it is you who persistently warns us that low nominal yields indicate tight money. If you disagree with K., however, what then is your model?

  15. Gravatar of Nick Rowe Nick Rowe
    11. August 2011 at 02:51

    amv: take a model like you describe, except that some fraction b of agents have adaptive expectations, and/or some faction c of prices are sticky, then take the limit of the predictions of that model as b and c go to 0. That model will not approach your model in the limit.

    This result is (to me) intuitively obvious. Suppose intitially inflation is 0%, and has always been 0%. And that the natural rate is 3% and has always been 3%. And the Fed has always set the nominal rate at 3% too. Assume a fraction b agents have adaptive expectations, and a fraction c of prices adjust slowly. Let b=c=1 initially. Then assume the Fed sets the nominal rate at 4%, and holds it there forever. We know what will happen. Slowly-accelerating deflation. Now what happens as we lower b and c towards 0? Some agents, having rational expectations, expect deflation immediately. And the price level stats to fall more quickly. In the limit the price level explodes downwards.

    All models are false. But a model whose predictions stay roughly true if its assumptions stay roughly true has at least a chance of being roughly right. A model that has the exact opposite predictions *in* the limit as its assumptions come true as its predictions *at* the limit is totally useless.

    What caused my total despair (“Oh Christ, Oh Christ”) at what you call “modern Macro” is the realisation that really really smart people, like NK and SW, just don’t get stuff like this. They could prove it in math in about 5 minutes. I would struggle for weeks to get the math right. But it’s bloody obvious to anyone with any horse sense.

    And these are the people teaching the next generation of idiot savants. It’s not just the US that is screwed; it’s macro.

    Plus the realisation that while Scott and me are struggling to get people to re-frame monetary policy as setting something other than a nominal interest rate, so that nominal interest rates become endogenous rather than exogenous, and to trying to explain why it could help escape the ZLB, we suddenly have to backtrack, and join the opposition, because NK doesn’t understand the difference between raising the nominal interest rate and doing something that would cause the nominal interest rate to rise. Because it’s all the same in the bloody equations that define the equilibrium. And if all you can see are those equations, you just don’t get it.

  16. Gravatar of Nick Rowe Nick Rowe
    11. August 2011 at 03:00

    TGGP: re SW’s 2 equilibria.

    But one of those equilibria (the too-low equilibrium) will not be an equilibrium *in the limit*. It only appears *at the limit*.

  17. Gravatar of Left Outside Left Outside
    11. August 2011 at 03:13

    Surely he’s a Cuckoo, he’s smuggled his way into a nest of monetary policy experts, but doesn’t belong there.

    Or is that too cruel?

  18. Gravatar of amv amv
    11. August 2011 at 04:29

    @ Nick Rowe:

    I know that some economist favor DeGrauwe-type models where they mix adaptive and rational agents. I don’t like that because it confuses RE with an assumption on agents. RE, however, is a general equilibrium notion. Like in case of the EMH, which Sumner definitely favors (like I do), not all agent have to be rational (defined as optimal forecasters). In one of my favorite papers, Sandroni (Econometrica 2000) shows under what restrictions a convergence process sorts out non-rational (in our case adaptive) behavior by transfering their wealth to relatively better forecasters such that adaptive agents’ forecasts loose any influence over equilibrium prices. In equilibrium, that is, in REE, prices fully reflect information !AS IF! agents were all rational, that is, as if agents knew the true model. This is essentially Milton Friedman’s view on how impersonal markets sort out stupid decision-makers.

    Now, given such a convergence process that transfers all wealth to optimal forecasters such that adaptive behavior cannot express itself, we can rely on models predicting REE allocations. This, of course, is just a convention. But if you accept it, as idiot savants like me do, then the ancient neutrality postulates (that Sumner at least seems to accept) suggests K.’s point of view.

    BTW, what does c=1 mean? All prices are sticky in period t=1,2,…? But then higher interest rates (given the natural rate) cannot – by assumption – induce even the slightest deflation (as I mentioned above, Wicksell’s model assumes prefectly flexible prices and zero output gaps). If deflation is ruled out by c=1, then adaptive agents (b=1) have nothing to learn from (c=0,99 would, however, work). But I’m sure that I misunderstood your model.

  19. Gravatar of Nick Rowe Nick Rowe
    11. August 2011 at 05:20

    amv: c=1 could mean an old-fashioned non-expectations-augmented Phillips Curve. Stable long-run trade-off. Or it could mean Calvo-pricing. Or overlapping Taylor-Style contracts. Whatever. (Agreed, it wouldn’t work if c=1 meant prices never change.)

