By the early 1980s, the perceived failure of monetarist policies had given rise to a new vogue—what can be called neomonetarism. The two basic tenets of neomonetarism—the natural rate hypothesis and the rational expectations hypothesis—are the most radical innovations in macroeconomic theory of the last 15 years. They are also, in combination, the most misleading and potentially the most dangerous.

The natural rate hypothesis implies that demand management is useless, since unemployment cannot be lowered permanently below some relatively high and stable “natural” rate by means of expansionary monetary policy. The rational expectations hypothesis, when added to traditional monetarism, suggests that people’s expectations about inflation can be systematically in error only as long as people are wrong about the growth rate of the money supply.

The problem is that politicians have interpreted the language of rational expectations to mean that inflationary expectations, hence inflation itself, can be rapidly reduced without a prolonged period of unemployment. Supply-siders, too, have called on rational expectations to argue that tax cuts will increase savings, investment and productivity virtually instantaneously.

Experience teaches otherwise. By 1982, inflation rates had begun to decline. But by 1982, the U.S. had just undergone the longest period of slack and the highest levels of unemployment it had experienced since the 1930s. Inflationary expectations take much longer to break than rational expectations theory suggests. And structural and institutional factors independent of expectations play a much larger role than neomonetarists, or indeed monetarists, would allow.