Debates in Macroeconomics
(Part D)
by
Charles Lamson
 |
New Classical Macroeconomics The challenge to Keynesian and related theories has come from a school sometimes referred to as the new classical macroeconomics (used because of the assumptions and conclusions of this group of economists resemble those of the classical economists---that is those who wrote before Keynes). Like monetarism and the Keynesianism, this term is vague. and no single model completely represents the school. The following discussion, however, conveys the flavor of the new classical views. The Development of New Classical Macroeconomics New classical macroeconomics has developed from two different, though related, sources. These sources are the theoretical and the empirical critiques of existing, or traditional, macroeconomics. On the theoretical level, there has been growing dissatisfaction with the way traditional models treat expectations. Keynes himself recognized that expectations (in the form of "animal spirits") play a big part in economic behavior. The problem is, traditional models have assumed that expectations are formed in naive ways. A common assumption, for example, is that people form their expectations of future inflation by assuming present inflation will continue. If they turn out to be wrong, they adjust their expectations by some fraction of the difference between their original forecast and the actual inflation rate. Suppose I expect 10 percent inflation next year. When next year comes, the inflation rate turns out to be only 5 percent, so I have made an error of 5 percent. I might then predict an inflation rate for the following year of 7.5 percent, halfway between my earlier expectation (10 percent) and actual inflation last year (5 percent). The problem with this treatment of expectations is that it is not consistent with the assumptions of microeconomics. It implies people systematically overlook information that would allow them to make better forecasts, even though there are costs to being wrong. If, as microeconomic theory assumes, people are out to maximize their satisfaction and firms are out to maximize their profits, they should form their expectations in a smarter way. Instead of naively assuming the future will be like the past, they should actively seek to forecast the future. Any other behavior is not in keeping with the microeconomic view of the forward-looking, rational people who compose households and firms.  On the empirical level, there was stagflation in the U.S. economy during the 1970s. Remember, stagflation is simultaneous high unemployment and rising prices. The Phillips Curve theory of the 1960s predicted that demand pressure pushes up prices, so that when demand is weak---in times of high unemployment, for example---prices should be stable (or perhaps even falling). The new classical theories were an attempt to explain the apparent breakdown in the 1970s of the simple inflation unemployment trade-off predicted by the Phillips Curve. Just as the Great Depression of the 1930s motivated the development of Keynesian economics, so the stagflation of the 1970s helped motivate the formulation of new classical economics. Rational Expectations In previous posts, we stressed households' and firms' expectations about the future. A firm's decision to build a new plant depends on its expectations of future sales. The amount of saving a household undertakes today depends on its expectations about future interest rates, wages, and prices. How are expectations formed? Do people assume things will continue as they are at present (like predicting rain tomorrow because it is raining today)? What information do people use to make their guesses about the future? Questions like these have become central to current macroeconomic thinking and research. One theory, the rational-expectations hypothesis, offers a powerful way of thinking about expectations. Suppose we want to forecast inflation. What does it mean to say that my expectations of inflation are "rational"? The rational expectations hypothesis assumes people know the "true model" that generates inflation---they know how inflation is determined in the economy and they use this model to forecast future inflation rates. If there were no random, unpredictable events in the economy, and if people knew the true model generating inflation, their forecasts of future inflation rates would be perfect. Because it is true, the model would not permit mistakes, and thus the people using it would not make mistakes.  However, many events that affect the inflation rate are not predictable---they are random. By "true" model, we mean a model that is on average correct in forecasting inflation. Sometimes the random events have a positive effect on inflation, which means the model underestimates the inflation rate, and sometimes they have a negative effect, which means the model overestimates the inflation rate. On average, the model is correct. Therefore, rational expectations are correct on average, even though their predictions are not exactly right all the time. To see this, suppose you have to forecast how many times a fair coin will come up heads out of 100 tosses. The true model in this case is that the coin has a 50-50 chance of coming up heads on any one toss. Because the outcome of the 100 tosses is random, you cannot be sure of guessing correctly. If you know the true model---that the coin is fair---your rational expectations of the outcome of 100 tosses is 50 heads. You are not likely to be exactly right---the actual number of heads is likely to be slightly higher or slightly lower than 50 but on average you will be correct. Sometimes people are said to have rational expectations if they use "all available information" informing their expectations. This definition is vague, because it is not always clear what "all available information" means. the definition is precise, if by "all available information" we mean that people know and use the true model. We cannot have more or better information than the true model. If information can be obtained at no cost, then people are not behaving rationally if they fail to use all available information. Because there are almost always costs to making a wrong forecast, it is not rational to overlook information that could help improve the accuracy of a forecast as long as the costs of acquiring that information do not outweigh the benefits of improving its accuracy.  Rational Expectations and Market Clearing If firms have rational expectations and if they set prices and wages on this basis, then, on average, prices and wages will be set at levels that ensure equilibrium in the goods and labor markets. When a firm has rational expectations, it knows the demand curve for its output and the supply curve of labor that it faces, except when random shocks disrupt those curves. Therefore, on average the firm will set the market clearing prices and wages. The firm knows the true model, and it will not set wages different from those it expects will attract the number of workers it wants. If all firms behave this way, then wages will be set in such a way that the total amount of labor supplied will, on average, be equal to the total amount of labor that firms demand. In other words, on average there will be no unemployment. There might be disequilibrium in the labor market (either in the form of unemployment or in excess demand for workers) because firms may make mistakes in their wage-setting behavior due to expectation errors. If, on average, firms do not make errors, then, on average, there is equilibrium. When expectations are rational, disequilibrium exists only temporarily as a result of random, unpredictable shocks---obviously an important conclusion. If true, it means disequilibrium in any market is only temporary, because firms, on average, set market-clearing wages and prices. The assumption that expectations are rational radically changes the way we can view the economy. We go from a world in which unemployment can exist for substantial periods and the multiplier can operate to a world in which (on average) all markets clear and there is full employment. In this world there is no need for government stabilization policies. Unemployment is not a problem that governments need to worry about; if it exists at all, it is because of unpredictable shocks that, on average, amount to zero. There is no more reason for the government to try to change the outcome in the labor market than there is for it to change the outcome in the banana market. On average, prices and wages are set at market-clearing levels.  *CASE & FAIR, 2004, PRINCIPLES OF ECONOMIC, 7TH ED., PP. 654-656* end |
No comments:
Post a Comment