by Adam S. Posen, Peterson Institute for International Economics
Remarks at the Institute for International Economics
January 15, 2002
© Peterson Institute for International Economics
Following a December 2001 trip to Tokyo, this is a significantly revised and extended version of remarks initially made at the conference "Where Does the Japanese Economy Go from Here?" Columbia Business School Center on Japanese Economy and Business, November 2, 2001, New York.
Japan is in very deep trouble, but the troubles are very obvious and easy to understand. Due to the frustration and desire to see Japan's economy as something very distinct, there has been a tendency over the last 10 years to constantly argue that there are deep-seated and complicated structural problems, technical changes, and political pressures working against Japan. But, it is not that complicated.
Japan is experiencing what others and I have called the Great Recession. The normal recession began after a stock market crash. That recession was not fully recognized or dealt with until 1994-95. A series of policy mistakes deepened the recession at that time. Japan also has a banking crisis. This is not unusual. We have had banking crises in the United States, France, and Sweden, for example. Japan had postponed dealing with it just as we did in the United States, France, and Sweden. For some reason, it postponed a lot longer. If you put these two together, you have 10 years of stagnation. You do not have to make life much more complicated than that.
Of course, it is possible to claim that the economic problems cannot be solved through demand management and financial clean-up. This would mean that the decline in growth in Japan in the 1990s was due to a shift in real factors, like the rate of total factor productivity (TFP) growth, and therefore this was not cyclical. Some noted Real Business Cycle economists, like Fumio Hayashi and Edward Prescott, have made this claim, suggesting that Japanese trend TFP growth declined by as much as 3 percent a year (from 4 percent to 1 percent) at the start of the 1990s. Yet, if the TFP growth rate went down in Japan during the 1990s, the question is why and how would we see it beyond simply labeling the decline in growth as that.
We can tell stories that Japan made it to the technological frontier, but that only gets you down to a TFP growth rate at US levels (now 2.5 percent), which is not the size of the decline the Real Business Cycle view needs here to explain the growth slowdown. Is it that Japan is supposed to be bad at new age technologies, like biotech and software? I have always found that to be a very odd, culturalist argument. I have also found it went wrong on the facts1. Japan's research and development capabilities and technological companies are still producing. Seven of the top 10 companies by number of patent applications in the United States in each of the last five years has been Japanese, for example. This has been the case throughout the 1980s and 1990s. There are whole realms of technology where Japan continues to be dominant and will be dominant.
It is very hard to tell a story why TFP growth should have dropped so sharply and so suddenly for reasons other than the recession itself especially when there are so many people who have documented that the institutions of Japan are arthritic and unchanging. That absence of domestic change implies that any non-cyclical TFP decline would require some kind of change in the environment around Japan that somehow only adversely affected Japan and no other OECD or East Asian economy, which is plainly nonsensical. It is particularly absurd when one recognizes that the size of the structural TFP growth decline spoken of here is three times the size of the US gains in TFP growth from the new economy (which were between 0.5 percent and 1.0 percent).
So if it is a great recession, but nothing more, what can you do about the situation? I think there are four elements: the BoJ's undertaking inflation targeting, the government shifting but not expanding fiscal policy by changing the composition of public spending, the FSA being tough and trying to put financially distressed real assets back on the market, and then—and only then—the Cabinet Office pushing ahead with structural reform beyond the financial system.
In terms of the BoJ doing inflation targeting, I am in complete agreement with such domestic Japanese critics as Takatoshi Ito. I think there is almost no one outside the walls of the BoJ right now who is not extremely critical of the BoJ. Not everyone buys into inflation targeting specifically, and not everyone is sure that rapid expansion of money supply will have immediate beneficial effects. Everyone, however, is certain that any claims that the current deflation are a good development are false. And everyone is certain that the claims that the Bank of Japan would stimulate disastrous inflation or an unfolding of fiscal discipline if it took aggressive action are false. There is no way to support this position.
