The foreign exchange value of the U.S. dollar, that delight for American travelers abroad and the bane of manufacturers at home, remains an elusive target for economic forecasters and currency traders alike.

There is a consensus that its value, which has soared since 1980, must come down, but there is no agreement on how far or how fast.

Compared to the currencies of other industrial nations, the dollar apparently reached a peak last winter at a level 54 percent higher than its average value in 1980.

Between then and August, the dollar dropped about 12 percent, pleasing the people concerned about the impact of the highly valued currency on the nation's deteriorating trade balance.

However, with the prospect of stronger economic growth and possibly higher interest rates in the United States, the dollar has been climbing again recently, particularly against the West German mark.

By the second half of 1986, some analysts expect that the dollar's decline, if resumed, will have begun to reduce the nation's trade deficit.

Other analysts, less sanguine, note that the dollar's value is still higher than its average for 1984 and that further large declines would be needed to have a significant impact on trade.

That is true, for instance, in the case of the Canadian dollar -- Canada accounted last year for nearly 18 percent of combined U.S. imports and exports, the most of any nation.

Compared to Canada's dollar, the American currency is down only about 1 percent from its peak and is about 6 percent higher than its average for last year.

Compared to the Japanese yen, the dollar is down about 6 1/2 percent from the peak and is still about 1 1/2 percent higher than its 1984 average.

Given the lags between changes in exchange rates and the impact on trade flows, it may be that last winter's dollar highs, which were quickly erased, had little effect on relative prices of traded goods and services, according to some experts.

If that is the case, then the fact that the dollar is still above its 1984 average is a more important point than the decline in the dollar from the peak, these experts argue.

Any future improvement in the trade balance will be dependent on either future declines in the dollar or other factors -- such as movement toward more nearly equal rates of economic growth in the United States and other major trading nations, this group of analysts says.

There is a wide variety of reasons given by economists as to why the dollar rose so far, but few if any are confident that they can explain the currency's behavior adequately. None of the theories they have used about exchange rates in the past seem to have worked very well recently, they say.

But there is widespread agreement that the dollar must come down because otherwise the United States will be faced with such large trade deficits and such a large and growing debt to foreign investors that the rest of the world will not be willing to finance them. Painful medicine?

At a conference on the dollar last month, sponsored by the Kansas City Federal Reserve Bank, economist Paul R. Krugman of MIT argued as follows:

"The current strength of the dollar, given that there are only modest differences between real interest rates that is, rates adjusted for expected inflation in the United States and in other industrial countries, amounts to an implicit forecast on the part of international investors that the dollar will decline only slowly, at a rate averaging less than 3 percent per year for the indefinite future.

"A dollar decline this slow would ensure huge U.S. current account deficits deficits in international transactions other than long-term capital investments for more than two decades.

As a ratio to exports or GNP, U.S. indebtedness to foreign countries would reach a level comparable to that of Brazil or Mexico.

"Whether one believes the strong dollar is sustainable depends on whether one views this level of U.S. external indebtedness as feasible. If as I believe, such a level of of debt is not feasible, at some point the market will realize that the dollar must fall more rapidly than it now expects. When this happens, by the usual logic of asset markets, the dollar will fall immediately."

The U.S. current account deficit, $102 billion in 1984, is forecast by Wharton Econometric Forecasting Associates to be $132 billion this year and $146 billion in 1986 -- assuming the value of the dollar averages 10 percent lower next year than this year.

Estimates included in a forthcoming book by Stephen Marris, senior fellow at the Institute for International Economics here, indicate that it would take a decline in the dollar of about 40 percent from its peak to bring the current account back into balance.

The institute's director, C. Fred Bergsten, told the Fed conference that Marris' figures show "that almost one-half of all world savings generated outside the United States would have to be moving into the dollar by the end of this decade to sustain the exchange rate at its current level."

Federal Reserve officials, including Chairman Paul A. Volcker, are concerned that a rapid drop in the dollar could reaccelerate inflation in the United States. For that reason, they would like to see it come down at a moderate pace.

But Bergsten believes that, even given the threat of more costly imports and therefore less price competition for U.S.-made goods here, the dollar needs to come down as fast as possible. "Early movement would help head off the risk of a 'dollar strike' by foreign investors and a renewed surge of U.S. interest rates . . . and help head off the risk of a protectionist outbreak, which could disrupt the entire world trading system."