Nothing to Fear but Fear (of Inflation) Itself

Policy Brief
October 1999

Adam S. Posen is senior fellow at the Institute for International Economics, and the author of Restoring Japan's Economic Growth (September 1998; Japanese translation, Toyo Keizai publishers, May 1999), as well as numerous other works on macroeconomic policy and on central bank independence.

In the months after it took office, the Obuchi government adopted many of the measures necessary to reverse the mistakes of the previous administration. These included large scale fiscal stimulus, a great share of which is in housing tax cuts, and funding of much of the additional public debt by short-term bonds. The activities of the Financial Supervision Agency to inject capital into weak banks on a strictly conditional basis, to close down or nationalize bankrupt banks, and to improve accounting have been real progress as well. Hence, the highly positive growth rate of Japan in the first half of 1999, the faster rise in the Nikkei than in the S&P 500 this year to date, and the disappearance of the “Japan premium” for international Japanese banks, all should not have come as a surprise to financial observers.

These positive results came readily because nothing fundamental has changed about Japan's ability to grow at close to 2 percent a year; the Japanese recession of the 1990s persisted and grew worse solely because of mistaken macroeconomic policies.1 Restoring positive growth was as simple as correcting fiscal and financial policy mistakes. Desire for the great should not make us fail to recognize the good—even if the current Japanese recovery falls short of the full rate of growth which could be attained through long-term structural reform, the current near 2% rate of growth is both a major improvement and sustainable.

Restoring positive growth was
as simple as correcting fiscal
and financial policy mistakes.

This policy brief explains in detail the economic reasoning why these policy shifts have had a significant positive impact, and why the importance of some negative factors for the prospects of Japanese recovery should not be overestimated. The policy agenda to assure a sustained recovery in Japan, however, is not yet completed. A responsible supplementary budget in Japan, consisting of housing and income tax cuts, is needed to finish the task in fiscal policy. More urgently, the mounting risks to the Japanese recovery from the remaining area of misguided policy—the course of monetary policy action resulting in deflation, as signified by the yen's recent appreciation—must be addressed.

The Sun Also Rises

Obviously, the Japanese economy cannot continue to grow at the torrid pace of nearly 8% annualized GDP growth at which it did in the first quarter of 1999. Still, the 0.9% annualized rate which followed in the second quarter confirms that enough has gone right in the Japanese economy to overcome any “givebacks” and cuts in public spending which were expected to drag down performance. There are many positive factors which must be given their due, and some negative factors which must be put in perspective.

The first positive factor is that when large fiscal stimulus was finally tried this year, it worked. In particular, the tax cuts for new primary housing and the income tax cuts have had a large effect, with spillover benefits for durable goods orders. This proves that people are not so anxious about the future of Japan as some budget hawks in the LDP and the Ministry of Finance have claimed—if they were, the tax cuts and the initial payments for public works would have been saved rather than spent. Instead, there has been a multiplier of greater than one on the fiscal stimulus so far this year, with most of the growth coming in domestic consumption, meaning that fiscal expansion has fed self-sustaining growth.

Second, the benefits of financial reform are just beginning to be felt. The conditional capital injection into the biggest banks was not completed until the end of March. Since then, people have switched more of their savings out of cash back into the banking system and the Japan premium has disappeared for those large banks remaining in international interbank transactions. Both of these developments indicate market faith in the clean-up, as far as it has gone. As would be expected with changed incentives, the re-capitalized banks have stopped making or rolling over bad loans. This combination of more available loanable funds and capital reduces the uncertainty for banks, stimulating productive lending and reducing risk-spreads on interest rates.

The policy agenda to assure a
sustained recovery in Japan, however,
is not yet completed.

Third, there are lasting structural improvements being made by these policies. The housing and income tax system still needs vast improvement—and some suggestions for using the upcoming supplemental budget to make further reductions are given in the next section—but the benefits of giving average Japanese more control over their own economic decisions have been made clear. On the financial side, the improved supervision of the banking system, the permanent move to transparent consolidated accounting, the efficiency gains from closing and merging banks, and the import of financial skills through foreign investment, all will enhance the functioning of the Japanese economy for years to come.

While these benefits have been underestimated, misplaced fears have been taken too seriously. The historically high (and rising) rate of Japanese unemployment is often cited as a drag on the economy, both directly and for the precautionary saving it is said to bring. Yet, as companies lay workers off, they gain in overall efficiency and they reduce uncertainty for the remaining workers about the future of the company. Unemployed workers continue to spend on basic consumption, even with Japan's too-small safety net, in part through family transfers. For all of Japanese society's distress about unemployment, its effects on growth should not be exaggerated. Remember that continental Europe has managed to combine rising unemployment and positive growth for years—not a desirable model, but one that shows growth can co-exist with labor market adjustment.