It is very discomforting to make these kinds of blunt declarative statements about public servants' efforts, but it is even more difficult to imagine how the BoJ allowed the Japanese economy to get to where it is now. Without crying over spilt milk, what is the BoJ left with as policy tools in today's economy? The BoJ is left with essentially two major tools. The first is jawboning statements making policy commitments. Since the governor and the senior officials of the BoJ have made such outrageous statements for so long, a clear commitment of different policies would have an effect on the market. The second tool is for the BoJ to accommodate fiscal policy. The Bank can print yen, and with those printed yen buy Japanese government bonds (JGBs) or other assets. There is no question that this eventually will lead to inflation and that it will increase the impact of fiscal policy.
I would emphasize that the printing of yen should be focused on monetizing government debt. This makes it easier for the government to take fiscal action until the limit when inflation expectations (and interest rates) rise is reached. I do not think buying of nontraditional assets is a good idea, mostly because the Bank would not know what to do with real estate or equities; I also do not like central banks mixing into markets where their decisions can bias allocation between sectors, which buying JGBs would only do secondarily. I also am less sympathetic to buying foreign exchange and/or foreign government bonds, both because the effects would be less direct and I worry about the political backlash of a conscious effort to depreciate the yen.
Thanks to President Bush's advisor Larry Lindsey, there are calls for the Bank of Japan specifically to purchase foreign (read US) treasury bonds. There is no particularly good argument for this on the economics—it does not expand the money supply any more than buying JGBs would. There could be a foreign exchange effect, by buying foreign assets, but unless an awful lot of them were bought, that f/x effect would likely be small2—and if an awful lot of US T-bills were bought, the f/x effect would be partially counter-acted because the gap between US and Japanese interest rates would come down and weaken the dollar. The reason this is being floated is political, because some people at the BoJ would rather do this than "erode fiscal discipline" and it looks better on the balance sheet than allowing the MoF to direct sterilized intervention3. These are not good reasons to pick a policy.
Still, if a declining yen were to come about as a side effect of looser domestic policy and a bank clean-up, I think everyone should try to live with that. Given the world economic situation, Japan has missed its opportunity to let the yen shift the brunt of its adjustment without suffering a political backlash. The Koizumi administration needs to not only have plausible deniability that it is not engaged in competitive depreciation, but also that it will have something (financial reform) to show for what ever it does to the world when its currency declines. Essentially, the endaka-logic has now been reversed, the pressure will be on Japan from a declining yen. One reason I suggest domestic JGB purchases is to avoid the political problems with the United States—and the real economic harm to Japan's neighbors—from yen depreciation.
But one can be stronger than that—since I do think there are many good things to be done on the fiscal front in Japan—I don't see any reason to look beyond JGBs for open market operations. Japan is facing a very bad deflationary spiral and BoJ Governor Hayami himself has said several times that you have to watch out for that (though he hasn't done anything about it). If that is the situation, then the alternative of aggressively seeking inflation is a better option—whatever costs you think you can incur by buying some unprecedentedly large number of bonds are overestimated. If especially we believe money multiplier is very low, and the transmission mechanism is unlikely to be effective (as the BOJ has argued with some reasonability), then we certainly don't have any inflationary spiral to fear from doing this.
Fiscal policy in Japan still works very well. The key mistake that people seem to make is to assert that because the deficits and debt in Japan have risen, but growth has not returned, therefore, fiscal policy must be ineffective. That is not true because a certain amount of debt accumulation is simply a reflection of the economy getting worse and tax receipts decreasing. If you control for this you can find out and isolate the true effect of fiscal expansion.
In a recent paper4, Kenneth Kuttner and I performed just such an analysis. Our results show that every time the Japanese government has undertaken fiscal expansion in the 1990s it has worked as predicted. Even the wasteful public works spending, much to my surprise, has a pretty solid positive multiplier. On our estimates even these ridiculous projects have a multiplier of approximately 300 yen over a four-year horizon for every 100 yen you put out, if you take into account the fact that the spending tends to persist. The few instances in Japan of (temporary) tax cuts have had a much larger multiplier and, obviously, structural benefits, so that would be better than public works going forward. For tax cuts, our estimate of what you get back for every 100 yen is in excess of 450 yen at a four-year horizon, but this is with the tax cuts being temporary. So in terms of bang for the yen, actually tax cuts beat public works in Japan by an even larger margin.
