EXPECTATIONS IN ECONOMICS
In comparison with some of the other important developments in our subject the literature on expectations in economics is of recent origins and can be traced back to the work of Knight (1921), Keynes (1936), Shackle (1949, 1955), Koyck (1954), Cagan (1956), Simon (1958) and Muth (1960, 1961). Uncertainty and expectations hardly figure in the work of the classical economists, and even where expectations and their importance for decision-making are discussed in Keynes’s work, they are treated as a datum and hence do not play a central role in the development of Keynesian macroeconomic theory.1 This is particularly striking considering that most decisions facing economic agents are surrounded by uncertainty. In an uncertain environment, analysis of economic behaviour inevitably involves expectations, and how expectations are formed and the effect that they may have on current decisions become issues of central importance for the development of dynamic economic theory and time series econometrics. Under the influence of the pioneering work of Muth (1961), the analysis of expectations formation and the explicit inclusion of expectational variables in theoretical and empirical research have assumed centre stage in many areas of economics. This is particularly true of intertemporal models of consumption, labour supply, asset prices, sales, investment and inventory decisions as well as in theories of money, information processing, search activity, labour contracts and insurance. The literature on expectations in economics is vast and growing; a single essay on the subject can at best provide only a partial account of some of the recent developments.2 The present chapter is not an exception. The focus will be on a discussion of alternative models of expectations formation including adaptive, extrapolative and rational expectations models. The learning problem is also briefly dis-cussed and it is argued that, given the extreme information assumptions that underlie the rational expectations hypothesis (REH), a greater emphasis needs to be placed on the use of survey expectations in empirical economic analysis. This in turn requires larger and better surveys and the development of more appropriate techniques suited to the use of survey expectations in econometric models.