Seldom have so many been so wrong about the future of the economy. For the first time in recent memory, the consensus forecast at the start of a quarter has been wrong five out of the last seven times about whether real GNP would go up or down during the following three months.

The problem began with the recession of 1973-75. Even in economics, pride goes before the fall, and by late 1973 economists had become proud of their ability to deliver accurate forecasts of the economy. Their consensus forecast for 1974 was a "soft landing" in the first half of the year followed by a strong resumption of real growth in the second half.

When the "soft landing" turned into the worst recession in 40 years, most economists were caught by surprise. In picking up the pieces to see what went wrong, economists focused on outside shocks, known in the trade as exogenous variables. The oil embargo and price hikes that followed the Yom Kippur War disrupted the economy and sent inflation soaring. Inflation soared even more when unusually bad weather in the Soviet Union and Western Europe forced those nations into the U.S. grain market, sending food prices up sharply. The European economies also turned down at the same time as our own, thereby reinforcing our recession rather than offsetting it as in the past.

Badly burned by their remarkable failure to anticipate the worst recession in a generation or even to recognize its severity as it unfolded, economists resolved not to be caught flatfooted again. Beginning in 1978, they began to predict another imminent recession.

The recession did not come in 1978. Or in 1979. It came swiftly as 1980 began and ended just as swiftly by midyear. Virtually no one predicted one of the shortest recessions in history.

Virtually no one predicted the sputtering economy since then, either. For the first time in U.S. history, recessions have begun in back-to-back years (1980 and 1981). Quarterly changes in the economy have become random to the point of perversity. During the last two years, predicting the next movement in the economy has been as frustrating as predicting the next undulation of a drunken snake on ice.

This time the cause of the forecasting error was not outside influences as it was in 1973-75. The oil price increase in early 1979 was unexpected, but it was not the cause of such consistently bad predictions. This time the cause was right here at home.

Monetary policy changed dramatically in the fall of 1979 and economists still don't know how to assess its impact. They don't know how to calculate the effects of the massive transfer of economic power from borrowers to lenders that took place when the real rate of interest (i.e. nominal interest rates minus inflation) became positive; it was negative for most of the last generation. Economists also don't know the impact of unusually volatile interest rates on spending decisions by consumers and businessmen. When credit controls were introduced and removed in 1980, most forecasters were unsure of what that would mean to economic activity, too.

Assessing the impact of monetary policy illustrates what the current state of the economic art does well and what it does poorly. As long as the future is reasonably like the past, economists can do a good job of forecasting.

Predicting when current trends such as real GNP growth will change is what economics does poorly. The Commerce Department's index of 12 leading indicators is the most accepted way to predict recessions and expansions, but it did not distinguish itself in 1973-75. That index is light on measures of monetary policy, so it comes as no surprise that it also did not distinguish itself during the last 18 months when monetary policy was notably convulsive.

If monetary policy were the only source of forecasting error, then inserting a few more equations would restore the econometric models to their former accuracy. Even the most optimistic forecasters do not expect a return to that former accuracy because the world has become a much colder place for reasons they are still groping to understand.

The U.S. economy lost its e'lan during the 1970s. Inflation and unemployment not only rose together, an event most economists once believed impossible, but they rose to astonishing levels. Productivity increases, economic growth and international competitiveness all dropped like stones.

Worse yet, the malaise spread to the rest of the world. European unemployment, once well below our own, is now higher as economies there stagnate. Socialist countries from Poland to Yugoslavia have gone into reverse. Even Japan has seen its formidable growth rate diminish. The world's golden age of growth now seems to be a thing of the past.

Precisely what will replace the golden age of growth is far from clear. The classic business cycle has been muted by more powerful secular forces whose dimensions are still unmeasured. With the rules that govern the economy changing around them, it is not surprising that economists are no longer able to see the future clearly.

Even in the best of times the accuracy of economic forecasts is limited by the stubborn tendency of human beings to remain only partially predictable. The strength of the economy in 1979 was due largely to an unpredictable willingness of consumers to go deeply in debt to maintain their consumption. Supply-side advocates currently predict dramatic productivity gains from tax cuts, but no one really knows how millions of workers will react to them.

The diminishing accuracy of economic forecasts would be merely amusing if so many important decisions were not based upon them. The businessman who invests in a new factory needs some degree of certainty that customers will be there to buy his products. The consumer who wants a loan for a home or car also wants the certainty of knowing he will be able to keep his job and pay off his loan. The legislator who ponders tax and spending bills wants to be reasonably certain where the economy will be in three years, to say nothing of three months.

Unfortunately the demand for certainty far exceeds its supply. The consensus forecast is for the current recession to end in the spring, but the poor track record of the consensus generates little confidence in the forecast. The consensus of competent economists is more likely to be right than a naive guess, but this is an age when no one has a very good chance of being right about the future of the economy.