Archive for the ‘e488-expectations’ Category

A Response to Kantor’s “Rational Expectations and Economic Thought”

Thursday, February 7th, 2008

According to Kantor, Keynesian economics began to be undermined following 1945 when the country experienced high levels of employment.  Keynes’ model assumed the rigidity (sticky) of real wages, but Don Patinkin claimed that “the involuntary unemployment and flexible money wages precludes the existence of equilibrium.”  Essentially, he “freed” the once static logic that decreases in employment led to increases in the real wage rate.  Instead, he noted that unemployment was a result of economic dynamics and that the real balance effect was due to flexibility, rather than rigidity, of the wage rate in equilibrium.  (Not knowing what the real balance effect was, I took it upon myself to look it up and this is what I found.  The real balance effect can be defined as an increase in somebody’s real wealth and thus an increase overall consumption due to a decrease in the aggregate price level.)

There is a large contrast between Keynesian economics and monetarists that incorporate expectations into their models.  Keynesians believed that in order to secure full employment, the only thing required would be to adjust real wages and increase the overall price level.  However, this would be proven later not to work because people, in fact, do pay attention to inflation.  Thus, the above-statement would result in only an increase in nominal wages, not real wages.  In the late 1960s, along came Milton Friedman and Edmund Phelps who both said that it is impossible to “fool all the workers all the time.”  That is, laborers are aware of the trade off between inflation and real wages and consequently, the inverse relationship between unemployment and inflation.  With the realization that expectations weren’t factored into the model, economists were soon to escape the “static general equilibrium.”

Early on, expectations, as modeled by Phillip Cagan, were assumed to be weighted averages of previous rates of changes in the aggregate price level.  While this seemed like a feasible model, there was still the understanding that the Phillips curve held true and policymakers could trade inflation for employment levels.  Eventually, the theory failed in the 1970s when the country experienced stagflation–that is, high levels of both unemployment and inflation.  (At this point, people began to abandon the Phillips curve for a short-run Phillips curve that said the relationship held true, but only in the short run.)

Soon, rational expectations came to the forefront of macroeconomic thought.  The inventor…John F. Muth.  He developed these expectations to “make dynamic economic models complete.”  Muth realized that there was great profit to be made by forecasting future expectations, especially if the predictions of economic theory were much better than the predictions developed by the firms.  Because rational expectations are built in to the current models, they can be assumed to be profit-maximizing since that is the goal of every firm.  However, if the past trends prove to be imperfect in forecasting current and future trends, then the practice of rational expectations will be inaccurate.  To guarantee the validity of the model developed, Muth had to assume that random disturbances in the model were normally distributed.  This is a very important assumption because otherwise hypothesis testing to check the statistical significance of variables in the model (i.e. through t-tests) would be useless.  The model in its simplest form says the difference between the actual rate of change of real aggregate output in time period t and the trend rate of the change of supply is equal to the difference between the realized price level in time period t and the expected price level in time period tplus an error term.  The forecasted error term in time period t-1 given the expected price level in time thas to be assumed to be zero because if that forecasted error term isn’t zero, then the actual output Yt will be different than the forecast, which will make it a useless tool for policymakers and firms (Kantor 1979, 1425).

Later on, these rational expectations were tested on other markets, such as the capital market.  It is from the rational expectations breakthrough that the Efficient Market hypothesis is born.  To remind everybody of that hypothesis, it claims that in an efficient market, all information is known and the prices (i.e. stock prices) fully reflect all of that available information.  There are many who believe in the Efficient Market hypothesis, and if their assumptions are valid, then investors can only beat the market in the short run.  That is, in the long run, due to the perfect availability of information, there is no room to make a profit.  However, there are many who refute these beliefs and claim that in fact, there is no such thing as perfect information and it is not readily available to all consumers.

Later in the article, Kantor does say that the major contribution that Keynes makes to economics is the understanding that the “basis of choice lies in vague, uncertain and shifting expectations of future events,” but he neither describes nor attempts to incorporate these expectations into a model.  According to Ludwig Lachmann, the incorporation of expectations by a business man is no different than a scientist attempting to prove a hypothesis.  “Both [business man and scientist] reflect an attempt at cognition and orientation in an imperfectly known world, both embody imperfect knowledge to be tested and improved by later experience.”  This is very true.  Economists are constantly refining models to more accurately reflect observations of trends in the real world.  These lagged models were tested in several situations, including Jacob A. Frenkel’s test for the demand for money during Germany’s hyperinflation era.  He found that people incorporated inflationary expectations into their demand for money.

Lastly, these lagged models used both autoregressive (lagging one variable)  and moving average parts (incorporating lags of other variables into the model) to better account for consumers’ expectations.  Robert E. Hall is one example.  He looked at the life-cycle and permanent-income hypotheses and found that consumption that is lagged one period can help predict current consumption because it is shown that “expectations of future marginal utility is a function of today’s level of consumption alone.”  Nevertheless, the one shortcoming that needs to be realized, which is apparent in forecasts for employment levels, is the fact that random shocks are real or nominal, as well as temporary or permanent.  Thus, due to the inability for people to determine which category the shock belongs in, forecasts go astray because these effects could not be anticipated.

 Source: Kantor, Brian.  1979.  Rational expectations and economic thought.  Journal of Economic Literature 17, no. 4 (December): 1422-1441.