(M.Allais, Kahneman, V. Smith)
Maurice Allais, a Ph.D. in Engineering and a Prof. of Mechanics at Lyons turned to Economics. Maurice Allais’s contributions to pure Theory and his first book Inquest of Economic discipline are in French and many do not know the contents. Allais formally reports of experiments in economics in his article in Econometrica as early as 1953 but the article is also in French. What is better known is his work on decision theory, and in particular the so called Allais Parodox
Allais’s Parodox
Utility measurement passed through several phases-cardinal utility, Ordinal Utility, Behaviouristic ordinalism and to neo-Classical utility under risk. Nuemann and Morgenstern (N-M) have devised a method of measuring utility under condition of Risk, According to N-M method, individuals do not maximize expected money but expected Utility. By way of criticizing N-M method, Allais raised a paradoxical decision situation.
Suppose a person is asked to choose between the following alternatives: Lottery L1 which offers Rs.2 crores for certain and another Lottery L2 which offers a 10% chance of winning Rs.10 crores and 89% chance of winning Rs.2 crores and 1% chance of getting nothing. In this case, anyone will choose L1. Now consider another choice situation. Lottery L3 offers 11% chance of winning Rs.2 crores and 89% chance of earning nothing. Another Lottery L4 offers 10% chance of earning Rs.10 crores and 90% chance of earning nothing. Let us choose between L3 and L4. Many others will choose similarly like us. The preference is for L4 over L3.
Our choices are not consistent with excepted Utility given by probability multiplied by utility. If the expected Utility from L1 is greater than L2, then, the expected Utility from L3 must be greater than L4 Denote the Utility values of the outcomes U10, U02 and U0 (the subscripts indicate the amount of winnings). Expected utility of L1 is presented on the left side and L2 on the right side of Eqn.(1).
Then the choice of L1 to L2 is represented by greater than symbol.
U02> (0.10) U10 + (0.89) U02 + (0.01)U0 Eqn.(1).
Adding (0.89) U0 – (0.89) U02 to both sides, we get
(0.11)U02 + (0.89) U0 > (0.10) U10 + (0.90) U0 Eqn.(2).
The expected utility of L3 is given on the left side and that of L4 on the right side of Eqn.(2).
As per Eqn.(2) L3 must be preferred to L4 (and not L4 to L3 as indicated by our choice)
Thus Allais has shown that certain kinds of risky choice could not be squared with expected utility theory. This and many other anomalies in choice behaviour have been thoroughly explored by both Psychologists and Economists.
Daniel Kahneman
Daniel Khaneman, a Professor of Psychology at Princeton University, have used insights from Psychology to study human behaviour and to conduct experiments in individual decision making under uncertainty. He argued that in complex decision situations under uncertainty, individuals do not make rational calculations, as assumed by traditional theory. Instead, individuals rely on heuristic short - cuts or rules of thumb. Khaneman (and Tversky) have developed the ‘Prospect theory’ of decision making under un-certainty. In this theory, individuals are assumed to be sensitive to the way an outcome deviates from statusquo than to the absolute level of outcome. And individuals are more averse to losses relative to the statusquo than they are partial to gains of the same size.
Suppose, you have invested in a start-up company, which is making profits (Company P). You have a 90% chance of winning Rs. 100 lakhs and a 10% chance of receiving nothing. If some one offers to buy the asset from you, for Rs. 85 lakhs, most likely you would accept the offer because the latter option has less risk. You would be exhibiting risk averse behaviour.
Now consider another situation involving huge losses. Suppose you had invested in a start-up company, which is incurring losses (Company L). There is a 90% chance of losing Rs. 100 lakhs but 10% chance of losing nothing (nil losses). Another investor offers to take-over the company if you pay him Rs. 85 lakhs (resulting in a certain loss of Rs. 85 lakhs). You would most likely reject the offer and choose to retain the loss making company.
The loss making unit case is exactly similar to the earlier profit making case. But, in the profit making case, you exhibit risk averse behaviour and in the loss making case, you don’t exhibit risk averse behaviour. This is called ‘a framing effect’.
This Prospect theory can explain why people take out expensive small scale insurance, why people buy expensive service contract for appliances that would be cheap to replace and such other individual (irrational) decisions.
Vernon L. Smith
Smith established laboratory experiments as a tool in empirical analysis, especially in the study of alternative market mechanisms. Smith (and Knez) tested a ‘strong market hypothesis’, which states that markets equilibrate as if agents were Utility maximisers even if the agents do not themselves behave as if they were Utility maximisers. They state this point of view as follows:
“The efficiency and social significance of markets does not depend on the validity of any particular theory of individual demand…. The empirical validity or falsity of efficient markets theory is a proposition that is entirely distinct from the empirical validity or falsity of theories of individual demand in markets”.
Smith (and Knez) conducted experiments to test the market hypothesis and the results confirm their hypothesis. The behaviour of some individuals might be irrational but the market behaviour of all is rational and efficient. Smith’s book Bargaining and Market Behaviour contains his experimental findings.
Smith’s latest interest is in Neuro-Economics. He uses brain scanning of experimental subjects playing economic games. The exponents of Neuro-Economics believe that by brain scanning of experimental subjects, they will be able to peer directly into the brain to predict behaviour.
The three Nobel Economists discussed in this chapter are pioneers and key figures in the experimental economics. Nowadays experimental work in economics is done in many areas such as investigating two-person bargaining problems; the free rider problem in the provision of public goods: and in examining auction markets and in privatization of public monopolies.
The study of human behavour based on human psychology and experiments falls into the category of a newly flourishing field of Behavioral Economics. The new field is again discussed in the Chapter on Markets.