The real significance of the yen's
appreciation against the dolar is as
an indicator of continued domestic
monetary tightening in Japan.

There is also too much concern about the recent rises in 10-year Japanese Government Bond interest rates and in the yen-dollar exchange rate. Investment decisions by companies depend on many factors, not just interest rates and exchange rates. These include the economy's current and expected growth rate, the company's own balance sheet, the health of the company's banks, and the price of assets. All of these are improving at the moment, and their positive effects on investment should more than offset any foreseeable interest rate rise.

A similar statement can be made about the Yen, that many other factors feed into companies' profits besides the exchange rate, except it is a stronger statement because less than 12% of the Japanese economy is exported. There certainly is a level of yen per dollar beyond which the effect on exports can overwhelm domestic growth, but any rate currently expected has not surpassed that level.2 The real significance of the yen's appreciation against the dollar is as an indicator of continued domestic monetary tightening in Japan.

Japan Still Needs a Supplemental Budget Now

Japan does need additional fiscal stimulus through a supplemental budget now. The Diet should pass an additional round of large housing and income tax cuts in September's session. Fiscal reform can wait until Japan has had a couple years of positive growth. A supplemental budget is necessary to avoid repeating the disastrous mistake of April 1997: cutting the structural deficit by raising the consumption tax and reducing public spending before the recovery gained speed. Without additional stimulus measures, public spending is set to fall precipitously (by 4% of GDP) by end of 1999.

One can take the analysis further—keeping the momentum of government spending going until mid-2000 will have ever larger benefits as the quarters of positive growth accumulate in people's confidence and planning. A stimulus program with a true spending content (mamizu) of 3% of GDP, and not just false claims of spending, led to sustainable growth.3 The insistence of some LDP members outside of the government on immediate fiscal reform has been proven unjustified, and would be destructive if implemented.

Of course, the structural budget deficit will have to come down eventually, and the Japanese government cannot keep spending huge amounts indefinitely. The goal should be to spend the minimum amount in the rest of 1999 that will still be felt and noticed by most Japanese, and to bring down government spending slowly over the next several years.

What Japan really needs is tax relief
for those average Salarymen who currently
pay too much of the tax burden.

Therefore, a supplemental budget in the realm of 5-8 trillion yen (1.0-1.5% of GDP) mamizu is called for, which should be big enough to have clear effects. This also will make sure fiscal stimulus declines only a few trillion next year so that the negative effect will not offset growth, and people will have time to adjust. By mid-2000, with four or five quarters of positive growth in hand, the economy should be strong enough to withstand a little cutback in government stimulus, if it is not too abrupt.

The entire 5-8 trillion yen stimulus in the supplemental budget should take the form of additional housing and income tax cuts. In general, tax cuts reach every Japanese person with no delays in implementation. They are structural reforms for the economy at the same time as they stimulate. They have much larger multiplier effects on the economy than old-fashioned public works when they are credible. This can be seen by the very large response in the first quarter of this year to tax cuts for new primary housing. To make the new tax cuts credible to the public, they should be permanent revisions in the tax code.

Both housing and income taxes are extremely unfair in Japan, as well as inefficient. Many of those excluded from paying income tax are simply people who call themselves “farmers” or “small business owners”, not people truly in need. The last income tax cut was in the right direction, but helped the higher salary people most. What Japan really needs is tax relief for those average Salarymen who currently pay too much of the tax burden. There should be a cut of 25% in all income tax withheld from people's paychecks. This reduction would total around 5 trillion yen per year, and require no time or cost to implement.

Similarly, high housing tax rates, combined with land use restrictions, unfairly put good housing out of reach for many average Japanese. The small home mortgage tax credit of five years ago, and last December's write-off for construction of new primary residences, are steps in the right direction. The Diet should go three steps further: they should give three years to buy a new residence, not one, for people who claim a tax credit from selling their houses at a loss; they should increase the mortgage tax deduction for primary residences; and they should extend the tax credit for people buying new residences to people improving their homes.

The entire 5-8 trillion yen stimulus
in the supplemental budget should
take the form of additional housing
and income tax cuts.

These housing tax measures will add liquidity to the Japanese real estate market, because people with accumulated losses will be willing to sell their homes, and individuals scared by the experience of what happens when mortgages rise in real terms will be more willing to buy. An increase in housing liquidity will have very beneficial effects on asset prices and on the banking system.

