Robert Feldman of Morgan Stanley powerfully argues that the center-of-gravity of economists' positions draws the wrong lessons from Japan of a decade ago for the U.S. today. The center-of-gravity is that what Japan's decade-long crisis teaches us is that monetary and fiscal policy need to be very aggressive in the aftermath of a bubble if depression or stagnation is to be avoided. Feldman thinks that is wrong.
Feldman thinks that the key lesson is that the collapse of a bubble produces substantial microeconomic problems--structural imbalances--that cannot be fought with macro policy.
As applied to the U.S. today, Feldman's argument is that the Federal Reserve can push nominal interest rates on Treasury Bills down to zero, but that still will not revive corporate investment as long as savers are skeptical of the accounting numbers reported by corporations and interest-rate spreads grow.
At some level, he's clearly right. The question is how right. How much of an impact will the corporate governance crisis have on the wedge between the terms on which the federal government can obtain capital and the terms on which private businesses can obtain capital? That's the big--and still unknown--question.
Japan: The Fed's Wrong Lessons from JapanSo what are the real lessons of Japan's experience for other countries that are flirting with deflation? The first lesson is to use a MACRO-economic model that accounts for MICRO-economic imbalances.... Japan's deregulation of the portable telephone industry in 1994 was an example of how brilliantly the right micro policy can improve welfare, just as the refusal to implement strict bank inspections is an example of how miserably the wrong policy can worsen welfare. The good news is that the microeconomic structures of Japan and the US are quite different. Japan suffers from less flexible labor markets, more intrusive regulation, less strict corporate governance, an inadequate judicial system, and many other structural problems.... In contrast to the lessons that I draw, the Fed paper concludes that Japan's experience implies that the US needs a very aggressive further easing of monetary policy, and yet more fiscal stimulus. My fear is that such policies would only exacerbate the imbalances in the US economy that my colleagues on the US economic team have discussed so cogently. Moreover, the political convenience of the Fed paper's recommendation is intergalactic. It is a lot easier to print money and to spend money than to reform accounting systems, to jail criminals, and to face angry voters...
The staff of the Federal Reserve Board has prepared a discussion paper on lessons from Japan on the question of preventing deflation. The boilerplate on the cover page of the Fed’s discussion papers says that such papers are "circulated to stimulate discussion and critical comment." This one sure has. Moreover, I find that two of the contributing authors are old classmates of mine from graduate school. So, with wistful memories of those halcyon days in the decrepit student lounge of building E52, I offer the following discussion and comment.
The main conclusions of the Fed paper are as follows. (1) Japan’s deflationary slump was unanticipated by policymakers and other observers. (2) As a result, there was a failure to provide sufficient stimulus to maintain growth and prevent deflation. (3) Once interest rates approach the zero lower bound, it became more difficult for monetary policy to reactivate the economy. (4) Neither monetary nor fiscal policy had lost the ability to support the economy in the early 1990s. The policy implication drawn from these conclusions is that, when inflation and interest rates are near zero, monetary and fiscal stimulus should go beyond conventional levels, in order to prevent outright deflation.
Much of this is quite reasonable and true. However, the analysis and conclusions suffer from some omissions. Some are theoretical, and others are informational. When added to the analysis in the paper, however, the policy conclusions are called into question.
The first conclusion, that the long slump was unanticipated, is true without a doubt. I confess. In 1993, in a paper entitled "Anorexia or Dyspepsia?," I concluded that Japan’s growth problem was merely dyspepsia, on the grounds of macroeconomic stimulus already in the pipeline. Even in December 1997, although I correctly predicted massive growth of the BoJ balance sheet, I did not even mention the possibility that such support might fail to revive the economy. Where did I go wrong? I was using the wrong model. Keynes once said that "economics is the science of thinking in terms of models, joined to the art of choosing models that are appropriate to the situation." In my case, I was vastly underestimating the importance of getting microeconomic policies correct as a precondition for macroeconomic policy to work. I think that the Fed paper makes the same mistake.
