With an obvious tinge of regret, the President's Council of Economic Advisers noted in its annual report last week that the Federal Reserve during 1987 continued to follow an "eclectic approach" in setting monetary policy.

Rather than using growth of the money supply as the principal guide for policy, Fed officials have been watching a wide range of economic indicators in determining whether to be more generous or less generous in supplying credit to the U.S. economy.

Even though it has produced steady economic growth with little increase in inflation over the past three years, the Fed's seat-of-the-pants, judgmental approach worries monetarist economists, such as CEA Chairman Beryl Sprinkel.

These critics of the Fed would much prefer that it paid far more attention to the money supply, which they believe is the key element in determining future levels of inflation and the gross national product.

It is plain from the tone of the report that the CEA believes the Fed was too tight-fisted in supplying money to the economy last year.

At one point last fall, Sprinkel was so concerned that slow money growth would cause a recession this year that he led a fight within the administration to put pressure on the Fed to get the money supply growing again.

Federal Reserve Chairman Alan Greenspan, who has kept an unusually low profile and made only a few public statements on monetary policy since taking over from Paul A. Volcker last August, is not likely to give Sprinkel much satisfaction on Tuesday when he makes his first semiannual report to Congress on monetary policy.

The Federal Open Market Committee (FOMC), the central bank's top policy-making group, tentatively decided in December that the link between growth of the money measure M1 and the economy is so uncertain these days that it would again not set a target for M1 growth this year. M1 includes currency in circulation, checking deposits at financial institutions and travelers checks -- so-called transactions balances that can be used to make purchases or pay bills on a moment's notice.

There is no reason to think the FOMC changed its mind when it met this month to set a policy course for the year. Greenspan will disclose the committee's action at this week's hearings before the House and Senate Banking committees.

The essential problem is that after more than a decade in which current-dollar GNP rose a fairly steady 3 percent a year faster than M1, the link was shattered early in this decade. No one has been able fully to explain why. However, as the CEA report explained, the abrupt break in inflation in 1982, the sharp decline in interest rates and the gradual ending of ceilings on what interest rates could be paid on most deposits at financial institutions probably all played a part.

When market interest rates fell, depositors were more inclined to leave money in noninterest-bearing checking accounts, which come under the M1 definition, rather than in interest-bearing savings accounts or certificates of deposit, which do not.

About the same time, Congress approved use of interest-bearing checking accounts on a nationwide basis. The negotiable order of withdrawal, or NOW accounts, spread rapidly. As they did, for the first time a portion of the transactions balances included in M1 began to pay interest, making them sensitive to changes in rates. Later, the ceilings on rates paid on NOW accounts were lifted.

A chart included in the CEA report showed one result of these changes. Between 1959 and 1981 the ratio of M1 to current-dollar GNP rose more or less steadily at about 3 percent a year. There was very little apparent sensitivity in this ratio -- known as the velocity of money -- to changes in interest rates.

But since 1981, not only did the velocity of M1 turn downward, but it also began to reflect the ups and downs in short-term interest rates.

The uncertainty surrounding M1 makes Fed policymakers extremely leery about trusting it as a guide to the future course of the economy or inflation. They do not exactly ignore it, but it is only one indicator among many that the FOMC members are watching.

The published policy record of the FOMC's December meeting noted, "While no decisions were made at this meeting, the members were not currently inclined to reestablish a {target} range for M1, given the continued large interest rate sensitivity of the demand for this aggregate and the associated wide swings in its velocity."

The velocity of a broader measure of money, M2, had been much more sensitive to shifts in interest rates all along and during the 1980s has varied somewhat less than the velocity of M1. However, it, too, has varied to the point that Fed officials are not willing to trust it very much either, though they are still setting targets for it and another measure, M3.

The implicit target at which the Fed now seems to be aiming is growth in current-dollar GNP, and so far its aim seems to have been excellent even in a time of great turmoil in financial markets and unprecedented shifts in the relationships between the U.S. economy and the rest of the world.

Over the past three years, interest rates generally were cut by a third before heading upward again in 1987, the value of the U.S. dollar fell sharply, the nation's trade deficit ballooned, hundreds of mostly small financial institutions failed, oil prices plummeted and recovered somewhat, major changes were made in personal and corporate income taxes and the federal budget deficit first soared and then dropped back. Last fall, stock prices took a historic plunge.

The month-to-month and quarter-to-quarter changes in a host of economic statistics have mirrored some of these large and often abrupt changes. For instance, since the beginning of 1985 quarterly changes in the GNP, adjusted for inflation, have ranged from a low of an annual rate of 0.6 percent to a high of 5.4 percent. Consumer prices went down at a 1.3 percent rate in one quarter and rose at a 5.3 percent rate in another. Money supply growth soared in 1986 and slowed sharply in 1987.

And yet if one steps back from these shorter-term changes, an almost hidden picture of stability emerges. After the economy's initial recovery in the two years following the deep 1981-82 recession, the annual increase in economic activity varied little: current-dollar GNP rose 6.3 percent in 1985, 5.6 percent in 1986 and 5.9 percent last year.

Take away inflation and the numbers were even flatter, with real economic activity increasing 3 percent in 1985 and 2.9 percent in 1986 and 1987. Moreover, the latest forecast from the Reagan administration is for an identical 2.9 percent rise in real GNP this year as well, though many private forecasters believe the figure will turn out to be lower, with a recession a possibility.

In the opinion of many economists, the Federal Reserve has had the major role in achieving such stable growth -- growth that has been sufficient to reduce civilian unemployment from more than 7 percent to less than 6 percent at the same time that inflation, excluding volatile energy prices, has gone up only slightly, from just below 4 percent to just above.

While monetary policy has been concentrated on facilitating just such stable growth, federal tax and spending decisions have generally been driven by other considerations, such as expanding or reducing particular spending programs and lowering marginal personal income tax rates, or simply a desire to cut the budget deficit. Certainly tax and spending policies have not been set with an eye to stabilizing economic growth or affecting inflation.

In terms of discretionary action, then, the Fed remains the only game in town and the potential for conflict with the Reagan administration is still substantial. At the moment, the administration has no public quarrel with the central bank.

Asked last week at a briefing on the 1989 federal budget if he thought the Fed was being too restrictive, Treasury Secretary James A. Baker III replied, "Well, my comment to you is simply that we feel that the current course of monetary policy being pursued by the Federal Reserve is adequate. And I'll not comment beyond that."

At an earlier briefing on the CEA report, Sprinkel said, "We have seen evidence of late -- declines in interest rates, which are welcome, and increases in {monetary} aggregate growth, which are welcome."

The potential conflict could arise out of the fact that Sprinkel and some other adminstration officials still tend to focus on fairly short-term movements in the money supply. Or it could arise from the fact the Fed and the administration may have a different idea about the appropriate future course of current-dollar GNP.

The latest administration forecast shows it rising 6.5 percent in 1988 and to 7 percent in 1989. Meanwhile, by next year inflation, as measured by the GNP deflator, would be rising about 1 percentage point faster than it did in 1987.

The Fed may not be very happy with an upward trend in current-dollar GNP that involves somewhat higher inflation.