All eyes are now on the budget. As President Reagan approaches the deadline in drawing up his tax and spending proposals for fiscal 1984, each day brings new and fascinating reports of what they may include.

The deliberations over fiscal policy are important. But decisions that are probably of at least equal importance have to be made soon about monetary policy, too.

Next month the Fed is to give Congress its monetary policy report and objectives for 1983. There is uncertainty about what the Fed will say. Monetary policy has undergone dramatic changes since the last policy statement to Congress six months ago, with the virtual abandonment--at least for the time being--of the targets for the narrow M1 money measure. This measure, which includes cash and checking accounts at banks and thrifts, was the centerpiece of the Fed's control of the money supply until a few months ago.

Last July, Fed Chairman Paul A. Volcker set tentative targets for M1 growth in 1983 of 2 1/2 percent to 5 1/2 percent, the same as the 1982 range. These targets have not been officially revised or revoked. But just weeks after Volcker announced them to Congress, the policy-making Federal Open Market Committee decided that it no longer made sense to judge monetary policy by what was happening to M1.

The reason for disenchantment is two-fold. First, the M1 figures have been distorted by changes in banking regulations to the point where no one knows what they mean. The purpose of controlling M1 is to control people's spending money. After the introduction of new deposit accounts at banks and thrift institutions, substantial flows of what are, in effect, savings--rather than deposits likely to be spent--are expected to show up in M1. It will be months, if not years, before analysts know what relationship this new M1 bears to total spending and gross national product. Until then, it will not be clear what growth rate of M1 is appropriate.

Volcker and other Federal Reserve Board members have concentrated in their public statements on the difficulties that banking upheavals have created in using M1 as a guide to policy. But there is another kind of reason for ignoring this money measure when setting policy.

That is that the signals M1 gave to monetary policy-makers last year were misleading, even before the changes in its composition as a result of the new accounts. The economy has ended up much weaker than the Fed expected would be the result of its monetary policy last year. During the early part of 1982, Volcker and other Fed officials gradually took their distance from the M1 figures. By mid-summer it was clear that the amount of money the Fed was aiming to supply to the economy was insufficient to support the amount of total spending--or GNP--that they and other forecasters expected.

Since then, M1 has grown way outside the official band of 2 1/2 percent to 5 1/2 percent. The economy is beginning to show some signs of recovery, but most experts still expect only a very slow upturn this year.

With M1 giving so little help to monetary policy-makers, the Fed has said it intends to pay more attention to broader measures of money--M2 and M3--that will be less affected by shifts between different kinds of deposits and savings instruments. But M2 also has problems.

It, too, was growing outside the Fed's target ranges at the end of last year, despite the evident weakness of the economy. At the fall FOMC meetings for which the minutes are available, the committee members decided to allow faster M2 growth rather than risk a rise in interest rates, if the choice came down to that.

More fundamental, the Fed cannot control M2 in the way that it can M1. It controls the money supply through management of reserves that banks have to hold against certain deposits. Most parts of M1 are subject to reserve requirements. But many M2 components are not. The Fed is thus likely to end up aiming at its M2 targets through a combination of controlling M1, which may not bear much relation to what is happening to the economy, and influencing interest rates and economic activity, which in turn affects other M2 components.

The object of controlling the money supply is to control, or at least influence, the course of GNP. But "the best way to control M2 is to control nominal GNP," one economist comments. In other words, the Fed cannot easily have any direct influence on this broad money measure, except via its influence on total spending and demand for money. In practice, with no reliable M1 figures, the Fed must depend heavily on other indicators of how the economy is faring when it decides how to operate in the money markets.

Volcker is unlikely to tell Congress that the Fed now plans to monitor money policy by looking at interest rates, inflation, growth and uenmployment rather than the money supply numbers. He may well give some new target ranges for money growth this year, with caveats about their uncertainty. But the FOMC will almost certainly continue its recent practice of mixing a heavy dose of judgment--based on the economic indicators--with whatever money numbers it produces to guide day-to-day policy-making.