The Fed paper’s conclusion is based on running a number of models, including Taylor Rule models and large-scale econometric models. Those who have estimated models for use in policy work know that the structure of the models often creates blind spots, even before the first regression is run. In the case of the Taylor Rule models, this is easy to see. Taylor Rule models are, without a doubt, a great advance in policy work. However, their very structure abstracts from two critical aspects of the Japanese situation.
First, Taylor Rule models assume the standard "one-good" structure of most macro models. When there are no relative price distortions in an economy, such an assumption is warranted. However, for Japan today, there are (at least) two huge relative price distortions, land prices (due to lingering effects of the bubble -- which have yet to be cleaned up) and traded goods prices (due to the ongoing emergence of China into the world trade structure). When weak real growth is the result of distortions of relative prices among goods, it is not clear that lowering the relative price of money versus goods can improve the situation.
Second, Taylor Rule models assume a given level of potential output (or, alternatively, a given path for potential output). The level of potential output is unaffected by monetary policy. In Japan over the last decade, easy money policy has in fact at least maintained and most likely expanded excess supply. The Fed paper acknowledges this point in a footnote, but does not include the idea in the models or in the policy analysis.
This problem with the Taylor Rule points up a significant difficulty in consensus macroeconomic models. Monetary and fiscal policies are assumed to affect only the demand side of the economy. While they certainly do affect the demand side, they can affect the supply side as well. In some circumstances, it is conceivable that the supply-side effects might even exceed those on the demand side. For example, in a liquidity trap, increased money printing would keep bankrupt firms in business, but would do nothing to demand. The output gap might even increase, and make deflation worse.
My contention is that Japan today inhabits exactly this kind of world. Because of the reluctance of the financial regulatory authorities to impose and implement strict standards on loan classification or capital adequacy for financial institutions, both the institutions and the inefficient borrowers have remained in business. Such a situation naturally raises questions of the second best. How should monetary policy be formulated in the absence of correct supervision policy?
The second major omission in the Fed paper is that of a model of political economy. At one point, the Fed paper acknowledges that "the Japanese stimulus packages may not have been designed so as to maximize their macroeconomic impact," a point I have made constantly for years on the basis of the political economy theories of Buchanan and Pelzman. In this context, not only must the Taylor Rule framework be modified, but so must all fiscal and structural economic models that rely on the assumption (implicit or otherwise) of a benevolent dictator forming policy. Central banks are not immune. There have been cases when central banks are looser than they should be, in order to protect their policy independence. Fiscal policy is particularly prone to abuse -- as the Fed paper acknowledges for Japan. Yet this acknowledgement vaporizes when the policy implication of the paper is drawn -- i.e., to be very aggressive with fiscal policy when deflation threatens. At the very least, the potential for abuse of fiscal policy (and the impact of such abuse on the economy) must be considered when drawing policy implications over the amount of desirable stimulus. The contention that "it cannot happen here" is ludicrous. It happens everywhere.
This point is critical in the context of the CRIC cycle (the cycle that describes the interaction of policy and the economy in a cycle of Crisis, Response, Improvement, and Complacency). This cycle arises because policymakers are seldom willing to take hard decisions when times are good. Even with a crisis, the feel-good policies of macro stimulus are usually preferred. This was clearly the case in Japan’s aggressive use of fiscal and monetary stimulus in the early 1990s. In the end, however, this stimulus only postponed the needed micro reforms, and worsened the microeconomic response.
For the US, there is a danger of falling into a CRIC cycle, I believe. For example, some observers have suggested that the US lower margin requirements, as a way to support the equity market. If this were successful, the pressure for more fundamental reforms (e.g., better quality earnings reports) would wane. Likewise, further interest rate cuts or tax cuts could push equity prices upward. A higher equity market would dull the edge of the crisis, and dampen the pressure on policymakers for action. It might also invite policymakers to manipulate markets, in order to avoid reform. Japan has fallen into some of these traps, and the US could as well.
