Must-Read: Paul Krugman: Rethinking Japan

Must-Read: Paul Krugman is musing about and rethinking his 1998 analysis of Japan and its macroeconomic problems:


Paul Krugman: Rethinking Japan: “[How] would change what I said in my 1998 paper…

…on the liquidity trap[?]… Japan and the world look different…. First, the immediate economic problem is… weaning the economy off fiscal support. Second… demand weakness looks… permanent…. Back in 1998 Japan… [was] operating far below potential…. This is, however, no longer the case…. Output per working-age adult has grown faster than in the United States since around 2000…. [But] Japan’s relatively healthy output and employment levels depend on continuing fiscal support… ever-rising debt/GDP…. So far this hasn’t caused any problems…. But even those of us who believe that the risks of deficits have been wildly exaggerated would like to see the debt ratio stabilized and brought down at some point…..

With policy rates stuck at zero, Japan has no ability to offset the effects of fiscal retrenchment with monetary expansion. The big reason to raise inflation… is to… allow… monetary policy to take over from fiscal policy…. But what would it take to raise inflation?… Back in 1998… [I assumed] the Wicksellian natural rate of interest… would return to a normal, positive level at some future date… [thus] the liquidity trap became an expectations problem…. But what is this future… normality[?]… [If] a negative Wicksellian rate is… permanent… [even] a credible promise to be irresponsible might do nothing: if nobody believes that inflation will rise, it won’t….

The only way to be at all sure… is… a changed monetary regime with a burst of fiscal stimulus…. While the goal… is… to make space for fiscal consolidation, the first part of that strategy needs to involve fiscal expansion… really aggressive policy, using fiscal and monetary policy to boost inflation… setting the target high enough… escape velocity. And while Abenomics has been a favorable surprise, it’s far from clear that it’s aggressive enough to get there.

In many ways, things are even worse than Paul says. Paul Krugman’s original argument assuming that the economy would eventually head towards a long-run equilibrium in which flexible wages and prices what make Say’s Law hold true, in which there would be a positive natural nominal rate of interest, and thus in which the price level would be proportional to the money stock. That now looks up for grabs. It is the fact that that is up for grabs that currently disturbs Paul. Without a full-employment Say’s Law equilibrium out there in the transversality condition to which the present day is anchored by intertemporal financial-market and intertemporal consumer-utility arbitrage, all the neat little mathematical tricks that Paul and Olivier Blanchard built up at the end of the 1970s to solve for *the* current equilibrium break in their hands. And we enter Roger Farmer-world–a scary and frightening place.

But there is even more. Paul Krugman’s original argument also assumed back-propagation into the present via financial-market intertemporal arbitrage and consumer-satisfaction intertemporal utility arbitrage of the effects of that future well-behaved full-employment equilibrium. The equilibrium has to be there. And the intertemporal arbitrage mechanisms have to work. Both have to do their thing.

But, as Paul wrote in another context:

Paul Krugman**: Multipliers and Reality: “Rigorous intertemporal thinking…

…even if empirically ungrounded, can be useful to focus one’s thoughts. But as a way to think about the reality of spending decisions, no…. Consider… what the public knows about the biggest new government program of recent years[, ObamaCare]…. If people are that uninformed about something that big, imagining that they do anything like the calculations assumed in DSGE models is ludicrous. Surely they rely on rules of thumb that don’t make use of the kind of information that plays such a large role in our models…

Suppose the full employment equilibrium is really out there. People still have to anticipate that it is out there, and then take account of the fact that it is out there and the way that a rational-expectations utility-maximizing agent would.

Now sometimes we get lucky.

Sometimes the fact that one can transact on financial markets on a large scale means that even if only a few are willing to bet on fundamentals, the fact that they can make huge fortunes betting on fundamentals on a large scale drives current asset prices to fundamental values, and those asset prices then drive the current behavior even those who do not know anything about the future equilibrium that is driving the present via this process of expectational back-propagation inductive-unraveling.

But when we are talking about inducing people to spend more now because they fear their money will be worth less then in the future when the debt will have been monetized–well, if that were an important and active channel, we would not now have our current sub-2%/year inflation in Japan and the United States, would we?

Time to drop a link to my New Economic Thinking, Hicks-Hansen-Wicksell Macro, and Blocking the Back Propagation Induction-Unraveling from the Long Run Omega Point: The Honest Broker for the Week of May 31, 2015.

In fact, let me just repeat the whole thing below the fold…

Over at Equitable Growth: In the long run… when the storm is long past, the ocean is flat again.

