I’m pretty late getting to this Wall Street Journal op-ed by John Cochrane (here’s an ungated version), and Noah Smith has already given it an admirable working over, but, even after Noah Smith, there’s an assertion or two by Cochrane that could use a bit of elucidation. Like this one:
Keynesians told us that once interest rates got stuck at or near zero, economies would fall into a deflationary spiral. Deflation would lower demand, causing more deflation, and so on.
Noah seems to think this is a good point, but I guess that I am less easily impressed than Noah. Feeling no need to provide citations for the views he attributes to Keynesians, Cochrane does not bother either to tell us which Keynesian has asserted that the zero lower bound creates the danger of a deflationary spiral, though in a previous blog post, Cochrane does provide a number of statements by Paul Krugman (who I guess qualifies as the default representative of all Keynesians) about the danger of a deflationary spiral. Interestingly all but one of these quotations were from 2009 when, in the wake of the fall 2008 financial crisis, a nasty little relapse in early 2009 having driven the stock market to a 12-year low, the Fed finally launched its first round of quantitative easing, the threat of a deflationary spiral did not seem at all remote.
Now an internet search shows that Krugman does have a model showing that a downward deflationary spiral is possible at the zero lower bound. I would just note, for the record, that Earl Thompson, in an unpublished 1976 paper, derived a similar result from an aggregate model based on a neo-classical aggregate production function with the Keynesian expenditure functions (through application of Walras’s Law) excluded. So what’s Keynes got to do with it?
But even more remarkable is that the most famous model of a deflationary downward spiral was constructed not by a Keynesian, but by the grandfather of modern Monetarism, Irving Fisher, in his famous 1933 paper on debt deflation, “The Debt-Deflation Theory of Great Depressions.” So the suggestion that there is something uniquely Keynesian about a downward deflationary spiral at the zero lower bound is simply not credible.
Cochrane also believes that because inflation has stabilized at very low levels, slow growth cannot be blamed on insufficient aggregate demand.
Zero interest rates and low inflation turn out to be quite a stable state, even in Japan. Yes, Japan is growing more slowly than one might wish, but with 3.5% unemployment and no deflationary spiral, it’s hard to blame slow growth on lack of “demand.”
Except that, since 2009 when the threat of a downward deflationary spiral seemed more visibly on the horizon than it does now, Krugman has consistently argued that, at the zero lower bound, chronic stagnation and underemployment are perfectly capable of coexisting with a positive rate of inflation. So it’s not clear why Cochrane thinks the coincidence of low inflation and sluggish economic growth for five years since the end of the 2008-09 downturn somehow refutes Krugman’s diagnosis of what has been ailing the economy in recent years.
And, again, what’s even more interesting is that the proposition that there can be insufficient aggregate demand, even with positive inflation, follows directly from the Fisher equation, of which Cochrane claims to be a fervent devotee. After all, if the real rate of interest is negative, then the Fisher equation tells us that the equilibrium expected rate of inflation cannot be less than the absolute value of the real rate of interest. So if, at the zero lower bound, the real rate of interest is minus 1%, then the equilibrium expected rate of inflation is 1%, and if the actual rate of inflation equals the equilibrium expected rate, then the economy, even if it is operating at less than full employment and less than its potential output, may be in a state of macroeconomic equilibrium. And it may not be possible to escape from that low-level equilibrium and increase output and employment without a burst of unexpected inflation, providing a self-sustaining stimulus to economic growth, thereby moving the economy to a higher-level equilibrium with a higher real rate of interest than the rate corresponding to lower-level equilibrium. If I am not mistaken, Roger Farmer has been making an argument along these lines.
Given the close correspondence between the Keynesian and Fisherian analyses of what happens in the neighborhood of the zero lower bound, I am really curious to know what part of the Fisherian analysis Cochrane finds difficult to comprehend.
Uneasy Money 
David, very good post! I must say when I read Cochrane I get the feeling that is just recently has discovered monetary theory and I am not even sure he really read Fischer or for that matter Milton Friedman or the young Keynes.
And you are of course absolutely right about the fact there is not “keynesian” about the debt-deflationary risks at the Zero Lower Bound.
What I miss from Cochrane is also answer to whether there is a liquidity trap or not. To me therefore of course is only a “mental liquidity trap” where central bankers chose not to ease monetary policy at the ZLB because they for some reason are uncomfortable with using the money base as a policy instrument.
