Administration officials say that they're happy with the present level and stability of exchange rates and that the worst of America's huge trade deficit is behind us. Federal Reserve Board chairman-designate Alan Greenspan agrees there will be a dramatic improvement soon.

Clearly, the decline of the dollar has helped make U.S. manufacturing exporters more competitive. Recently, there has even been a trickle of exports to Japan and elsewhere in Asia from Japanese-owned plants in America. But how realistic is the prognosis for a major decline in the U.S. trade deficit without basic changes in the structure of American industry or a further devaluation of the dollar?

To be sure, many economists believe that with a further decline in the dollar, prospects will greatly improve. Former Economic Council chairman Martin Feldstein said the other day that another 10 to 15 percent drop in the dollar against the Japanese yen "would do wonders" for American manufacturers' belief in their ability to compete.

Yet too much faith may have been placed in the theory that America's huge trade deficit can correct itself through the exchange rate mechanism alone. The dollar has already dropped about 50 percent against the Japanese yen and the German mark, but most experts confess that they're disappointed with the meager correction so far, even when taking expectable time lags into account.

As Stephen S. Cohen and John Zysman, who direct the Berkeley Roundtable of International Economy (BRIE) at the University of California, point out in a stimulating new book, "Manufacturing Matters," there is the critical matter of "reversibility" when talking about exchange rates.

Cohen and Zysman are willing to chalk up as much as half of the 1984-85 trade deficit increases to an overvalued dollar. But they add: "The problem is that however much the rise in the dollar contributed to the loss of position in international markets, an equal decline of the dollar will not leave the national economy where it was before the rise began."

In his confirmation testimony, Greenspan explained that only limited progress has been made so far in bringing down the trade deficit, despite a cheaper dollar, because Japanese and other foreign producers and distributors chose to hold down prices, protecting market share at the expense of profit margins. That, he suggested, won't go on forever.

A new Brookings Institution study by economists Paul R. Krugman of the Massachusetts Institute of Technology and Richard Baldwin of Columbia University agrees that the lower dollar rate that began to show up in 1985 will be reflected in a substantially reduced trade deficit "in coming quarters."

The Brookings economists think that up to two-thirds of the increase in the trade deficit may be exchange rate-related. But they speculate that the competitive advantage American exporters might gain from the decline in the dollar so far will be offset after 1988 by an increase in productivity among some of our competitors abroad. Thus, they argue that a further decline in exchange rates is necessary.

In plain language, what they are telling us is that countries such as Japan, Taiwan, South Korea and the other star performers in Asia have consciously boosted their productivity in the export-manufacturing sector far over their national averages -- hence, well above American performance.

Cohen and Zysman come at it a different way, but arrive at essentially the same conclusion. The BRIE analysis is that Japan and West Germany have managed to create and sustain production efficiencies, and spread them into such simple goods as shoes and textiles. Thus, both Japan and West Germany continue to enjoy export surpluses in shoes and textiles, although America has trouble in keeping competitive in these areas.

Cohen and Zysman detect "a radical inability" in this country, relative to our competitors, "to apply high technology to the production of traditional goods and to maintain our competitive position by diffusing technology and know-how widely through the economy. If this is the case ... . then our trade problem is deep and serious."

If either the BRIE or Brookings authors are anywhere near right, we face a long-term erosion of American ability to compete not only in ordinary goods, but in high-tech goods -- whatever the exchange rate. The BRIE solution is to aim at a combination of "advanced technology with high-skilled labor and innovative management to create high-wage, high-productivity, flexible production capabilities." There is little in the Senate or House trade bills that focuses on that kind of forward-looking strategy. Instead, Congress would look backward toward protecting the weak and inefficient.

On July 1, in his last congressional appearance before his fatal riding accident, Commerce Secretary Malcolm Baldrige recognized that something beyond exchange-rate adjustments will be necessary to achieve an acceptable trade balance. He stressed the "central role" that must be played by American manufacturers, and said that improved competitiveness could be derived only from "increased productivity, efficiency and quality."

After a column was published here recounting that testimony, Baldrige asked his aide, B.J. Cooper, to phone me to make sure that I understood that he knew that exchange rate changes, by themselves, would not do the job. In an interview just before his death, and published July 28 in USA Today, Baldrige made it explicit this way:

"The Japanese are working as hard as they can to adjust to 140 or 150 yen to the dollar. I'd wager they're working to adjust to an even lower figure for the yen. That makes it incumbent on U.S. managers to be doing something of the same thing. So improvement of American industry -- cost-cutting and quality improvements -- must go on, and on, and on. We can't get out of it. And we shouldn't. We should lead the world. We should fight back ... . and win."