Again, to repeat something I have said many times in many places, and others like Takatoshi Ito and Marcus Noland have also argued, the ideal sort of tax cut would be a housing tax cut. We saw this work very well in the first half of 1999 when it was oriented toward new condo construction. My current proposal is to do many things at once: announce that a large number, say 30 trillion yen, in public works will not be done this year, and all that money will be put into a trust fund. That trust fund will be for private households who sell their real estate at a loss and have to pay off their higher mortgage-if they do so, they can deduct their losses on their taxes. People can do this (for their one major residence) until the total deductions use up the trust fund. This would have the benefit of restoring liquidity to the real estate market (letting prices finally find a bottom) and of encouraging people to act now on the tax cut, in addition to shifting stimulus from low to higher multiplier form. If it is successful, one can always move more money into the trust fund, and still remain revenue neutral.
These general analyses of tax policy and public spending in Japan are very robust results. The longer the decade has gone on, the more the evidence has accumulated that fiscal policy still works in Japan. If you look back at 1997 and the tax increase at that time, it is very clear that that had a contractionary effect. The temporary new condo tax break of early 1999 was clearly expansionary.
What about the long-term debt? The basic way in which this matters is twofold. First, what are you going to do when Japan gets really old as opposed to merely old? Second, will this demographic trend lead savers to offset current attempts to use fiscal policy? Let me address the first concern with a fable and the second one with direct evidence.
The fable is what I call the story of the drought ridden rice paddy5. You have a rice paddy. You have heard on TV that there is going to be global warming in 20 years and it is not going to be possible for you to continue to farm rice because there is not going to be enough water. You also hear on TV the next month there is going to be a drought. Luckily, you have a reservoir up the street. Do you take the prospect there is going to be global warming in 20 years so seriously that you do not use the reservoir this year to save this year's crop? Absolutely not. Holding off using the reservoir does nothing to prevent global warming.
The same case could be made with fiscal policy in terms of balancing the short and long runs. If counter-cyclical fiscal policy works, i.e., you have a greater than one multiplier, that means that literally you are getting back more growth and, therefore, more tax revenue, for the amount you put out. There is no acceleration of the demographic threat by using your fiscal resources today. To address the demographic problem, the underlying major forces have to be addressed by raising the retirement age, the participation of women in the workforce, the utilization of immigrants and guest workers, or raising the potential growth rate.
In terms of evidence regarding savers' offsets, the basic concern is for what is called Ricardian equivalence. In this view, the reason Japanese people will not spend tax cuts or public works proceeds is because they are scared about the future demography and their pension plans, and the need to raise taxes eventually, and, therefore, fiscal policy will be ineffective. We have ways of looking for this effect in the data. There was a big debate in the United States in the 1980s over Ricardian equivalence and econometric techniques were developed to go out and test for its effects.
It turns out that even in the extreme case of Japan in the 1990s, there is almost no Ricardian offset. You have an offset over a long period (six-plus years) of about 30 yen for every 100 yen spent by the government—a large part of which is attributable to the fact that many of these tax cuts are temporary rather than permanent in the first place. What does this mean in the short term? There is no reason for the Japanese government not to engage in aggressive tax cuts to offset the contractionary effects of resolving the banking problem as well as the general recession.
With respect to financial policy, it is important to emphasize how the macroeconomic effects of the banking problems lead one to much the same policy conclusions as evaluations of their microeconomic effects. Even though shutting down banks would be contractionary, I think people vastly overestimate how contractionary it would be beyond the present conditions. There are credit crunch effects even when you do not close the banks6. This makes sense because the undercapitalized mismanaged banks have the threat of future closure, runs, or outright failure over them. Savers also know that threat is there. It turns out these effects of potential market discipline on undercapitalized banks are large—even if regulators temporize.
The microeconomic list of what is involved in a banking sector clean up is very well known and does not need extensive recapitulation here: close a large number of banks, trying to distinguish the bad from the good on market signals; recapitalize the remaining banks, conditional on them firing management, losing original shareholder equity, and writing off bad loans; put the accumulated foreclosed collateral, like real estate, back on the market as soon as possible to get a bottom to the market; take the bad loans into a government agency for evaluation and resale.
