Estimating budgets for fine-tuning

Congressmen were asking: ''How large would the deficit be if the economy was growing at about the same average rate it had in the past several decades, instead of just emerging from a deep recession?''

Two Commerce Department economists, Frank de Leeuw and Thomas M. Holloway, attempted to find out. They put together an econometric model, wound up their computers, and produced some numbers that were included in a paper presented at a Federal Reserve Bank of Boston conference last month.

The way they measure their hypothetical deficit, it amounts to 4 percent of gross national product (GNP) this calendar year, compared with about 5.6 percent for the deficit as recorded by the federal government.

This may be mysterious to the uninitiated. But it is only the latest in a series of what economists term a ''cyclically adjusted budget.''

Looking at the actual budget facts, the Reagan administration reported last week that the deficit for fiscal 1983, which ended Sept. 30, was $195.4 billion, or $14.4 billion below its July estimate. Revenue was up slightly and spending down substantially from that earlier estimate.

Also, the Congressional Budget Office (CBO) projected deficits in each of the next fiscal years of about $190 billion. But such projections require assumptions. One is that Congress will enact only small tax increases and spending reductions. Another is the growth in the economy - a real 3.1 percent this year, 5 percent next, 4 percent in 1985, and 3.5 percent in 1986. It already appears that growth will be somewhat higher this year. But an extra 1 percent in GNP growth would reduce the deficit only about $10 billion the first year. ''It is next to impossible to grow our way out of these deficits,'' a CBO aide said. Congress will eventually have to do something about them.

As for these ''cyclically adjusted'' hypothetical budgets, they are designed to remove the impact of the business cycle from the deficit numbers. They in effect ask, ''What would be the deficit if a recession had not reduced tax revenues and raised spending?'' The names for this type of budget deficit include ''full-employment deficit,'' ''high-employment deficit,'' and, more recently, the ''structural deficit.''

All these budgets aim to help administration and congressional policymakers decide how much stimulus they could apply to the economy in the form of tax cuts or extra spending without stimulating renewed inflation. They may also want to know whether the deficit will disappear once the economic recovery has moved ahead for a few years, or whether the deficit is ''structural'' - will it last indefinitely?

At the Federal Reserve System, these cyclically adjusted budgets might give the Federal Open Market Committee (FOMC), the monetary policymaking body, some idea of whether it could afford to create somewhat more money without prompting another bout of bad inflation. But these hypothetical budgets are somewhat out of vogue nowadays. One central banker said the FOMC hasn't looked for years at ''potential GNP'' - an element of most of these cyclical budgets.

Once you start playing such mathematical games with the budget, much depends on your assumptions. The high-employment budget, for instance, can measure what the deficit would be if employment was at a high level. The key is which high level you assume. The Bureau of Economic Analysis at the Department of Commerce has one such budget assuming a 4.9 percent unemployment rate. It shows a deficit equivalent to 1.5 percent of the high-employment GNP, the nation's output of goods and services, in calendar 1983 if unemployment was that low. In fact, most economists believe such a low employment rate would today mean considerably higher inflation.

If the high-employment budget is based on a 6 percent unemployment rate, the 1983 deficit would be 2.5 percent of the corresponding high-employment GNP, the bureau reckons.

So these calculated budgets show the deficits are still there - even if the nation's economy has been working at its ''full potential,'' whatever that may be. But the deficits aren't quite so bad in terms of proportion of high-employment GNP as the actual deficit is of actual GNP.

Now Mr. de Leeuw and Mr. Holloway have come up with a budget that says what the deficit would be, not if the economy was running at full steam, but at a sort of ''normal'' rate. It does this by assuming a ''mid-expansion trend GNP.'' The two chart the level of the economy at the middle of each of the past nine cyclical expansions; figure out from this where GNP would be this year if it was itself in the middle of such an economic expansion (actually it is early in the expansion); and then calculate the deficit. It turns out to be $138.7 billion, or 4 percent of ''trend GNP,'' which is far higher, by this measure, than in earlier years. This budget doesn't need to estimate the full potential of the economy.

Commenting on the Commerce Department economists' paper, Barry P. Bosworth, an economist with the Brookings Institution, argued it was more important for policymakers to know how far they can go in stimulating the economy without stirring up inflation than what will happen to the deficit if the economy plugs along at its usual rate. Figuring that it will take years for the economy to reach its full potential, he is interested in the ''speed limit'' on traveling in that direction without prompting more inflation. So he favors the ''high-employment deficit'' concept.

Mr. Bosworth is an economic activist who believes it possible to improve economic performance by careful control of fiscal and monetary policy.

Robert M. Solow, an economics professor at the Massachusetts Institute of Technology, likened the policymaker considering these different cyclical budgets to a golfer, saying it was more important for him to know how far short his scores were from those of Arnold Palmer than whether he was short of his own trend of scores.