Top 4 Theories of Expectations Formation

Read this article to learn about the four theories of expectations formation in economic theory.

Theory 1 # Cobweb Model:

As a model of expectation, the ‘Cobweb Model’ of a market is familiar to practically all students of economics.

While this model is known as an example of dynamics and market stability; it is the first formulation of expectations in an economic model.

It, thus, makes a useful starting point. The essence on the Cobweb Model is some delay between the formation of production plans and their realization.

It is, therefore, often applied to agricultural markets, where farmers decide in one season how much land to plant under a crop which will be harvested in the following season. It is only at the time of the harvest that the actual quantity of the crop available (the actual supply) is known and the price which prevails is the one which clears the market. But how does the farmer decide how much land to plant under a crop? In order to make this decision he has to form some expectations of the price that will prevail when the crop is ready, harvested and marketed.

The farmer will base his expectations of future prices on the price ruling at the time of planting the crops. Since all the farmers think and behave in the same manner, the model predicts that year of ‘glut’ (plenty) will be followed by a year of ‘scarcity’; again followed by another year of plenty and so on in the familiar Cobweb fashion (a feature that gives) the model its name.

Under certain conditions, the oscillations will converge towards equilibrium under other conditions, these may diverge. But, the Cobweb Model does not attract much empirical support since it assumes that farmers conduct their business in a most naive manner, because their behaviour ignores the impact of similar actions of all the other farmers. Again, even if they are not smart enough to consider the impact of their joint actions, one might well expect the farmers to learn from their experience and to benefit from that.

The Cobweb Model follows a regular pattern, namely-over supply-under supply over supply- under supply. The assumptions of the Cobweb model do not allow any learning on the part of the farmers and therefore as a Model of expectations, the Cobweb Model is unsatisfactory. Despite these limitations the model does illustrate the importance of expectations and highlights the advantages of alternate model of expectations.

Theory 2 # Extrapolative Expectations:

In an attempt to overcome the limitations (naivety) of Cobweb Models, Metzler (1941) introduced the idea of extrapolative expectations. He reasoned that future expectations should be based not only on the past level of an economic variable, but also on its direction of change. The extrapolative expectations in any period is equal to the price level in the previous period plus (or minus) some proportion of the change between the previous two periods.

In this way extrapolative mechanism allows for more variety in the behaviour of the model, which, in turn, depends upon the underlying economic structure of the model. Extrapolative expectations, thus, make it clear that the structure of the model plays a very important part in the appropriate expectations mechanism.

A more sophisticated version of this model was proposed by Hicks (1946). Hicks’ original Model asserted that the expected rate of inflation equals the current inflation rate plus an adjustment factor which allows for the rate of change of inflation.

In other words, people are forming their expectations not simply about the rate of inflation, but also the rate of change of that. This is basically a second order expectations mechanism, and as such seems rather over-sophisticated for times of normal inflation. Since Hicks wrote—the term extrapolative hypothesis has been used to describe any method of expectation formation which is based upon a distributed lag of actual prices.

Theory 3 # Adaptive Expectations:

Yet another approach to expectations formation, which can also be viewed as a special case of the extrapolative hypothesis has come to dominate much of the work done on expectations. This is the adaptive expectations hypothesis, first put forward by Cagan (1956) and Neriove (1958). It states that expectations are revised in accordance with the last forecasting error; hence its alternative name, the error learning hypothesis.

Another variation of the extrapolative theme, which has received some prominence recently, is the regressive (effect-cause relationship)—extrapolative expectations hypothesis. This was first put forward by Duesenberry (1958) and expanded by Modigliani and Sutch (1966).

They suggest that there might be both extrapolative and regressive elements present in the process by which expectations are formed. The latter implies a reversion of expectations towards a long run ‘normal’ level, which may in itself be given parameter of the system, or a lagged function of actual price changes, where the lag may extend over several years. In the latter case, the hypothesis once more becomes a special case of the general extrapolative hypothesis.