    I agree that RE is a system property. But we cannot assume an instant switch to RE when the regime changes. Just assume that the Darwininian process that kills off agents with Adaptive Expectations works slowly.

    If you have an RE model, where the DeGrauwe version of that model approaches the same predictions as the original model in the limit as mix approaches all RE, then that RE model is “robust”. If it approaches some totally different model in the limit, then that RE model is “fragile”. “Fragile” models suck. “Robust” models might be good models.

  20. Gravatar of Skepticlawyer » Postmodern Conservatism – guest post by Lorenzo Skepticlawyer » Postmodern Conservatism – guest post by Lorenzo
    11. August 2011 at 05:30

    […] ‘fighting the last war’ in monetary policy. Yet, in much of conservative US (reaching all the way to the Fed), showing how much of an ‘inflation hawk’ you are has become the path of righteousness. […]

  21. Gravatar of George Selgin George Selgin
    11. August 2011 at 06:33

    This thread is very important, so I hope that SS will forgive me by clogging it up with a longish reply to amv.

    I ver much hope, amv, that you and I can come to a better understanding of why we differ, if not to any actual agreement. So allow me to state some starting assumptions that I hope are not part of the disagreement.

    (1) the Fed has pegged the nominal FFR at 0-.25; let us call it “zero” for simplicity’s sake.

    (2) the commitment is to maintain the peg for 2 years, if not longer.

    (3) at present the natural or neutral nominal FFR is also either at 0 (if the Fed has got it just right) or somewhere below 0 (if Scott and other doves are right in claiming that policy remains too tight). Again for simplicity’s sake, let us assume that FFR*=0, so that we are in stationary equilibrium initially.

    (4) the Fed’s move is intended to raise, and certainly not to lower, inflationary expectations, which is to say that it is intended as a move toward easier rather than tighter monetary policy.

    For the sake of argument I will also concede the following points, the though I actually reject them for reasons I will eventually make clear:

    (5) RE holds, and prices are perfectly flexible; from which it follows that

    (6) there is no equilibrium in which actual and natural rates diverge;

    (7) for the right rate of (actual and expected) inflation or deflation, the nominal natural rate will conform with the pegged rate, that is, will be zero.

    Now, my claim is that, even under these conditions, should real FFR* increase while the peg remains in effect, the result will be inflation that continues so long as the peg remains in effect.

    How can this be the case, you will ask, if under the conditions described the Fed is incapable of making FFR and FFR* diverge persistently, and if the two can only remain equal by means of deflation sufficient to keep FFR*=0?

    The key to the answer is that inflation results, not because the Fed is able under the conditions given to keep FFR below FFR*, but precisely because it it is seeking to do the impossible, that is, keep FFR down by means of easy money. The consequence isn’t an inflationary “equilibrium” in which FFR remains < FFR*, for no such equilibrium exists; but neither is it K's deflationary equilibrium, despite the fact that that equilibrium does indeed exist, because that equilibrium isn’t consistent with the Fed’s actual monetary policy stance. The inflation is, instead, a disequilibrium phenomenon that only comes to a halt if the Fed gives up the peg.

    I understand that in making appeals to disequilibrium, I am rejecting “modern macro.” But I’m not bothered by that. I think it’s modern macro that has a lot to answer for, and that is proving untenable. Indeed, it seems to me that consistent appeal to “modern macro,” understood as implying RE + Walrasian prices + equilibrium doesn’t allow one to say very much about the proper conduct of M policy at all, for the imaginary world to which it refers is precisely one in which the sort of real outcomes that M policy is now struggling to combat (and which it has perhaps helped cause) simply don’t arise. Is it wise for people to take advise concerning the likely consequences of this or that M policy from persons relying on “modern” models according to which money is neutral, if not superneutral, even in the short tun?

    I’ll end by quoting here a comment my friend Jerry O’Driscoll made recently in connection with a related discussion at Coordination Problem:

    “I’ll end my contributions with a point frequently driven home by Fritz Machlup. Optimality is a property of models, not the real world.

    “Models may be applicable to real-world problems, but to know that you must have information beyond the model. Institutional knowledge is required.

    “This is important because we now have policy-making being conducted by people with little or no institutional knowledge. They only know the model. It is positively dangerous.”

  22. Gravatar of Scott Sumner Scott Sumner
    11. August 2011 at 06:52

    TGGP, Thanks. I’ll do a post on that.

    Nick and George, Thanks for those excellent explanations. I’m going to do a post on SW, which comes at his from a slightly different perspective.