At that time–or, rather, in that logical state to which the economy will converge if values of future shocks are set to zero–expected inflation will be constant at about the 2% per year that the Federal Reserve has announced as its target. At that time the short-term safe nominal rate of interest will be equal to that 2% per year of expected inflation, plus the real profits on marginal investments, minus a rate-of-return discount because short-term government bonds are safe and liquid. At that time the money multiplier will be a reasonable and a reasonably stable value. At that time the velocity of money will be a reasonable and a reasonably stable value. Why? Because of the powerful incentive to economize on cash holdings provided by the the sacrifice of several percent per year incurred by keeping cash in your wallet rather than in bonds. And at that time the price level will be proportional to the monetary base. READ MOAR

That was and is the logic behind so many economists’ beliefs. Their beliefs before 2008 that economies could not get stuck in liquidity traps (because central banks could always create inflation by boosting the monetary base); beliefs in 2008 and 2009 that economies’ stays in liquidity traps would be very short (because central banks were then boosting the monetary base); and beliefs since then that (because central banks had boosted the monetary base) those who believe will not taste death before, but will live to see exit from the liquidity trap and an outburst of inflation as the Federal Reserve tries and fails at the impossible task of shrinking its balance sheet to normal without inflation–all of these beliefs hinged and hinge on a firm and faithful expectation that this long run is at hand, or is near, or will soon draw near (translations from the original koine texts differ). Because the long run will come, increases now in the monetary base of sufficient magnitude that are believed to be permanent will–maybe not now, but soon, and for the rest of our lives, in this long run–produce equal proportional increases in the price level, and thus substantial jumps in the inflation rate as the price level transits from its current to its long-run level.

Moreover, there is more to the argument: The long run is not here. The long run may not be coming soon. But the long run will come. And so there will will a time when the long run is near. At that time, those who are short long-term bonds will be about to make fortunes as interest rates normalize and long bond prices revert to normal valuation ratios. At that time, those who are leveraged and short nominal debt will be about to make fortunes as the real value of their debt is heavily eroded by the forthcoming jump in the price level .

And there is still another step in the argument: When the long run is near but not yet here–call it the late medium run–investors and speculators will smell the coffee. This late medium run will see investors and speculators frantically dumping their long bonds so as not to be caught out as interest rates spike and bond prices collapse. It will see investors and speculators frantically borrowing in nominal terms to buy real assets and currently-produced goods and services so as not to be caught out when the price level jumps. Thus even before the long run is here–even in the late medium run–their will already be very powerful supply-and-demand forces at work. Those forces will be pushing interest-rates up, pushing real spending levels, and pushing price levels and inflation rates up.

The next step in the argument continues the induction unraveling: When it is not yet the late medium run but only the medium run proper, rational investors and speculators must still factor the future coming of the long run into their decisions. The long run may not be near. But it may be that soon markets will conclude the long run is near. Thus in the medium run none will want their portfolios to be so imbalanced that when the late medium run does come and with it the time to end your exposure to long-term bonds and to nominal assets and leverage up, you are on the wrong foot and so last person trying to get through the door in the stampede. There may be some short run logic that keeps real spending low, prices low, inflation quiescent, and interest rates at zero. But that logic’s effects will be severely attenuated when the medium run comes, for then investors and speculators will be planning not yet for the long run or even the at-handness of the long run, but for the approach of the approach of the long run.

And so we get to the final step of the induction-unraveling: Whatever may be going on in the short run must thus be transitory in duration, moderate in their effects, and limited in the distance it can can push the economy away from its proper long run equilibrium. And it certainly cannot keep it there. Not for long.

This is the real critique of Paul Krugman’s “depression economics”. Paul can draw his Hicksian IS-LM diagrams of an economy stuck in a liquidity trap:

The Inflationista Puzzle NYTimes com

He can draw his Wicksellian I=S diagrams of how the zero lower bound forces the market interest rate above the natural interest rate at which planned investment balances savings that would be expected were the economy at full employment:

Lifestream vpdoc 2015 06 02 Tu Krugman Feldstein

Paul can show, graphically, that conventional monetary policy is then completely ineffective–swapping two assets that are perfect substitutes for each other. Paul can show, graphically, that expansionary fiscal policy is then immensely powerful and has no downside: it does not generate higher interest rates; it does not crowd out productive private investment; and, because interest rates are zero, it entails no financing burden and thus no required increase in future tax wedges. But all this is constrained and limited by the inescapable and powerful logic of the induction-unraveling propagating itself back through the game tree from the Omega Point that is the long run equilibrium. In the IS-LM diagram, the fact that the long run is out there means that even the contemplation of permanent expansion of the monetary base is rapidly moving the IS curve up and to the right, and thus leading the economy to quickly exist the liquidity trap. In the Wicksellian I=S diagram, the fact that the long run is out there means that even the contemplation of permanent expansion of the monetary base is rapidly moving the I=S curve up so that the zero lower bound will soon no longer constrain the economy away from its full-employment equilibrium.