Furthermore, it is bizarre that Cochrane claims to be a Fischerian and then at the same time dismiss the dangers of deflation. If anybody understood these dangers it was of course Irving Fischer (Read the book Cochrane!)
LikeLike
David: I am one of those “Keynesians” who was scared of the dangers of a deflationary spiral. What stopped it, I think, was the refusal of central banks to let base money fall, and their doing the exact opposite (“QE”) making base money rise.
But I still don’t properly understand why inflation didn’t fall further than it did, given the amount of unemployment and apparent excess supply of labour and goods. There are a number of theories, but none is wholly convincing to my eyes. The most plausible theory, I think, is that inflation targeting over the previous years had made expected inflation very sticky. Unlike the 1930’s.
I am with you in this post up until your paragraph that begins “And, again…” I do not follow you in that paragraph, but it sounds to me that you are saying something very similar to what the Neo-Fisherites are saying. But we can’t reverse the causality in the Fisher equation. The Fisher equation tells us that if the central bank create an increase in inflation, the nominal interest rate will rise one-for-one. But we can’t use the Fisher equation to tell us what determines inflation. And there has to be some sort of Phillips Curve out there somewhere, And it should be roughly vertical, once expected inflation adjusts to actual inflation. So if there’s high unemployment, and it persists, why doesn’t inflation eventually fall without limit?
LikeLike
Lars, Yes, I think David Laidler would be appalled.
Nick, I understand and to some extent share your puzzlement that there was not more deflation in the wake of the financial crisis. But I wouldn’t necessarily attribute the difference between the Great Depression and the Little Depression to greater price stickiness. It seems to me that it’s necessary to distinguish between equilibrium deflation (or inflation) and disequilibrium deflation (or inflation). The lower bound on equilibrium inflation is given, according to the Fisher equaiton, by the real rate of interest. But there can be disequilibrium deflation, precipitated by some monetary policy shock or an appreciation of gold under the gold standard, where prices have to readjust and measured inflation in the transition is greater than the equilibrium rate of deflation. In the Little Depression, the monetary authorities were able to limit the extent of disequilibrium deflation by pursuing an accommodative policy so that inflation stayed close to the inflation target for most of the episode. That’s why I was careful to speak about “equilibrium” expected inflation in the paragraph that’s giving you trouble. Does this help you at all?
LikeLike
Lars, Sorry to be picky (well, I guess I’m not, or I wouldn’t be picky), but it’s Fisher. No c.
LikeLike
in chapter 18 (of the general theory) keynes describes scenarios where economies function long time below full potential
which we would now call secular stagnation
at the end of chapter 18 he says this
“But we must not conclude that the mean position thus determined by “natural” tendencies, namely, by those tendencies which are likely to persist, failing measures expressly designed to correct them, is, therefore, established by laws of necessity. The unimpeded rule of the above conditions is a fact of observation concerning the world as it is or has been, and not a necessary principle which cannot be changed.”
LikeLike
David,
Fisher explicitly lays out that:
“When a deflation occurs from other than debt causes and without any great volume of debt, the resulting evils are much less. It is the combination of both – the debt disease coming first, then precipitating the dollar disease – which works the greatest havoc.”
The end of Fisher’s debt-deflation scenario (or spiral) occurs whenever the real cost of servicing debt falls or the debt level falls to a manageable level (through repayment, liquidation, bankruptcy, or default). Once the liquidations / bankruptcies stop happening, the “spiral” if you want to call it that, is over. Bottom line – the debt level matters just as much as the interest rate.
Was it the rise in interest rate or the rise in leverage ratio that did in Lehman, etc.?
From the article on MIT’s website:
“The output gap feeds expectations of deflation, and since the nominal interest rate cannot fall this implies a rising real interest rate, worsening the output gap.”
A lot of things wrong with this statement but a rising output gap means that fewer goods are being produced and fewer goods are being sold. That does not “feed” expectations of deflation in the slightest. The output gap is a measurement of how many goods are both produced and consumed.
No mention of debt in this article at all. No mention that the output gap rising spiral only works with infinite life agents.
LikeLike
I haven’t personally read any of Fisher’s stuff, and I’m encouraged to do so after reading this post (I don’t get the Neo-Fisherites, in any case). I’m reading right now Hyman Minsky’s Stabilizing an Unstable Economy, and one of the things he mentions is how automatic deficit spending in a downturn ‘rigs the game’ for big business, propping up falling profits as the economy is entering a recession. While we certainly didn’t spend on the scale that I would have liked (being of the Keynesian persuasion myself), the $1 trillion annual deficits we were running were certainly nothing to snicker at in a $15 trillion economy: I’m curious if this additional spending coupled with more people using their incomes to pay down debt in the course of the Great Recession could explain the low but stable inflation we saw. Do business profits over the last several years support this view?
