Wednesday, 23 February 2011

Role of Expectations in Economics

Expectations play a very important role in economics. One of the major limitations of our understanding about real world economic phenomena arise from our inability to deal with expectations in an efficient manner and incorporate them in our empirical analysis. Our overdependence on the rational expectations assumption has only worsened this limitation.

Most economic decisions of all economic agents, be it consumers, producers or traders, are dependent upon their expectations of the future in one way or another. A consumer who expects the prices to rise tomorrow will be willing to buy more of goods today than another who expects the prices to fall tomorrow. A producer who expects the prices of a particular good to increase over the next year would often be more willing to invest in to a production unit for that good compared to another producer who expects the prices to fall. A trader who expects the price of a good to increase sharply after a month will buy and store the goods for a month, even if that comes at a cost.

Expectations and asset markets

In a stock market, every bit of news can lead to a heuristic reassessment of future leading to frequent changes in expectations about the future. Such news may pertain to any new development related to any factor that may affect asset prices. It can be a new report analysing the broad trends of profitability of corporate sector, or may just be the report of a major investor dumping some of his investments in the market. As would be predicted by the theory of bounded rationality, signals like price play a significant role. A sharp fall in the price of a particular share may signal to the investors that all may not be well with that company. The resultant expectation among a segment of investors that its prices may fall in future may lead to panic selling by many, even though the initial fall in price may not have anything to do with the economic viability and financial health of that stock. Such selling will actually bring down the prices of that stock. This is a modern version of ‘Bank run’, wherein rumours of a bank in trouble can lead to depositors rushing to withdraw their deposits – a phenomenon that may actually lead to the collapse of that bank. In all such cases, expectations act as a self fulfilling prophecy.

Expectations and Recession

When an economy is in severe recession, investors are highly pessimistic. As growth in demand is not expected, not many are willing to invest. Since the expected income prospects are not very bright, the consumers tend to save a significant part of their earnings for the rainy day. Reigning in of consumption and investment means a constrained demand. Simultaneously, the workers are unwilling to switch jobs, while the entrepreneurs are equally unwilling to take greater risks of a new enterprise. When perceived risk is higher, a greater risk premium is charged by the Banks, making the cost of capital costlier and eventually contributing to a supply constraint. In this way, continued pessimism can hold back both demand and supply and keep the economy in continued recession. In the light of the role of expectations, one can view the whole of Keynesian economics as a means to break the vicious cycle of pessimistic expectations in case of recession.

Importance of Expectations

The role of expectations in asset markets and economic growth exemplify their powerful influence over the eventual economic outcome. Expectations are often at the core of investment decisions, and to a great extent, contribute to the business fluctuations as well as asset bubbles and busts. Expected inflation, expected exchange rate and expected growth of income and demand are invariably the deciding factors in all such decision making. Understanding how these expectations are formed, to what extent they are rational and how much do we know and understand them has become the single biggest challenge in all kind of macroeconomic analysis, forecasting and regulation. It would not be wrong to say that expectations may hold the key to efficient economics in future.

Theories and Debates about Expectations

One of the major outcomes of the recent Global Economic Crisis has been the re-ignition of the debate on how expectations need to be incorporated in the Economic models to analyse real life events and forecast the future. This debate has largely been focussed so far on whether the assumptions of rational expectations can be safely continued for economic analysis in the way it has been done so far, and whether there is any alternative which can be used in place of rational expectations assumption in quantitative exercises.

Rational Expectations Hypothesis

The theory of rational expectations is the very basis of “efficient markets”. Based largely on the works of John F. Muth and Robert Lucas Jr., who was awarded the Nobel Prize for his work in 1995, this theory treats every economic agent as a rational being, who always behaves in the rational manner. The errors, if any, are treated as random, and nullify each other as a result of which the mass expectation or their average expectation always coincides with the stable market equilibrium. Notably, this theory is closely related to the theory of ‘rational choice’ propounded by Gary Becker, who also won the Nobel in 1992 for his works in that field.

Why is Rational Expectations Hypothesis so Important?

The simple assumption of rationality of expectations has very significant implications on both the economic theory as well as all empirical research. At a theoretical level, it allows the macroeconomic phenomena to be built upon microeconomic behaviour of economic agents. At the empirical level, this theory allows all interrelated variables to be placed in simple mathematical equations removing all uncertainties resulting from human errors or erroneous judgement, thus making use of data for conclusions and validations infinitely easier than it would have been with the uncertainties of human errors thrown in the equation.