The “depression economics” equilibrium Paul plots on his graph is a matter for today–a month or two, or a quarter or two, or at most a year or two.

But it will soon be seven years since the U.S. Treasury Bill rate was more than whispering distance away from zero. And it is now more than two decades since Japan’s short-term bonds sold at less than par.

Paul has a critique of the extremely sharp Marty Feldstein’s latest over at Project Syndicate (parenthetically, I must say it is rather cruel for Project Syndicate to highlight Feldstein’s August 2012 “Is Inflation Returning” in site-searches for “Feldstein”):

Paul Krugman: The Inflationista Puzzle: “Martin Feldstein has a new column on what he calls the ‘inflation puzzle’…

…the failure of inflation to soar despite the Fed’s large asset purchases, which led to a very large rise in the monetary base. As Tony Yates points out, however, there’s nothing puzzling at all about what happened; it’s exactly what you should expect when interest rates are near zero…. This isn’t an ex-post rationale, it’s what many of us were saying from the beginning. Traditional IS-LM analysis said [it]… so did the translation of that analysis into a stripped-down New Keynesian framework that I did back in 1998, starting the modern liquidity-trap literature. We even had solid recent empirical evidence: Japan’s attempt at quantitative easing in the naughties, which looked like this:


I’m still not sure why relatively moderate conservatives like Feldstein didn’t find all this convincing back in 2009. I get, I think, why politics might predispose them to see inflation risks everywhere, but this was as crystal-clear a proposition as I’ve ever seen.

Still, even if you managed to convince yourself that the liquidity-trap analysis was wrong six years ago, by now you should surely have realized that Bernanke, Woodford, Eggertsson, and, yes, me got it right…. Maybe it’s because those tricksy Fed officials started paying all of 25 basis points on reserves[?] ([But] Japan never paid such interest[.]).

Anyway, inflation is just around the corner, the same way it has been all these years.

Unlike Paul, I get why moderate conservatives like Feldstein didn’t find “all this convincing” back in 2009. I get it because I only reluctantly and hesitantly found it convincing. Feldstein got the Hicksian IS-LM and the Wicksellian S=I diagrams: he just did not believe that they were anything but the shortest of short run equilibria. He could feel in his bones and smell in the air the up-and-to-the-right movement of the IS curve and the upward movement of the S=I curve as investors, speculators, and businesses took look at the size of the monetary base and incorporated into their thinking about the near future the backward induction-unraveling from the long run Omega Point. My difference with Marty in 2009 is that he thought then that the liquidity trap was a 3 month-1 year phenomenon–that that was the duration of the short run–while I was much more pessimistic about the equilibrium-restoring forces of the market: I thought it was a 3 year-5 year phenomenon.

And it was not just me. Consider Ben Bernanke. I have no memory any more of who was writing [Free Exchange] back in 2009. But whoever it was was very sharp, and wrote:

????: Person of the next five to ten years: “There are those who blame [Bernanke] for missing all the warning signs…

…those who blame him for managing the crisis in the most Wall Street-friendly way… those who blame him for laying the groundwork for a future asset bubble or inflation crisis…. I think his defining decision… has been to conclude that 10% unemployment is acceptable–that having averted a Depression-style 25% unemployment scenario, his countercyclical work is complete… that the risk of sustained high unemployment is outweighed by the risk of… efforts to boost the economy… by asking for more fiscal stimulus… targeting nominal GDP or… committing… to some [higher] level of inflation….

Bernanke believes most of the increase in unemployment… to be cyclical… does not think that pushing… unemployment… down to… 7% would overextend the economy…. He simply seems to think that leaving his primary job half done is acceptable. That’s a pretty momentous choice, affecting millions of people directly and billions indirectly. It will shape American politics and economics for the next decade, at least…. He deserves… person of the year…. But reappointment? That’s another story entirely.

