No single piece of macroeconomic advice given by the experts to their government has ever had the results predicted.
Debates in Macroeconomics
(Part F)
by
Charles Lamson
Evaluating Rational-Expectations Theory
From the last few posts, it should be clear that the key assumption concerning the new classical macroeconomics is how realistic is the assumption of rational expectations. If it approximates the way expectations are actually formed, then it calls into question any theory that relies at least in part on expectation errors for the existence of disequilibrium. The arguments in favor of the rational-expectations assumption sound persuasive from the perspective of microeconomic theory. If expectations are not rational, there are likely to be unexploited profit opportunities---most economists believe such opportunities are rare and short-lived. The argument against rational expectations is that it requires households and firms to know too much. This argument says it is unrealistic to think these basic decision-making units know as much as they need to know to form rational expectations. People must know the true model (or at least a good approximation of the true model) to form rational expectations, and this is a lot to expect. Even if firms and households are capable of learning the true model, it may be costly to take the time and gather the relevant information to learn it. The gain from learning the true model (or a good approximation of it) may not be worth the cost. In this sense, there may not be unexploited profit opportunities around. Gathering information and learning economic models may be too costly to bother with, given the expected gain from improving forecasts. Although the assumption that expectations are rational seems consistent with the satisfaction-maximizing and profit-maximizing postulates of microeconomics, the rational-expectations assumption is more extreme and demanding because it requires more information on the part of households and firms. Consider a firm engaged in maximizing profits. In some way or other, it forms expectations of the relevant future variables, and given these expectations, it figures out the best thing to do from the point of view of maximizing profits. Given a set of experiences, the problem of maximizing profits may not be too hard. What may be hard is forming accurate expectations in the first place. This requires firms to know much more about the overall economy than they are likely to, so the assumption that their expectations are rational is not necessarily realistic. Firms, like the rest of us---so the argument goes---grope around in a world that is difficult to understand, trying to do their best but not always understanding enough to avoid mistakes.  In the final analysis, the issue is empirical. Does the assumption of rational expectations stand up well against empirical tests? This is difficult to answer. Much work is currently being done to answer it. There are no conclusive results yet, but it is one of the questions that makes macroeconomics an exciting area of research. Real Business Cycle Theory Recent work in new classical macroeconomics has been concerned with whether the existence of business cycles can be explained under the assumptions of complete price and wage flexibility (market clearing) and rational expectations. This work is called real business cycle theory. As we discussed in earlier posts, if prices and wages are completely flexible---that is not fixed and open for negotiation, then the aggregate supply (AS) curve is vertical, even in the short run. If the AS curve is vertical, then events or phenomena that shift the aggregate demand (AD) curve (such as changes in the money supply, changes in government spending, and shocks to consumer and investor behavior) have no effect on real output. Real output does fluctuate over time, so the puzzle is how these fluctuations can be explained if they are not due to policy changes or other shocks that shift the AD curve. Solving this puzzle is one of the main missions of real business cycle theory. It is clear that if shifts of the AD curve cannot account for real output fluctuations (because the AS curve is vertical), then shifts of the AS curve must be responsible. However, the task is to come up with convincing explanations as to what causes the shifts and why they persist over a number of periods. The problem is particularly difficult when it comes to the labor market. If prices and wages are completely flexible, then there is never any unemployment aside from frictional unemployment (the result of voluntary employment transitions within an economy). For example, because the measured U.S. unemployment rate was 9.7 percent in 1982 and 4.2 percent in 1999, the puzzle is to explain why so many more people chose not to work in 1982 than in 1999.  Early real business cycle theories emphasized shocks to the production technology. Suppose there is a negative shock in a given year that causes the marginal product of labor to decline. This leads to a fall in the real wage, which leads to a decrease in labor supply. People have been led to work less because the negative technology shock has led to a lower return from working.The opposite happens when there is a positive shock: The marginal product of labor rises, the real wage rises, and people choose to work more. This early work was not as successful as some had hoped because it required what seemed to be unrealistically large shocks to explain the observed movements in labor supply over time. Since this initial work, different types of shocks have been introduced, and work is actively continuing in this area. To date, fluctuations of some variables, but not all, have been explained fairly well. Some argue that this work is doomed to failure because it is based on the unrealistic assumption of complete price and wage flexibility, while others hold more hope. Real business cycle theory is another example of the current state of flux in macroeconomics.  *CASE & FAIR, 2004, PRINCIPLES OF ECONOMICS, 7TH ED., PP. 657-658* end |
No comments:
Post a Comment