Escape from Global Deflation, A Commentary By Kenneth Rogoff, Economic Counsellor and Director, Research Department, IMF

July 17, 2003

Escape from Global Deflation
A Commentary By Kenneth Rogoff
Economic Counsellor and Director, Research Department
International Monetary Fund
Nihon Keizai Shimbun
July 17, 2003

The International Monetary Fund has recently issued a much-discussed research paper on global deflation. Although we view the risk of deflation in the United States as low, our indicators suggest that the prospect cannot be entirely dismissed, and the same would be true for the euro area as a whole, although there are higher risks for Germany. Given apparently tepid growth now in the United States, and with Europe virtually stalled, growing excess capacity is putting downward pressure on prices. Despite the recent rise in long-term rates in Japan, interest rates and inflation rates will remain near or at fifty year lows throughout the globe, and a sufficiently strong negative demand shock could tilt the balance towards worldwide deflation.

To be sure, there are significant differences between the United States, the Eurozone and Japan. Deflationary expectations are deeply embedded in Japan—people expect deflation to continue for at least two to three more years. This despite an economy with among the highest (and rising) government debt levels in the OECD, and with potentially huge future costs of recapitalizing the banking system. Long-term deflation expectations persist despite the predictable drop that is coming in Japan's savings rate as its baby boomers retire, and despite the obvious future pressures on the government to inflate. Things are different in the United States, though less so than one might think. It is true that, by a variety of measures, inflation expectations in the United States had been relatively stable around 2 percent, but they have been dropping notably of late. The picture is similar in Europe. It is also true that the financial sectors in the United States and Europe are much stronger than in Japan, overall. However, Germany's banking system is certainly vulnerable if the downturn continues, and its insurance industry is already experiencing widespread problems.

So how to get out of global deflation in the unlikely scenario that it happens? For Japan—where despite recent developments, the IMF reports warning that deflation could easily still worsen remain valid—many observers believe that only way out of its deflation is through yen depreciation. At some level this must be right. Whatever route the Bank of Japan ultimately takes to conquer deflationary expectations—whether it be buying stocks, JGBs, or foreign exchange—the yen will likely fall sharply. Such action may be seen as manipulating the exchange rate in beggar-thy-neighbor fashion. After five years of deflation, however, the rest of the world—and certainly the IMF—should not complain too much about such a policy.

But if all the major regions were mired in deflation, they cannot all simultaneously depreciate their exchange rates, so this door out is closed. Would this be fatal? No, it should not be. A clear communication strategy that all three major central banks are deeply committed to restoring positive inflation, would go a long ways towards putting a floor under deflation expectations. It is not necessary to adopt any narrowly-construed version of inflation targeting—indeed escaping from deflation is sufficiently tricky it would be very hard to deliver on a target. But central banks can and should specify their long-term (five- to ten-year) inflation objectives. Aggressive quantitative easing worked in the 1930s, when those countries that abandoned the shackles of the gold standard and inflated their economies were largely far more successful than those that clung to gold. If all three major central banks acted in concert, there would not have to be dramatic exchange rate movements, although admittedly the yen might still depreciate since Japan is now in the deepest hole.

I admit that the central banks of the world would have a hard time calibrating their escape from deflation. Quantitative easing does eventually lead to inflation, but the relationship is not nearly as smooth or reliable as economists such as Milton Friedman once believed. Modern central banks rightly prefer to work with interest rates. Of course, as interest rates approach zero, this method begins to fail. Printing piles of currency will surely end deflation: the problem is how to avoid overshooting and ending up locked in high inflation. Should people in Japan worry that an overly aggressive exit from deflation will lead to the kind of horrific inflation levels that Japan suffered after the War? I don't see it, especially given the large output gap. Inflation might temporarily spike, but as soon as it does, interest rates will shoot up above zero, the BOJ will be out of its liquidity trap, and will once again be able to use standard monetary policy tools. Any inflation overshoot should be fairly temporary and relatively manageable. One has to acknowledge there are many risks to this strategy. However, there may be even greater risks to trying to live with deflation indefinitely.

Some commentators have opined that there is no way for an economy to escape deflation without fiscal easing. Given Japan's already high government debt levels, Japanese are rightly wary of outsiders telling them to embrace further fiscal profligacy. Nor is advice to run bigger deficits likely to resonate in the United States, where government deficits are already approaching five percent per year, or in Europe where the largest economies are already routinely breaching the three percent limits of the eurozone's increasingly strained Stability and Growth Pact.

It is not essential to support monetary easing with current fiscal easing. Yes, when the central bank buys long-term government bonds, it is directly reducing the nominal quantity of interest-bearing debt held by the public. Because the government's medium-term fiscal position is strengthened, this means that taxes can eventually be lowered. (More precisely, given Japan's formidable medium-term fiscal difficulties, it means taxes will not have to be raised as much.) Expectations of lower future tax hikes should provide enough of a lever for aggressive quantitative monetary easing to be effective in fighting deflation; there is no need to relive the failed fiscal experiments of the 1990s.

Implicitly I am assuming that when the BOJ buys JGBs, it is relieving the government's budget constraint. Is this right? Why, the reader may ask, should the Ministry of Finance care whether it is making interest payments to Japanese citizens or to the Bank of Japan? Isn't the BOJ now fully independent? Yes, it is, but BOJ's profits must eventually be remitted to government. For purposes of government debt management, having the BOJ purchase government bonds is not so different than retiring them. I am not suggesting a special form of accounting for Japan. This point applies with equal force to the United States Federal Reserve and even to the European Central Bank, whose profits are remitted pro rata to its member governments.

The fact that we should focus on the consolidated government balance sheet also explains why it is nonsense for the public to worry excessively when the Bank of Japan suffers capital losses on its holdings of long-term government bonds. The Bank's loss is the Ministry of Finance's gain, and both are ultimately owned in full by the Japanese taxpayer. The independence of the Bank of Japan is of the utmost importance, but if capital losses on JGBs were the only problem standing in the way of escape from deflation, there should be many ways to deal with it.

In sum, the odds of global deflation are not high, but if it were to set in, the world's central banks have the tools needed to address the problem. The path out, however, is neither easy nor riskless. Let's hope that pre-emptive action in the United States and Europe staves off the problem. The best thing for Japan would be to have its trading partners growing crisply again as it tackles its own home-grown problems.


Public Affairs    Media Relations
E-mail: E-mail:
Fax: 202-623-6278 Phone: 202-623-7100