What this leaves out is that Bernanke was willing to take his foot off the gas in late 2009 with an unemployment rate of 10% because, like Marty, he could smell the back-propagation of the induction-unraveling of the short run equilibrium. He us expected that, with his foot off the gas, unemployment would be 8.5% by the end of 2010, 7% by the end of 2011, 6% by the end of 2013–and thus that further expansionary policies in 2010-2011 would run some risk of overheating the economy in 2013-2014 that was not worth the potential game. He didn’t see the liquidity trap short run as as brief as Marty did. But he also didn’t see the short run as as long as I did–and I have greatly underestimated its duration.

(Someday I want Christina Romer to write up her memoir of late 2009-late 2010, as she wandered the halls of the White House, the Federal Reserve, the IMF, and the OECD, trying to convince a bunch of economists certain that the short run was a year or two that all the historical evidence we had–the Great Depression and Japan’s Lost Decades, plus what we dimly think we know about 1873-9, and so forth–suggested, rather, that it the short run would, this time, be a five to ten-year phenomenon. Yet even with backing by Rinehart and Rogoff on the short run equilibrium duration (albeit not the proper fiscal policy) front, she made little impression and had next to no influence.)

Ahem. I have gotten off track…

My point:

Back in late 2009 I thought that the liquidity-trap short run was likely to be a three-to-five-year phenomenon. It has now been six. And the Federal Reserve’s proposed interest-rate liftoff now scheduled for the end of 2015 appears to me profoundly unwise as a matter of technocratic optimal control, prudent policy, and recognition of the situation. The duration of the short run thus looks to me to be, this time, not three to five years but more like ten. Or more. The backward-propagation of the induction-unraveling of the short run under pressure of the healing rays of the long run Omega Point is not just not as strong as Marty Feldstein thought, is not just not as strong as I thought, it is nearly non-existent.

Thus I find myself getting somewhat annoyed at Paul Krugman when he writes that:

Paul Krugman: Choose Your Heterodoxy: “A lot of what I use is 1930s economic theory…

…via Hicks. And I should be deeply ashamed…. [But] plenty of physicists who still use Newtonian dynamics, which means that they’re seeing the world through the lens of 17th-century theory. Fools!… Farmer is trying to explain an empirical regularity he thinks he sees, but nobody else does–a complete absence of any tendency of the unemployment rate to come down when it’s historically high. I’m with John Cochrane here: you must be kidding…

Or that:

Paul Krugman: Nonlinearity, Multiple Equilibria, and the Problem of Too Much Fun: “Was the crisis something that requires novel multiple-equilibrium models to understand?…

…That’s far from obvious. The run-up to crisis looks to me more like Shiller-type irrational exuberance. The events of 2008 do have a multiple-equilibrium feel to them, but not in a novel way… pull Diamond-Dybvig…. And since the crisis struck, as I’ve argued many times, simple Hicksian macro–little equilibrium models with some real-world adjustments–has been stunningly successful…


Paul Krugman: Learned Helplessness: “We knew all about liquidity traps, and had at least thought about balance-sheet crises…

…a decade ago…. The Return of Depression Economics in 1999. The world we’re now in isn’t that different from the world I suspected, back then, we’d find ourselves in. Oh, and about Roger Farmer and Santa Fe and complexity and all that: I was one of the people who got all excited about the possibility of getting somewhere with very detailed agent-based models–but that was 20 years ago. And after all this time, it’s all still manifestos and promises of great things one of these days…

The problem is that the macroeconomics that Paul Krugman learned at Jim Tobin’s knee wasn’t just 1930s-style Hicks-Hansen Keynesianism. It was the 1970s adaptive-expectations Phillips Curve neoclassical synthesis–nearly the same stuff that I first learned at Marty Feldstein and Olivier Blanchard’s knees in the spring of 1980. That is the framework that Marty is using know, and that generates his puzzlement. That framework had a short run of 1-2 years, a medium-run transition-dynamics phase of 2-5 years, and a long run of 5 years or more baked into it. You cannot–or at least I cannot–just throw away the medium run transition dynamics* and the declaration that the long run Omega Point is five years out, and say that mainstream economics does well. You need to explain why the back-propagation induction-unraveling worked at its proper time scale in the 1970s and the 1980s, but is nowhere to be found now.

And so I am much less confident that I have solid theoretical ground under my feet than Paul Krugman does.

October 20, 2015


Brad DeLong
Connect with us!

Explore the Equitable Growth network of experts around the country and get answers to today's most pressing questions!

Get in Touch