During the last 15 years, Congress has largely lost control of the federal budget. As a result, the United States has not had a real fiscal policy but merely a fiscal result, and the country has had to depend on monetary policy to deal with ups and downs in the economy.

This dependence has made it very important that the public know what monetary policy is being pursued by the Federal Reserve in order to have some idea of where the economy is going. Unfortunately, there is no obvious or reliable way to evaluate current monetary policy, and recent developments in the economy have made this problem worse than it was in the past.

The level of interest rates has been the most common measure of monetary tightness, but the distinction between nominal and real interest rates makes this a very misleading approach. If a nominal interest is 15 percent when inflation is 12 percent, lenders are earning a 3 percent real interest rate -- that is, 3 percent after allowance for the reduction in their capital caused by inflation. High nominal interest rates often reflect low real yields and an expansionary monetary policy when prices are rising rapidly. An 18 percent prime rate is not high in a period of 16 percent inflation. The lender is earning 2 percent in real terms.

This approach would seem to suggest that the current rate of inflation should be subtracted from market interest rates to find a real interest rate that accurately reflects monetary policy. Unfortunately, it is more complicated than that. This subtraction process indicates what lenders have earned in the immediate past, but says nothing about what they might earn in the future. Current borrowing and lending decisions are based on expected real yields, which means the nominal interest rate minus the expected rate of inflation over the life of the loan. A 15 percent bond yield is very attractive to lenders if they expect an inflation rate of 5 percent during the next few years, but is equally unattractive if 20 percent inflation is expected.

There is no way to know what reates of inflation borrowers and lenders expect, so there is no way to measure the real interest rates that determine current borrowing and lending decisions. The recent decline in nominal interest rates may represent an easing of monetary conditions, or merely a reduction in the expected rate of inflation due to an approaching recession.

If interest rates are an uncertain guide to monetary conditions, rates of growth of the money supply may be an alternative. The problem is, which money supply? Economists used to think that they knew what money was; namely coins, currency and checking accounts, known as M-1. Then a few monetarists decided that time deposits (savings accounts) at commercial banks filled so many functions of money that they ought to be included, giving us M-2. Time deposits at savings banks and thrift institutions were added to produce M-3. Then things really became confused.

Because of government limitations on the interest rates that banks and thrift institutions can pay on deposits, strong market incentives developed a few years ago for financial institutions to create new assets that had many, if not all, of the characteristics of money, but that escaped legal limits on interest rates. The result was money market mutual funds, checkable deposits at some savings banks, repurchase agreements at commercial banks and a range of other assets that looked like money. Large negotiable certificates of deposit (CDs) grew in importance, and were widely viewed as being similar to money, producing the M-4 and M-5 definitions of the money supply. The original definition of M-1 was recently split into M-1A and M-1B by allowing for problems in defining total checking accounts in the economy.

The result is a tremendous uncertainty about just what constitutes money and, consequently, about just how fast the money supply has grown. M-1 (and both M-1A and M-1B) has grown rather slowly in recent years, so if you believe in that definition, monetary policy has been tight. M-2 and M-3, however, have been growing much faster, suggesting a quite expansionary policy. M-4 and M-5 grew rapidly in the mid-1970s, but have recently slowed somewhat. Since there is no agreement as to what constitutes money, there is no way of knowing how rapidly the money supply has been rising, particularly since the data for various definitions of money show very different results during recent years.

You can find a definition of the money supply that will support almost any view of past monetary policy. A few cynics have suggested that monetarists like having a number of definitions of money so they can always find one times series that will fit their models.

President Carter recently signed a bill into law that will phase out regulations on interest rates paid on deposits by banks and thrift institutions. This should change the rates at which the various Ms grow, but it is not clear how fast these changes will occur as the regulations are phased out, making it even more difficult for the public to evaluate growth rates for various definitions of the money supply during the next few years. One can only hope that the Federal Reserve Board knows what it has been doing; the rest of us are far from certain.

Since it has become almost impossible for Americans to know what monetary policy is being pursued, perhaps we should rely more on a countercyclincal fiscal policy. At least a federal government deficit can be measured. Or can it? How should off-budget items such as loans by the Export-Import Bank be treat? Or federal laws that require various state and local expenditures? Or federal commitments for future expenditures (such as indexed Social Security) that affect current private spending decisions?

Maybe we should all return to subsistence farming.