Rumors suggest that the Bank of Japan may actually be considering adopting some form of inflation targeting, with a positive upper bound. This is good news, because it means that the Japanese are finally starting to understand the nature of their situation. But do they still get it? I am not sure, because the numbers being floated are almost surely too low. And in that case perhaps they still don't get it - because the basic logic of a liquidity trap says that the choice facing a country in such a trap is between a significant positive rate of inflation and grinding deflation. There is no middle ground.

The point is simple, but apparently hard to grasp. Suppose that the equilibrium real interest rate - the rate at which savings and investment, including net foreign investment, would be equal at full employment - is negative. (As I have tried to explain, in  Japan: still trapped , that is what a liquidity trap is all about). And suppose also that prices do not fall quickly in the face of unemployment. Then if the expected inflation rate is too small to allow a low enough real interest rate - if, for example, the economy "needs" a minus 3 real interest rate, and expected inflation is only 1 percent - the actual result will be an underemployed economy, and hence continuing slow deflation. This will soon mean that the inflation target loses credibility, and you are back where you started.

So an inflation target can work only if it is high enough that, if believed, it would produce a strong enough economy to actually generate inflation. A target that is set too low is doomed to failure.

This is why people who want to water down proposals for "managed inflation" - why can't we just aim for price stability, or 1 percent inflation, instead? - are missing the point. As I tried to say in my original piece on  Japan's trap , the deflationary pressures we actually see in Japan represent an economy that is "trying" to achieve the inflation it needs, by reducing current prices compared with expected future levels; the reason the economy is depressed is that such deflation does not come quickly and painlessly.

I know that all of this is painful to think about; policymakers are certainly not used to dealing with such paradoxical-sounding problems, and their instinct is always to go for some sort of middle ground. But the law of the excluded middle here is not some abstract professorial quibble. It is the unavoidable conclusion of any reasonable analysis.

So let me say it in capitals:

INFLATION TARGETING IN JAPAN WILL FAIL UNLESS THE TARGET IS SET HIGH ENOUGH.