A New Rational Expectations Hypothesis: What Can Economists Really Know About the Future?

In implementing REH, economists have relied on a class of models that rest on a key premise: an economist can ignore the importance of quantitative changes in the economy’s structure that he himself could not specify ex ante with a probabilistic rule. In the early 1970s, Robert Lucas pointed out that once such models are upheld as “the relevant economic theory,” it follows on purely logical grounds that REH models represent rational forecasting and its implications for market outcomes.

There is ample evidence that quantitative structural change is an important driver of outcomes in real-world markets, most notably those for assets. The evidence also shows that this change often is triggered by historical events that are not exact repetitions of similar events in the past, for example, the appointment of a new Federal Reserve chair or company CEO. The timing of these events may be unknown, and their quantitative impact on the economy’s structure depends on the extent of their novelty and the particular historical context in which they occur. As such, no one can fully anticipate — even in probabilistic terms — the structural change triggered by such events.