The need for a lender-of-last-resort function is one qualification to the discipline of inflation targeting, but it is a necessary one. There is a second line of argument that seems to me much less compelling. It starts with the fact that monetary policy necessarily affects future inflation rates, not the current rate: That has already been determined when the open market committee meets. We also know that whatever funds rate target is chosen, all kinds of others forces -- anything that happens to the real economy -- will affect next quarter's rate of inflation, or next year's. So we would like to forecast these other forces as well as possible and take them into account.

There is nothing wrong with this logic, but how useful it is depends on how good we are at forecasting the non-monetary determinants of prices. In fact, inflation forecasting is notoriously one of the squishiest areas of economic statistics. In this situation, it is all too easy for easy money advocates to see a recession coming and rationalize low interest rates. They could be right -- who really knows? -- and in any case we may not know enough to prove them wrong.

So I am skeptical about the argument that the subprime mortgage problem will contaminate the whole mortgage market, that housing construction will come to a halt, and that the economy will slip into a recession. Every step in this chain is questionable and none has been quantified. If we have learned anything from the past 20 years it is that there is a lot of stability built into the real economy.

To me, inflation targeting at its best is an application of Milton Friedman's maxim that "inflation is always and everywhere a monetary phenomenon," and its corollary that monetary policy should concentrate on the one thing it can do well -- control inflation. It can be hard to keep this in mind in financially chaotic times, but I think it is worth a try.