So what are the real lessons of Japan’s experience for other countries that are flirting with deflation? The first lesson is to use a MACRO-economic model that accounts for MICRO-economic imbalances. In econo-jargon, the endogenous nature of aggregate supply must be modeled as part of the macro policy process. The second lesson is to address microeconomic problems with microeconomic solutions. Japan’s deregulation of the portable telephone industry in 1994 was an example of how brilliantly the right micro policy can improve welfare, just as the refusal to implement strict bank inspections is an example of how miserably the wrong policy can worsen welfare.
The good news is that the microeconomic structures of Japan and the US are quite different. Japan suffers from less flexible labor markets, more intrusive regulation, less strict corporate governance, an inadequate judicial system, and many other structural problems. In contrast, the US has a far more flexible if socially harsher microeconomic structure, which should obviate some of the problems of macro/micro coordination. History gives some hope: After all, when the Fed lowered real rates to zero during the S&L crisis, there was no diminution of regulatory zeal. (The US did have a zeal problem in the 1980s, but this was solved, after several false starts, before the big Fed easings.)
In contrast to the lessons that I draw, the Fed paper concludes that Japan’s experience implies that the US needs a very aggressive further easing of monetary policy, and yet more fiscal stimulus. My fear is that such policies would only exacerbate the imbalances in the US economy that my colleagues on the US economic team have discussed so cogently. Moreover, the political convenience of the Fed paper’s recommendation is intergalactic. It is a lot easier to print money and to spend money than to reform accounting systems, to jail criminals, and to face angry voters (who, by the way, just loved the US bubble on the way up but want to blame someone else for the collapse). This political convenience alone should at least make one skeptical of expansionist recommendations. In my opinion, neither monetary nor fiscal expansion would contribute to the heavy lifting that the US must do, in order to reduce its structural imbalances. Macro policies will hasten the adjustments if and only if proper micro policies are taken. In short, the lesson from Japan is that macroeconomic convenience is no substitute for microeconomic sweat.
Posted by DeLong at August 12, 2002 05:14 AM | Trackbackblah, blah, the "corporate governance crisis" is a consequence rather than a cause of the stock market slump, blah, blah, the economy is not the stock market, blah, blah, since retained earnings are the major source of financing and most companies (ie; all except regulated utilities) don't even *know* their cost of capital, how can this affect investment in the long term.
Blah, blah, there was no "corporate governance crisis" last September, blah, blah, Krugman has a perfectly adequate demand-side explanation of Japan's problems which everyone has ignored ...
Blah, blah, the article is off by an order of magnitude in comparing the Japanese banking crisis to the S&Ls, blah blah, what the hell do "flexible labour markets have to do with this, blah ...
Sorry, Brad. I'm getting my own weblog soon, when I can figure out how to do so consistently with my religious beliefs (specifically the belief that it is immoral to pay any money for anything on the Internet), so I'll be filling up your comments section with less of this crap. The attached article is not really all that bad ... but it suffers from not getting to grips with the Japanese data, and with a really bad case of supply-side hand-waving. Does anyone really believe that the reason that millions of Japanese are unemployed is that "corporate governance is weak"?
Posted by: Daniel Davies on August 12, 2002 05:42 AMBrad
Morgan Stanley has never considered Japanese government macro policy to have contributed to the lost decade. Krugman - Smithers - have had it right for years. Japan has had macro policy that has accentuated the stock market decline from 1989. The macro policy has produced deflation that shows no sign of abating. Much like England in the decade before Keynes or the Hoover administration before FDR. There is a liquidity trap in Japan. The only thing that has kept Japan from a fearful fearful recession is the high employment level and good income distribution.
Krugman has written several important essays on Japan. Look to the unofficial Krugman site.
Morgan Stanley has never even bothered to argue with Krugman.
Posted by: on August 12, 2002 08:49 AMBrad
Why is it that Krugman has long made so much sense on Japan? At the same time that you have been properly urging stimulative monetary policy for America, arguing that a weaker dollar would be of assistance as well, Japan stifled the economy with a restrictive monetary policy far into the slump of the 90's. Also, Japan has had a fairly strong yen for a decade Compare the yen in 1989 and today. Japan is a test case gone sour for Hooverism in the face of a stock market and economic slump. Krugman seems to have long been right on Japan and also on the peg in Argentina.