Alternate Explanations

There are three major alternatives to rational expectations theory.

The first of them can be termed as a “random Walk” theory, which in broad and simple way can be summarised as the lack of any necessity for expectations to be always rational. In this model, the expectations may depend upon any of the factors, genetic or environmental; may be based on information or the lack of it; could be a result of extensive analysis or pure instinct; and can even be a result of an earlier experience, culture or psychological state. Its major limitation is that human mind and decisions do not look like being devoid of all methods even in their madness. Accepting it would mean that all economic outcomes significantly dependent or affected by prevailing expectations will always be random in nature.

The second is the theory of “Adaptive expectations” which predicts the expectations to be a function of past experiences. If this were absolutely true, then all future behaviour will continue to follow the patterns of the past with little scope of any deviations there from. Clearly, this is not true. While history does repeat itself, the future is seldom a simple extension of past without any deviations. In fact, deviations from past are as much a rule in time series data as are any other accepted phenomena in nature. This theory, which had its days of glory once, subsided rapidly once the theory of rational expectations came in to vogue.

The third alternative is the theory of “Bounded Rationality”, which states that economic agents, i.e.. human beings are basically rational, but are unable to invariably arrive at the most rational expectation, due to the high costs of collecting information and extensive analysis of the same that is required to arrive at the most rational expectation of a future occurrence. Instead, people often resort to easily available signals and simple thumb rules or heuristic decision making for arriving at their own expectation of a future event. These concepts are largely a result of incorporating behavioural and psychological aspects in economic decision making. Pioneering work in this regard was contributed by Daniel Kahnmen, Amos Tversky and Herbert Simon. This theory is, in many ways a better alternative than the ‘rational expectations theory’. Unfortunately, so far it has been of little use for researchers playing with figures and equations, and hence unable to get the recognition that it deserves.

According to the theory of ‘bounded rationality’, the expectations and as a consequence, the actions of rational agents are determined by the signals that are available to them. In the market, the most common signal is price. Price is a signal to the producers about the prevailing balance between demand and supply, and most decisions regarding future production capacity and its utilisation are based on it. Interestingly, price is also the signal to a wealthy consumer willing to pay for a better product about the quality of a good, something that luxury brands often exploit for promoting exclusive brands that come to be recognised as status symbols.

Why are we unable to look beyond ‘Rational Expectations’

Rational Expectations assumption lies at the core of all theoretical models used in macroeconomic analyses and forecasting. Almost invariably, at the macroeconomic level, the effect of any dependent variable (e.g.. inflation) leads to an impact on another variable (e.g.. Economic growth) through human decision making involving a large number of participants (i.e.. the mass) and unless we presume that they are all rational agents and likely to behave in the same way, no meaningful conclusions can be drawn from the analyses of data related to them. The assumptions that everyone of those agents, or at a more practical level, the average behaviour of all such agents is always rational allows the analyst to reduce their behaviour to mathematical equations which can be easily quantified and studied allowing scope for concluding simple relationships between those factors.

Unless we assume that expectations of the masses are always rational, there is very little scope of econometric methods like multivariate regression. Thus, for the usual researcher, planning to validate his hypothesis by subjecting primary data to econometric analysis, falling back on the rational expectations assumption is a necessary prerequisite, avoiding which would actually mean giving up the only tool of research and study that he often has at his disposal. To a large extent, this situation is due to the almost pathological obsession of current day economic researcher with mathematical processes, equations and figures. Equally absurd is the tendency to rely on statistical results as the only validation for any economic proposition, even though the statistical results themselves are neither perfect fit for the given hypothesis nor are ever replicable – the two prerequisites for using them as scientific evidence. Most fallacious, however, is the use of over- simplistic equations with a multitude of assumptions, which are then conveniently ignored when the results are analysed. As a result, very often the econometric research is reduced to a blind search for that set of data and equations which can possibly be used for validating some hypothesis. Till the time we are able to get out of this obsessive trap, it would be difficult to approach and understand the limitations of rational expectations hypothesis and find a way of removing them.


Expectations are far more significant in economics than seems to be accepted till date. They can be the difference between efficient and inefficient markets and have the potential to lead to market failures of all kinds. That makes regulating expectations as a major means of stabilising and regulating economy. Unfortunately, most of empirical has bypassed this important aspect, probably due to its overdependence on assumption of rational expectations and efficient markets as a tool of economic research.

No comments:

Post a Comment