    Left Outside, That’s what I was thinking. But I didn’t put in in the title because I thought it might be too cruel. One can call the President any name one wishes. But I don’t like to insult my fellow economists. The question is whether an economist at the Fed is important enough to be fair game like Obama, or a fellow economist who should always be treated with respect. Anyone have any thoughts?

  23. Gravatar of Left Outside Left Outside
    11. August 2011 at 07:32

    Yes, name calling has its place in politics, but in academia? That is a more complicated question.

    Whether or not it is fair or cruel you have to ask, “is this sensible?” Name calling can be easily dismissed as an ad hominem attack and your substantive argument can be ignored.

    If you want to convince the people you’re arguing with, or convince the people who agree with your opponents, it might not be the best tactic.

    If you want to discredit the people you’re arguing with and make it less likely for people to agree with them, then it may be a worthwhie tactic.

    If you’re on the FOMC are you an economist or a politcian? Congressional Republicans seem to be treating nominees like they’re politicians, not economists, in refusing to confirm eminently qualified people.

    Its a toughie. But a *insert policy subject* Cuckoo is a political phrase whose time has surely come.

    I tell you what. E-mail DeLong, Krugman and Thoma with the phrase and let them do the dirty work. You’ll keep your hands clean and they’ll get to enjoy themselves to boot.

  24. Gravatar of Riots, Recessions and Fear « Left Outside Riots, Recessions and Fear « Left Outside
    11. August 2011 at 10:10

    […] understand how to do his job, he has snuck into a nest of monetary policy experts, but definetely doesn’t belong there. Normally this would not matter, the Fed would err slightly from time to […]

  25. Gravatar of amv amv
    11. August 2011 at 13:04


    I answered over here:

    Thanks for your detailed response!

  26. Gravatar of Nick Rowe Nick Rowe
    11. August 2011 at 14:02

    George: I see where you are coming from. Why not say this instead:

    “When FFR* rises above FFR, the ***equilibrium*** rate of inflation rises, but that equilibrium is unstable. The economy will not move towards that equilibrium. Instead, it will more away from it. Actual inflation will fall.”

    (The trouble is, with perfectly flexible price, inflation won’t just *fall*; it will instantly collapse to minus infinity.)

    George, you are wanting to do Hayekian process analysis, and I sympathise. But Hayekian process analysis does not compute in a world where RE equilibrium analysis implicitly assumes the existence of a divine auctioneer who coordinates individuals’ plans and expectations instantly. That’s the world of “modern macro”, God help us.

    Any model with perfect price flexibility, RE, superneutrality, and the Fed pegging the nominal interest rate, is a really, really daft model. The equilibrium exists, but is totally unstable. It’s explosively unstable. It’s a knife-edge with a precipice either side. And “modern macro” is so utterly corrupt that it can’t even make sense of what I have just said.

  27. Gravatar of George Selgin George Selgin
    11. August 2011 at 14:24

    George: I see where you are coming from. Why not say this instead:

    “When FFR* rises above FFR, the ***equilibrium*** rate of inflation rises, but that equilibrium is unstable. The economy will not move towards that equilibrium. Instead, it will more away from it. Actual inflation will fall.”

    (The trouble is, with perfectly flexible price, inflation won’t just *fall*; it will instantly collapse to minus infinity.)

    Well, I agree with the first restatement of my position. But I disagree that actual inflation must drop. I don;t think my position is due to Hayekian thinking as such. Rather (as I indicate on Scott’s other post) I think it has to do with the way in which the Fed responds to the increased FFR. That is, in my view the response is likely to be one that will actually raise the equilibrium inflation rate still further.

    The question is, what instructions does the Fed send to the open-market desk when FFR* (and equilibrium inflation) rise. I’ve said that I believe that if the instruction is, buy more bonds, actual inflation will increase. I recognize know that you doubt this to be the case in the moddel world so described, presumably because changing the money-bond portfolio mix doesn’t change people’s demand (because it doesn’t change the effective supply of exchange media). So, let’s see if we can narrow down the disagreement a bit more: suppose the Fed abandons its “Treasuries only” stand (not a bad idea, I say), thereby changing its rule for implementing the peg to, “buy more securities of whatever sort whenever the rate increases.” Would you agree that this rules out deflation? How about a “buy goods so long as FFR is above target”? Would that do it?

    I’m not being provocative. I’m trying to see where the real disagreement resides. I suspect we are honingin on it.

  28. Gravatar of Scott Sumner Scott Sumner
    11. August 2011 at 18:39

    Left outside, Good analysis.

    Nick and George, For some reason I am having trouble figuring out what you guys are debating.

Leave a Reply