Fiscal reform can be achieved later by cutting public spending while leaving taxes low. That combination of tax cuts first followed by spending cuts is what Reagan followed by Clinton did for the U.S. to great benefit. This sequence forces the public sector to shrink over time, and it is better to cut spending than to raise taxes. This sequence is also the only lasting way to reduce government expenditures. Historical evidence shows that only those governments which cut spending keep deficits down.4 Those governments which simply raise taxes always either lower them later or increase spending to match the higher revenues.

The Japanese political system with its over-representation of rural districts favors wasteful public works and construction. If the current government wishes to earn the vote of the cities of Tokyo, Osaka, and so on, and thus be popular with the majority of the Japanese people, the government will concentrate the supplemental budget on tax cuts instead. This is probably a good political investment in the long run, even if it means a little more difficulty in the budget negotiations. For the Obuchi government to pursue (mostly) tax cuts rather than wasteful public works would be a clear message of leadership to the Japanese people and to the world at large.

People ultimately judge central banks
on policy outcomes. Deflation is not an
outcome which will build faith
in any central bank.

Fears of a rise in interest rates are no excuse not to cut taxes in this permanent, structurally-sound, manner. Interest rates on long JGBs have already risen in anticipation of the supplementary budget because the government has been preparing markets for months, and a package of more than 10 trillion yen is expected. Thus, rates should not rise much higher and may even fall when the budget actually comes. More importantly, as discussed above, the effect on investment of such a rise in interest rates will not be large at all because interest rates are only one of many factors in company's decisions about whether to invest.

Avoiding One Last (Monetary) Mistake

The failure of the Bank of Japan to take sufficient steps to fight deflation represents the largest continuing danger to the Japanese economic outlook in 1999 and beyond. The contrast between the progress on the fiscal and financial policy fronts and the ongoing monetary mistakes, and between the Bank of Japan's deflationary stance and developments in other central banks around the world, could not be sharper.

It is true that nominal short-term interest rates have been near zero for some time, and that the BoJ has committed to “continue the zero interest rate policy . . . until deflationary concerns subside.”5 Nevertheless, real interest rates remain high (greater than 2%) for an economy with enormous unused capacity, and money supply growth has been limited (less than 5% in base money). Worse, real interest rates have been rising, given the negative trend in prices, and the yen has been appreciating of late against the dollar (see Figure 1, at end). The effect of central bank policy on an economy is the combined effect of nominal interest rates, real interest rates, and exchange rates. For a central bank to say it has done all it can by dropping its own nominal short-term rates is disingenuous.

The Bank of Japan remains capable of raising inflation—and reducing real interest rates—by printing yen to buy Japanese Government Bonds (JGBs) and U.S. dollars. To see why this is the case, some practical matters about monetary policy must be clarified. Even though a central bank cannot directly control long-term interest rates and the exchange rate, which are set by the market, it can significantly influence them. It is not necessary for the financial system to be in perfect condition, or for money demand to be perfectly stable (admittedly neither the case in Japan today), to get lending and exchange rates moving in the desired direction.

The only concern is that a central bank would not be able to hit an announced target accurately on a precise timetable. Yet, this is an unreasonable worry because central banks never precisely hit their announced targets, even when times are good—research shows that central banks which set numerical targets, even if only to miss them and then explain why in terms of ultimate goals, tend to gain more credibility from their transparency than banks which refuse to announce a target.6 People ultimately judge central banks on policy outcomes. Deflation is not an outcome which will build faith in any central bank.

The failure of the Bank of Japan
to take sufficient steps to fight deflation
represents the largest continuing
danger to the Japanese economic
outlook in 1999 and beyond.

The best way for the Bank of Japan to counter deflation is to be transparent about its intent, and let private sector expectations do some of the work for them—in other words, to announce a positive inflation target.7 Of course, the announcement itself will not create inflation, but it will provide a nominal anchor for expectations to hold on to as monetary policy turns expansionary.

If agreement cannot be reached on an appropriately positive level for this inflation target, as unfortunately seems to be the case judging from released Monetary Policy Board minutes, the Bank of Japan should at a minimum replace its April interest rate commitment with one to an actual policy goal. The Bank should shift from a target of maintaining the level of nominal short-term interest rates (an instrument) to reducing the current degree of monetary tightness (an outcome). This would involve counteracting through unsterilized intervention any further rapid rise in the yen, as well as expanding the money supply through buying JGBs, in addition to keeping the nominal short-term rates it controls at zero. Sterilized intervention does not alter domestic monetary conditions or the price level, and therefore allows real interest rates (and the yen) to continue to rise.

