In today's Wall Street Journal, Greg Ip takes a trip to exchange-rate policy's past.

On the morning of March 2, 1995, Ted Truman, then top international staffer at the Fed, was getting reports of massive dollar sales, some triggered by derivative strategies, driving the U.S. currency down sharply against the deutsche mark and yen. Bond yields were rising.  Mr. Truman went to see Mr. Greenspan and recommended the Fed and Treasury intervene in the markets to buy dollars.  "I don't think it's going to do any good," Mr. Truman recalls telling Mr. Greenspan, "But by not being there we are saying we totally don't care what the conditions of the markets are."

Interesting theory, that, but seemingly consistent with the belief that monetary policy is as much about what people think you will do, as it is what you actually do. (In the short-run anyway -- see here and here, for example.)  Ip, in any case, says it worked.

Mr. Greenspan agreed, and that afternoon the Fed and the Treasury waded in, buying $600 million of marks and yen.  The next day it repeated the action, joined by thirteen central banks.  The dollar stabilized.

Maybe, but I think I'll need a little more convincing about the cause and effect.  Here's what the data look like.

Exchange_rate
Just casually, I'd say that looks like a picture supporting the conclusions of this article -- "On the Rotation of the Earth, Drunken Sailors, and Exchange Rate Policy" -- by my colleague Owen Humpage.  Here's the abstract:

A growing number of observers seem to believe that official foreign exchange intervention offers a useful tool for managing the dollar's descent. In particular situations, official transactions can sometimes produce temporary changes in exchange rates, but intervention does not permit countries to avoid or substantially modify trends in the movements of their exchange rates. At best, intervention is of very limited value.

That probably explains Mr. Truman's sense that the intervention was not likely "to do any good."  Nonetheless, he says we should expect more of the same.

Mr. Truman, now a scholar at the Institute for International Economics in Washington, predicts that in the next five years, the U.S. will have to intervene again "either because it's a period of disorder or because we can't withstand the political criticism from our partner countries."