Uncertainty

Surprisingly, uncertainty has a rather short history in economics. In this paper I will trace attempts to differentiate it from risk and show how particular interpretations are grounded in the assumptions of the paradigm from which they derive.

In the time between 1921 and 1936 the economic climate changed dramatically for capitalism. The success of the Russian experiment in reducing the uncertainty of unemployment and hunger was in stark contrast to the boom and then depression that enveloped the capitalist economies. Against this background of increase and then decrease in confidence, theories of uncertainty and risk arose.

There has been much controversy about Keynes’s views on uncertainty. One reason for this is that Keynes presented several different arguments regarding uncertainty, including a possible shift in his view over time toward more emphasis on its absolutely unquantifiable nature. Keynes initially argued that probability is a logical relation and so it is objective. He said a statement involving probability relations has a truth-value independent of people's opinions. With regard to probabilities themselves, Keynes distinguishes four cases, that there are no probabilities at all (fundamental uncertainty), that there may be some partial ordering of probable events but no cardinal numbers can be placed on them, that there may be numbers but they cannot be discovered for some reason, and that there may be numbers but they are difficult to discover. (Keynes, 1921 p. 33) In contrast and at about this time, Knight argued uncertainty referred to situations involving randomness cannot be expressed in terms of probability and risk is a more accurate term for specific probabilities. (Knight, 1921 p20)
In the General Theory, Keynes returned to the role of uncertainty in forming long run expectations in the context of arguing the case for rational long term investment in contrast to destabilizing speculation. He saw a large proportion of our economic activities depend on spontaneous optimism and animal spirits. Faced with matters about which there is no basis on which to form any calculable probability, agents rely on confidence. The management of this confidence then becomes a crucial issue. (Keynes, 1936 p162)

Given that people base their expectations on the weightier of probable events they foresee as possible, and that those perceptions depend greatly on what other people expect, expectations can change very suddenly. The idea of other people’s expectations forming the model for agent’s perceptions of uncertainty needs teasing out. Russia, Germany and the United States provided examples of very different political and economic arrangements as possible futures. The absolute conviction in ideological correctness and the demands it placed on citizens would have provided some sense of security after the tumult since 1900. But at what cost? The economic and human devastation of this period demonstrates the lengths humans will go to avoid uncertainty.

The war demonstrated capitalist economies needed the increased involvement of the state to maintain confidence. Post war voters had raised expectations of governments to provide employment and a safety net under poverty and old age. The welfare state was to be funded by the controlled animal spirits of entrepreneurs. In contrast, rational choice theory emphasizing individual utility maximizing was also being developed. In 1944 Von Neumann and Morgenstern lay a rational foundation for decision making under uncertainty or risk according to expected utility rules using axioms in game theory. Agents are assumed to be rational with fixed preferences unable to be altered by exogenous events. This implies risk and uncertainty are the same thing. The agent chooses to assign a probability to all potential outcomes within a bounded reality. It is therefore a problem of knowledge of the relevant probabilities not of their existence. This line of research continues today with increasingly complex econometric solutions to finding the relevant probabilities and assigning them accurately.

Post Keynesians have argued that Knight’s distinction remains crucial. Uncertainty, particularly when it is tied up with the issues of time and information may be the only relevant form of randomness in economics. They suggest situations of Knightian risk are only possible in gambling where the probabilities are contrived and controlled. This has no relevance where the situation is unique and the alternatives are not really known or understood. Thus decision rules in the face of uncertainty ought to be considered different from conventional expected utility. Post Keynesians have also drawn on different viewpoints in forming their explanation of the nature of uncertainty and expectation formation. One line of argument emphasizes the element of potential surprise associated with crucial decisions such as major capital investments. In this view fundamental uncertainty is equated with free will and choice by decision makers. Followers of this line also emphasize the unidirectional and irreversible nature of historical time. Another line argues that the essence of fundamental uncertainty derives from nonergodicity of time series and stochastic processes. Yet another emphasizes the nature of the interaction of people in forming expectations and various complexities of group dynamics. In the face of this uncertainty expectations are seen as arising from bounded rationality and conventions. (Rosser, 2000)

New challenges to the rationality assumption came in the work of Kahneman and Tversky since the 1970’s. In the development of prospect theory they demonstrated agents, faced with uncertainty, will change their preferences depending on how the choice is framed. The startling result is risk avoiders will switch to risk seeking behavior irrespective of the rational calculation of probabilities when known. They found people are myopic in their decisions, may lack skill in predicting their future tastes, and can be led to erroneous choices by fallible memory and incorrect evaluation of past experience. (Kahneman and Tversky, 2000 p758)
In my view this implies agents are unable to distinguish between risks of the measurably probabilistic kind and uncertainty covering the rest of possible outcomes. It may be useful to continue research on risk on the basis it will cover uncertainty if the concepts are not distinguishable for modern agents.

References

http://cepa.newschool.edu/het/profiles/knight.htm
accessed 22 March 2004
http://cepa.newschool.edu/het/essays/uncert/intrisk.htm

accessed 22 March 2004
Duffie, D (2001) Dynamic Asset Pricing Theory Third Edition Princeton
University Press
Kahneman, D., & Tversky, A (eds) (2000) Choices, Values and Frames
Russell Sage Foundation Cambridge University Press
J.M. Keynes, (1936)The General Theory of Employment, Interest and Money,
Macmillan,
J.M. Keynes, (1921) A treatise on probability London, Macmillan and Co

Crocco, Marco, (2002) The concept of degrees of uncertainty
in Keynes, Shackle, and Davidson. Retrieved: 16 April 2004 from http://www.google.com.au/search?q=cache:l11T8vjzjc4J:
www.face.ufmg.br/
novaeconomia/sumarios/
v12n2/Crocco.
pdf+
keynes+uncertainty&hl=en
J. Barkley Rosser (2000) ALTERNATIVE KEYNESIAN AND POST KEYNESIAN
PERSPECTIVES
ON UNCERTAINTY AND EXPECTATIONS Journal of Post Keynesian Economics,
Summer
2001, vol. 23, no. 4, pp. 545-566 retrieved 16 April 2004 from http://www.google.com.au/search?q=cache:HKbHmueEOrEJ:cob.jmu.edu/rosserjb/
UNCRTEXP.ECT.doc+
nonergodicity+&hl=en
Peter Lichtenstein (1983) An Introduction to Post-Keynesian and
Marxian
Theories of Value and Price. London: Macmillan