This mechanism of adaptive expectations formation is more frequently used in economics. According to this mechanism of adaptive expectations agents revise their expectations in each period according to the degree of error in their previous expectations—hence the name adaptive expectations. The speed at which the expectations adjust to past error is called the coefficient of adaptations. This coefficient may fluctuate between zero and one. Thus, with adaptive expectations, the expected value in the next period is equal to the expectations for the current period plus or minus a proportion of the error in the expectations for the current period.

Until the idea of rational expectations was introduced in economics, adaptive expectations were the most common method of formulating expectations in economics. Its popularity was due to its conceptual simplicity and the ease with which, it could be implemented empirically. Statistical estimates for the coefficient of adaptive expectations can be easily obtained.

Moreover, according to Carter and Maddock the adaptive behaviour in the face of an uncertain environment appears intuitively very plausible and appealing. Adaptive expectations model worked well in a climate in which the change was gradual—a characteristics of the 1950s and the 1960s when the inflation rates were low and relatively stable and when inflation rates underwent fast changes and increased rapidly, adaptive forecasts were left behind.

Thus, adaptive expectations are effective when the variable being forecast is reasonably stable, but adaptive expectations’ are of little use in forecasting trends. Moreover, there may be additional or supplementary information available to the forecaster which is a highly relevant to the variable being forecast for example knowledge of which party has won a general election may be used to forecast the rate of inflation which is otherwise based only on past price data.

Mechanical application of an adaptive expectations mechanism, therefore, does not essentially make the best use of all the scarce information available. It is for this reason that this mechanism as a for caster of economic behaviour is not very dependable. Under some peculiar circumstances, it has been observed that the adaptive expectations mechanism performs poorly. Rather than converging to zero, the expectations errors increased from year to year.

Theory 4 # Rational Expectations:

The naive Cobweb model of expectations and extrapolative and adaptive mechanisms of expectations suffer from a common defect that they are essentially arbitrary rather than based on any underlying theory of economic behaviour. This led to the publication in 1961 of a classic paper by John Muth in which he advanced the theory of rational expectations.

In his own words, “I would like to suggest that expectations, since they are formed predictions of future events, are essentially the same as the predictions of the relevant economic theory….. in particular the hypothesis asserts that the economy does not waste information and that expectations depend specifically on the structure of the entire system.”

The hypothesis asserts three things:

(i) Information is scarce, and the economic system generally does not waste it;

(ii) The way expectations are formed depends specifically on the structure of the relevant system describing the economy;

(iii) A ‘public prediction’—will have no substantial effect on the operation of the economic system (unless it is based on inside information).

Thus, the essence of the idea of rational expectations is that over time, economic agents accumulate a wealth of information concerning the relationship governing economic variables and the behaviour of other agents, particularly of the government. This information can be used in forming expectations about future values of economic variables. Muth reasoned that information should be considered as just another of the resources available to be allocated to maximum advantage.

Utility maximizing individuals should use all the information available to them in forming their expectations. Part of the information which is relevant to the behaviour of any economic system is the structure which underlies that system. The fundamental premise on which most economic analysis is based is that economic agents “do the best they can with what they have? Muth, therefore, concluded that rational economic agents would use their knowledge of the structure of the economic system in forming their expectations.

The rational expectations hypothesis, in itself, should not be provocative to economists. It merely brings expectations within the scope of individual maximizing behaviour. Expectations were handled within economic models on very casual and ad hoc basis. Rational expectations provides a way of incorporating expectations which is consistent with the orthodox economy theorizing.

The development of rational expectations theory will make a more significant contribution to economics in the impetus it gives to research on the vital areas of learning and expectations formation. It brings to the fore question about the availability and use of information. Instead of being the finale of the monetarist’s case against policy intervention, it should be seen as the prologue for a revitalized theory of expectations, information arid policy.