Other thoughts that come to mind: Could it simply be that lackluster expectations by consumers and businesses fed off of each other, leading to the suboptimal equilibrium that we’ve been stuck in? Perhaps changes in productivity, induced by the recession, have sustained prices without leading to an increase in employment? Or could it be, as Krugman and Bernanke seem to suggest, that a stagnant housing market has acted as a persistent weight on an otherwise healthy economy?
The persistent, low inflation is definitely one of those puzzles I’ve been scratching my head about.
LikeLike
changes in productivity
increases in productivity could keep employment lower but i would suspect if would decrease not increase prices, unless it led to monopoly
decrease in productivity should case increased employment
LikeLike
When Cochrane says he is Fisherian, he means Richard Fisher, the Dallas Fed President and FOMC board member who also seeks mild deflation as an economic cure-all.
LikeLike
djb, Thanks for the quotation. Keynes was not a fatalist.
Frank, Lehman failed, in my not particularly well-informed view, because it overinvested in mortgage backed securities which started to lose value fast in 2007, so its downfall was not due to a classic debt-deflation process, though there are similarities.
Not sure what your definition of output gap is. My definition, and I think Krugman’s, is the difference between actual output and potential output. In Krugman’s model, the rate of change in prices is negatively correlated with the output gap.Krugman’s paper is not a debt-deflation paper, it is a model of deflation. There is nothing wrong with having a model of deflation in which debt does not have a role. Even though Fisher showed that debt may have a central role in some deflations, I don’t think he asserted that there can’t be a deflation when the level of debt is low.
Matt, There has been a lot written about balance-sheet recessions, in which households and businesses that are more indebted than they want to be cut back spending as they pay down their debt. However, when everybody tries to pay down debt at the same time, output and income fall, so total debt goes down more slowly than income, which only makes people feel more overindebted, causing spending to contract even more.
Benjamin, Interesting thought, but, seriously, what sane person, let alone a Ph.D. economist, and a Chicago economist at that, would choose to associate himself with Richard Fisher instead of Irving Fisher? But that is exactly why I made a parenthetical insertion in the title of this post.
LikeLike
David-
OK, but I asked Cochrane how does the Volcker success against inflation fit the Neo Fisher model.
Cochrane replied that Reagan and Volcker had a coordinated fiscal and monetary policy and that Reagan actually ran operating or primary surplus budgets.
However, when I checked the figures, the history is that Reagan’s deficits were so large that even his primary or operating budget was in large deficit in the years that Volcker beat inflation.
Cochran never replied.
I still cannot understand the Fisherian explanation of Volcker’s success.
LikeLike
Benjamin, Well, good for you. It’s always inconvenient to have someone actually check your “facts.” No wonder he didn’t want to continue that conversation. Robert Eisner, a pretty orthodox Keynesian, argued that if you counted the government’s seignorage revenue, the government was running a surplus, so that it was fiscal policy that was tight and was responsible for bringing inflation down.
LikeLike
David,
Your statement,
“Even though Fisher showed that debt may have a central role in some deflations, I don’t think he asserted that there can’t be a deflation when the level of debt is low.”
But Fisher did assert that deflations in conjunction with low / zero debt levels are less corrosive to economic growth than deflations with high debt levels.
“When a deflation occurs from other than debt causes and without any great volume of debt, the resulting evils are much less.”
The reasoning is fairly straight forward – If a farmer with no debt receives less money for his goods than he expects in one year, lesson learned, he plants other crops or leaves his land fallow the next year. The farmer in debt can lose his farm after a year of deflation.
And so, taking Fisher’s words to be true (being an Irving Fisherian), it seems reticent for economists to decry deflation as an evil in and of itself or omit debt from any discussion of a deflationary spiral (http://web.mit.edu/krugman/www/spiral.html).
LikeLike
Paraphrasing Fisher – debt was the disease, deflation made it worse.
LikeLike
Frank, No one, including Krugman, denies that Fisher made a seminal contribution in his debt-deflation theory. The fact that in a specific paper, he went through an analysis of deflation without including a discussion of debt doesn’t mean that he was in any way disputing Fisher.
LikeLike