Posted by: on August 12, 2002 02:50 PM>> Japan stifled the economy with a restrictive monetary policy far into the slump of the 90's<<
It's not entirely self-evident that Japan's monetary policy has been so tight. E.g. real rates were negative 2% back in 1997 and have been functionally close to zero ever since. And the BoJ has even resorted to quantitative easing of the Milton Friedman kind to get money out there in spite of a money transmission (banking) system that is broken.
( e.g.: http://www.clev.frb.org/research/Et2001/0401/Html/japmonpol.htm )
Some quick thoughts come to mind:
[] Years of negative-to-minimal real interest rates generally aren't associated with deflation-causing tight money policy. E.g, the US deflation of the 1930s followed years of very high real rates.
[] Almost all textbooks these days -- including I'd wager Prof. Krugman's -- say that "in the long run money is neutral", and we're getting into a pretty long run here, 10+ years during at least half of which (maybe more, I didn't check) real rates have been negative-to-minimal. If the textbooks are right ...
[] The Japanese economy obviously has serious structural problems. E.g., if the money transmission system wasn't broken the BoJ wouldn't have to resort to quantitative easing like no other central bank. (Then there are all the features of one-party government, protected rice growing in Tokyo, etc.)
[] The BoJ warns that relying too much on monetary easing could actually make matters worse, because it does nothing to alleviate the structural problems of the economy but does give a temporary stimulus that makes it easier for politicians to ignore them and let them fester.
It's usually a good idea to give at least a fair hearing to "the troops on the ground" in a distant location, and here that's the BoJ.
The Economist recently reported that Japan has lost more relative to economic capacity than the US did during the Great Depression, albeit with a different pattern -- the US took a huge fall then climbed back up, while Japan avoided the huge fall up front but has been stagnating for a decade so far falling ever further behind with no end in sight.
It's hard for me to understand how such a decade-plus of stagnation can be caused primarily by tight money when real rates have been negative-to-minimal for half that time, at least. Something structural looks broken. ("Liquidity trap" looks a bit glib to me. There wasn't the necessary deflation for most of the time, until just the last few years. And I'm not sure even that's the 'necessary' deflation. Keynes saw 10% deflation and still considered the liquidity trap theoretical).
Corporate governance? Japan hasn't had a single Enron, Worldcom, Global Crossings, yet. It should have had a bunch.
If the US banking system was to be used to prop up all those firms and their fellows for the next decade (and to keep their leaderships in place, in a deal with the one-party government) what would the effect be on funds available for real productive investment (especially if the US had no real bond market), and on the banking system's solvency, and money transmission, etc.?
I dunno. Looks to me like it might be a part of the problem.
'[] Almost all textbooks these days -- including I'd wager Prof. Krugman's -- say that "in the long run money is neutral", and we're getting into a pretty long run here, 10+ years during at least half of which (maybe more, I didn't check) real rates have been negative-to-minimal. If the textbooks are right ...'
The other possiblity, which Krugman has advocated, is that interest rates still aren't negative enough. There's got to be *some* point that'll drag demand out of the gutter.
Posted by: Jason mccULLOUGH on August 12, 2002 07:34 PM>>Japan hasn't had a single Enron, Worldcom, Global Crossings, yet<<
Jim, this is a pretty ignorant statement ...
Posted by: Daniel Davies on August 12, 2002 11:10 PMNOTES ON DEPRECIATION, JAPAN, AND ARGENTINA
By Paul Krugman
In today’s world, there are two big “deflation” stories – Japan and Argentina. In both cases the way out almost certainly involves, among other things, a major depreciation of the currency. And in both cases, as the final act draws near, there is a chorus of skeptics.
The argument – which you now hear about both – goes like this: exports aren’t a very large share of GDP. That means that currency depreciation can’t have much impact on aggregate demand; so depreciating the currency is ineffective, and one might as well maintain the current exchange rate.