In fact this financial clean-up would be a great situation for combined fiscal and monetary operations: have the Japanese government recapitalize the banks that stay open by buying the bad loans at discount, funded through a special bond issue, and have the Bank of Japan monetize most of that specific bond issue. This would be the perfect thing for the BoJ to buy if it were to do monetary expansion (better than foreign bonds), and it would break the deadlock because it would ensure that the banks were being cleaned up and the government did not have free choice about what it could spend money on.
I would like to say one very quick additional thing about the JGBs and the banks. The point is the banks have not yet fully taken the losses they were supposed to from their years of mismanagement and bad loans. So, the losses have been floating there, rolled over to others, adding to the costs of the credit crunch, waiting to be taken. By raising the interest rate on the JGBs and forcing banks and their shareholders to finally take the losses they earned, you are finally putting paid to the bad behavior that was in the past. The money is already gone, you are going to have to pay for it sometime, so you might as well do it that way.
Over the longer term of four or five years, I think it is inevitable that foreign, particularly American, banks and financial firms will play a more important role in Japan. There already is a great deal of interest and foreign direct investment—as management gets turned out of the Japanese banks and more banks get consolidated and closed, there will be demand for even more. In essence, Japan needs the competition, the capital, and especially the transfer of skills and practices in banking from the US firms in the exact same way that the United States needed the competition, the capital, and the transfer of skills in the auto industry from Japanese firms in the 1980s7. Already some Japanese multinational firms are forcing the issue by switching to international financial partners, just as they forced the issue in the mid-1980s by taking advantage of deregulation. This is not a stable situation, especially if Japanese savers start to move money abroad as well.
What do I think is going to happen as opposed to what should be done? My fear is that Japan may finally be running out of time. Not primarily because of the public debt, although that is an increasingly illiquid market subject to shocks from downgrades—primarily because of the external international environment, political as well as economic, and because the private-sector bad loan problem has simply become so large. Rather than speaking of the sustainability of public debt, we should be concerned with the sustainability of the bad loan totals: we have reached a situation where the feedback loop of the bad loans on the economy through deflation and credit contraction/distortion has led to them being accumulated ever faster, far faster than they can be written off. Even on the official FSA numbers, which vastly understate the NPLs total, the classification IV loan totals have been rising faster than the banks' provisioning (the reverse was true until 1998). The private debt, not the public, is the one which is in imminent danger of becoming explosive.
What does running out of time mean in terms of impact? It means that if the Prime Minister or the Cabinet Office does not manage to cut a deal in the next couple of months that is a package, such as monetary loosening, bank consolidation and fiscal expansion combined, we may see a very severe contraction in Japan over the next two years-meaning negative 2 percent real growth or worse. We may see a rapid outflow of capital and a rapidly declining yen (with nothing to show for it to trading partners). And we may see support for more radical political parties because the Liberal Democratic Party, with all its Koizumi-led popularity, will have blown it. I would prefer not to see that happen. On the economics, the problematic outlook is very solvable. Unfortunately, I am less than optimistic that it will be solved.
Given my gloomy forecast of policy, why has the Japanese government done nothing? I think there are three obvious factors. The first is a game of chicken. The BoJ is playing with the MoF who is playing with the Diet who is playing with the FSA. Each one has its own agenda and is trying to make the other ones blink. You can go to a deeper political science explanation and say this is occurring because there is more independence and less coordination among these formerly unified agencies. The Cabinet Office in theory should be able to coordinate, but without a clear political base has been unable to do so as yet.
The second factor is ideology. There are people with positions of power who believe in specific theories of the world and these things matter. If you have an independent central bank and your governor believes a certain point of view, that matters, be it Greenspan deciding to gamble on the American new economy or Hayami deciding that he cannot afford to gamble on inflation. In work in progress, we have established that both fiscal and monetary policy in Japan in the 1990s became significantly more pro-cyclical rather stabilizing, and argue that ideological factors rather than institutional, international, or economic factors seem to have been at work.8
The third factor is the international political context. Until Professors Heizo Takenaka, Takatoshi Ito, Fumio Hayashi, Takeo Hoshi, Mitsuhiro Fukao, and a few others, got involved in the last couple of years, the major macroeconomists of Japan were not engaged in the public debate over policy. It was, instead, the great macroeconomists of the United States like Stanley Fischer and Larry Summers, and there was ample political reason why they as Americans (and officials) were not going to be listened to. If you put these three factors together it is a sorry tale. However, I do not find it a great mystery why action has not been taken.