The Bank of Japan's Monetary Policy Board statement following its meeting of September 21, however, made clear that currently the Bank has no intention of so doing. In fact, the Bank appears to being committed to doing the opposite—sterilizing whatever minimal effect on prices would arise out of exchange rate intervention, and targeting nothing but the short-term nominal interest rate. The Bank clearly fears more from the good news of signs of financial reform and consumer recovery than from the downside risks of deflation. This is a dangerously mistaken assessment.

Japan in the 1990s has reacquainted us with all the arguments from the 1930s of why deflation is a bad thing: it encourages people to withhold consumption because they expect prices to fall. It raises the real burden of outstanding debt on already borrowed loans. It puts pressure on the financial system. On the other hand, just from looking at the other nations of the OECD it can be seen that a temporary inflation rate of 2-3% has minimal or no costs to growth.8

The theoretical concern, which seems to be of greatest concern to the Bank of Japan at the moment, is that any central bank monetization of government debt would erode the credibility of the central bank. Public demands by Japanese cabinet officials last December through February for bond purchases have given the Bank the impression that it would be seen as caving in if it pursued this policy. Central bank independence is, however, a means to an end of good monetary policy, not a blind commitment to refusing any policy request, no matter how reasonable, just because it happens to come from elected officials.

Central bank independence is, however,
a means to an end of good monetary
policy, not a blind commitment to
refusing any policy request . . .

While the Bank of Japan is indeed newly independent of Ministry of Finance control (since April 1997), it is not a maiden whose reputation will vanish overnight if it is touched once. In fact, central bank credibility is not as fragile as some would make it out to be. The German and Swiss central banks, the two most credibly anti-inflationary central banks in the world, have set long-term inflation goals of 2% rather than zero. They have also flexibly accommodated times of weakness in their economies by temporarily raising that target level.9

Moreover, the effect of money creation on inflation and on credibility is a matter of the economic environment. Economies where prices are falling and unemployment is rising are much less fertile grounds for inflation than economies which are pushed into operating above capacity, The current Japanese economy—despite the very real good news I mentioned at the start—is far from overheating. One can only be burnt by touching an oven if the oven is on. The reality in the current Japanese context is that the Bank of Japan would have to print even more money, and buy even more JGBs, than would typically be necessary in order to have the same effect on prices.

It was an independent central bank, the Federal Reserve, which made the American depression of 1929-33 “Great” by overestimating the dangers of expansion, and underestimating the dangers of deflation. That same central bank lost much of its independence into the 1950s, beyond the needs of war finance, because the results of its earlier deflationary policy eroded support for its independence. Let us hope that the Bank of Japan recoginizes that credibility is built on performance not symbolism, and becomes more aggressive in fighting deflation in the near future. Otherwise, the very real progress that Japanese economy is making due to good fiscal and financial policy might be offset by a monetary mistake.

Figure 1. Monetary Situation


1. The arguments in support of this analysis are given in Restoring Japan's Economic Growth.

2. Simon Wren-Lewis and Rebecca Driver argue that an equilibrium exchange rate for the yen is at 90 to the dollar (see the Institute study, Real Exchange Rates for the Year 2000, 1998), and Keidanren has expressed an ability to live with a dollar-yen rate of around 110 (likely indicating an actual ability to cope with a yet stronger yen). Such a rate would still leave Japan with a substantial current account surplus on the order of 2% of GDP.

3. For a discussion of past fiscal policies in Japan in the 1990s, especially of the implementation of stimulus plans that were too small and less than announced, see Restoring Japan's Economic Growth, Chapter 2.

4. See the studies for the IMF by Alberto Alesina and Roberto Perotti summarized in the May 1996 IMF World Economic Outlook.

5. Bank of Japan, “On the Current Monetary Policy,” Monetary Policy Board Public Statement, September 21, 1999.

6. For a discussion of central bank targets and transparency, see Kenneth Kuttner and Adam Posen, “Does Talk Matter After All? Inflation Targeting and Central Bank Behavior,” (PDF file) Institute Working Paper 99-10 (September).

7. I advocated a 3% target for Japanese CPI at a two-year horizon, then to be reduced to 2%, in Restoring Japan's Economic Growth, Chapter 5.

8. One of several econometric cross-national studies documenting this fact is Michael Sarel, “Nonlinear Effects of Inflation on Growth,” IMF Staff Papers, March 1995.

9. Thomas Laubach and Adam S. Posen, “Disciplined Discretion: Monetary Targeting in Germany and Switzerland,” Princeton Essays in International Finance, December 1997.

More From

Adam S. Posen Senior Research Staff