If you think about it, there must be something wrong with this argument; it seems to suggest that a fixed exchange rate between two economies is more desirable (or at least less costly) the less trade they have with each other. The dollar zone and the euro zone are both largely self-sufficient? Good – let’s fix the dollar-euro rate. Huh?
What’s wrong with this argument? It misses the point that a fixed exchange rate also, under conditions of capital mobility, deprives a country of independent monetary policy. The point of depreciation is to get that monetary policy back; the lift from increased net exports is only part of the story.
Here’s one way to think about it. Suppose that your country is committed to a fixed nominal exchange rate – and that the currency is overvalued. What I mean by “overvalued” is that the price of domestic goods, relative to foreign goods, is higher than it would be if all prices were completely flexible. What one would expect to happen in this case is a process of deflation – certainly relative deflation, with the domestic inflation rate less than the foreign, and quite possibly actual deflation too.
Now ask what this implies for interest rates. Even if the nominal exchange rate is completely credibly pegged – if, say, your economy is dollarized – the nominal rate in your country will be the same as the rate abroad. But because of the ongoing relative deflation, your real interest rate will be higher than it is abroad. And your economy will be depressed both because of depressed exports (the direct result of overvaluation) and because of a high real interest rate.
Nominal depreciation, if all goes well, allows you to go immediately to the equilibrium real exchange rate, eliminating the overvaluation – and therefore allows a reduction in the real interest rate. If exports are a small share of GDP, this real interest rate effect, rather than the “competitiveness” effect, will be the main source of gains from depreciation.
In fact, if the economy faces temporary adverse shocks, it is possible with a flexible exchange rate to depreciate the real exchange rate beyond its long run level, allowing a lower real interest rate in your country than abroad. This is the reason people like Lars Svensson and myself advocate a weak yen policy: the point is that a weak yen is part of a strategy to lower the real interest rate. Focusing only on the direct competitiveness effect misses that point.
Now back to Argentina. The problem with the overvalued peso has been not just the export weakness but the high real interest rate – admittedly a problem exacerbated by the peg’s increasingly shaky credibility. And the point of any new monetary policy should be to get rid of the real overvaluation that is at the root of the problem.
I know it’s hard for any Argentine government to face up to the ugly truth that a devaluation – a real devaluation, involving a real currency – is necessary. But the longer this truth is denied, the worse it gets.
Posted by: on August 13, 2002 08:38 AMThe drift of discussion here catches the key point:
- monetary policy, alone, can neither explain Japan's situation, nor find a rectification for it
BUT
money is the lever that is left to the Japanese government, and Krugman and Smithers are right that they need to break all the 'rules' in using it.
One consequence will be a fall in the yen. This at the very least should increase the domestic Japanese inflation rate, which should make more structural adjustments possible.
The loser *has* to be the existing holder of Japanese government debt (this is very like Argentina).
The winners *have* to be the holders of Japanese real assets (chiefly property).
The macro adjustment will be inflation, achieved through unscrupulous printing of money.
But there has to be a micro adjustment as well: the market needs to be convinced that the property market has bottomed. Effectively, the Japanese government will have to do a Roosevelt, and nationalise the banks, foreclose their bad loans and dump the assets to whatever buyers emerge (yes, including American vultures). This will have to take place against a background of vastly increased unemployment benefits (to prevent civil apocalypse). All of this to be funded by cuts in the construction budget and by the BOJ buying government securities directly (aka irresponsible printing of money).
Its this last step that the decentralised nature of Japanese political power makes difficult. The LDP is too cronyist and corrupt to manage it: could a socialist alternative be the new Thatcher?
Is the US parallel? Hopefully not: the US has more flexibility in nominal everythings than virtually any other economy. But a combination of government spending cuts, tax cuts and a tightening of monetary policy could be *very* bad.
If one wanted to look for a weak point in the US financial system, I would be tempted to look at Fannie Mae/ Freddie Mac and the whole housing market/ housing finance structure. I am not sure this could survive a prolongued deflation.
Posted by: John on August 14, 2002 03:12 AM