How do we get beyond this now that the American economists are out of it and some of the Japanese people and scholars have picked it up? The answer is that the US government has to continue making the case that Japan is playing with some very seriously bad outcomes. There is a lot of loose talk, starting five or six years ago through the present, from Japanese and US commentators suggesting that what we really need is a crisis to shake things up in Japan. There are two big mistakes to this kind of thinking. The first is that crises have a way of getting out of hand—the responses that are held in reserve for when a crisis comes turn out to ineffective or not even implemented once things turn sour. The second is that we know there are examples of countries that manage to execute reforms and change policies without crisis—for example, New Zealand in the 1980s or the United States in the 1990s.
All we can hope to do, since Japan remains a relatively closed power elite in terms of who makes the decisions on these policies, is to try to influence those specific decision makers—that is why it is worth continuing to yell about these issues, even if there is not that much new to say analytically, just new political tactics and assessments of threat to Japan from inaction. It would be worthwhile considering issuance of an explicit US government threat to the Japanese government of what will happen if the yen were to go down unaccompanied by the economically and symbolically important bank reforms.
The problem is that trade threats are no longer feasible under WTO, and might unsettle the already difficult efforts to launch a new round. The threat in question would have to come from another part of government affecting another issue area. Perhaps we should consider more public snubbing of Japan in favor of China, or more shutting Japan out of discussions having to do with US Asian policy at a working level, given that Japan is making our foreign policy tougher? Alternatively, in the economic sphere, we could consider encouraging devaluations from the countries surrounding Japan to counteract their efforts (though that has some negative feedback on the United States), and the downgrading of yen-denominated assets.
The United States has a clear national security stake in the Japanese economic situation. A financial crisis on the order of 1998 in Japan will draw capital out of Asia, and scare money into the United States from around the world. The intermediated credits of bank loans and FDI to industry throughout East Asia will be cut off. It is likely that the direct economic effects on the region will be smaller than the surprisingly limited effects of Japanese contraction in 1998, because the countries involved are both off of exchange rate pegs and are less leveraged.
The political fallout, however, is likely to be much higher: frustrated middle-classes in Korea and elsewhere will complain that Japan is making them suffer again without having reformed when Korea, et al, have tried; China will play this opportunistically, featuring the fact that they maintained their peg throughout recent years; and the US Congress will be less willing to countenance what looks like a competitive devaluation while unemployment continues to rise. This will be destabilizing in the region, and work to China's advantage. Everyone will say that it is Japan's turn to bear the costs of its own adjustment, and everyone will be largely justified.
In the end, even if the risk of outright crisis in Japan remains below 50 percent in the near term, given the country's passive-aggressive savers, it seems silly for policy not to pre-empt that sizable possibility of a very bad outcome—especially since the pre-emptive policies are also ones that would improve the current performance of the Japanese economy. Sad to say, the Koizumi administration may end up being seen by the United States and other Western decision-makers in the same way in economic policy that Arafat's Palestinian Authority is seen in security policy: it does not matter whether the government cannot or just will not deliver on needed policy commitments, the government becomes simply irrelevant. Then the discussion of "Japan passing" will have real force behind it.
3. It is also political in the US, in my opinion, with Lindsey trying to widen his turf against Treasury and particularly to cut in on John Taylor (whom he views as a rival). With September 11, it is difficult for Lindsey to find areas where he has such room for political maneuver and it seems Japan popped up.
5. See Adam S. Posen, Restoring Japan's Economic Growth, IIE (1998), Chapter 3.
Op-ed: Will the Crisis Create a New Japan? March 16, 2011
Op-ed: Japan Must Dip into Its Rainy Day Fund March 24, 2011
Speech: Seven Broad Lessons for the United States from Japan’s Lost Decade March 26, 2009
Policy Brief 04-6: What Went Right in Japan November 2004
Working Paper 03-9: It Takes More than a Bubble to Become Japan October 2003
Working Paper 07-4: Industrial Policy, Innovation Policy, and Japanese Competitiveness May 2007