The move away from a hands-off policy toward exchange rates started with the New York Plaza agreement among the group of five last September.

It received a boost from the Bradley-Kemp Congressional Summit on the Dollar in November.

President Reagan's State of the Union address went on to ask the Treasury to study possible ways of dealing with international currency fluctuations and report back by December.

And last week the French Treasury published a paper on "reference zones" that demonstates French sympathy for the new trust of U.S. policy and should contribute to the Treasury's forthcoming study.

Meanwhile, events in the exchange market have been demonstrating once again the need for systematic management of exhange rates. The dollar has come a long way back toward a level consistent with medium-run equilibrium and adequate U.S. international competitiveness, but already fears are being expressed that it may overshoot.

This danger deserves to be taken seriously. Downward correction of the dollar will inevitably produce a temporary blip in the inflation rate as the past gains on the inflation front from an appreciating dollar have to be paid back. But if the dollar goes into a free fall, what should be a temporary blip could instead rekinkle inflationary expectations, revive wage inflation and cause us to lose the anti-inflation gains of the early 1980s.

The authorities would be in a far more convincing posture to resist the danger if, while the dollar was still too strong, they had already spelled out (and agreed on) how big a fall was needed and where they would start to resist a further decline.

Such thoughts lie behind the propsal for "target zones" or "reference zones," which was one of the two principal reform proposals discussed at the Bradley-Kemp conference and which has now been backed by the French. A target or reference zone would be a wide zone of exchange rates (perhaps plus or minus 10 percent) that are "not clearly wrong," rather than a single rate that the authorities decide is right.

This system would provide more flexibility than the old fixedrate system of Bretton Woods, in three dimensions. The zones would be wider. They could be adjusted automatically to avoid being outdated by diffewrential inflation. And the rate might be allowed to go temporarily outside the zone in response to strong speculative pressures while the authorities were deciding whether the market was reflecting fundamentals that should prompt a change in the target rather than a speculative bubble that they should resist.

A target zone system could provide the basis for internationally agreed management of exchange rates. Various instruments would be used to keep exchange rates within the target zones, once these had been agreed on. Merely publishing the zones should help, by leading to better informed private speculation than we have been seeing.

Intervention in the foreign exchange market can also be of assistance. But if the authorities are serious about managing exchange rates, they must be prepared, when the need arises, to modify their policies -- especially interest rates -- with a view to influencing the exchange rate.

Since the dollar is still too high for a healthy U.S. balance of payments, a target zone system would point to a need for further reductions in U.S. interest rates at the present time. The problem is, of course, that a reduction in interest rates could rekindle inflationary expectations unless it were accompanied by the action to cut the U.S. budget deficit, which everyone knows to be urgently needed.

This illustrates a basic truth that is one of the principal attractions of the target zone approach: that pursuit of sensible exchange-rate targets should provide a catalyst for the timely adoption of sound economic policies.

The other school of thought at the Bradley-Kemp conference advocated reintroduction of a gold standard. The basic motivation for this propsal is a distrust of government and in particular a desire to have some automatic rule to guide monetary policy and stabilize exchange rates.

Most economists think that gold involves too narrow and speculative a commodity base to provide efficient monetary control. Moreover, the proposal confronts the old problem that a government willing to be disciplined does not need such arbitrary rules, while an undisciplined government will certainly not be bound by them for any length of time. For such reasons the proposal was decisively rejected by the majority of the Gold Commission that reported to Congress as recently as 1982.

At the same time, the Treasury's forthcoming investigation should ask whether there is not a germ of truth in the argument of the gold standard proponents that might usefully be incorporated into the target zone approach.

Afterall, there is at times a problem of deciding which country should act when exchange rates approach the edges of their target zones. The gold standard claims to offer a solution to this problem: the country with a weak currency at the bottom of the band should act to restrict credit if the gold price is high and rising, and vice versa, on the argument that a high and rising gold price is a symptom of developing inflation.

While gold may not be a very adequate guide to whether the world is facing a threat of inflation or deflation, the price of a broad bundle of commodites might provide a useful indicator. I would myself need a lot of persuading that an automatic rule along those lines deserved adoption, but a presumptive rule is at least worth careful consideration.