Trading in the ‘Pit Market’

1. Trading in the ‘Pit Market’ Dr. Nikos Nikiforakis The University of Melbourne Pricing and Efficiency in Competitive Markets

2. Overview • Purpose of this talk: • Use experiments and experimental results as a vehicle for discussion • Explain what we mean by ‘markets’ • Derive and explain theoretical predictions of market outcomes and desirability of competitive markets • Discuss (briefly) the importance of trading institutions • Ultimately, goal is to show the complementary role of economic theory and experiments

3. What is a market? • “A place or institution in which buyers and sellers of a good or asset meet”. (Oxford Dictionary of Economics) • Markets facilitate trade • Bring buyers and sellers together • Aggregation of information • Help traders strike agreements

4. Theoretical framework • Optimization principle – people try to achieve the best deal they can. • Units are homogeneous. • Equilibrium – the price(s) where neither buyers nor sellers have an incentive to change their behaviour.

5. Demand side (buyers)

6. Demand side (buyers)

7. Supply side (sellers)

8. Supply side (sellers)

9. Market

10. Market

11. Market Consumer surplus Producer surplus

12. Market predictions • Let p* be the price where supply and demand curves cross. • If p>p* => excess supply will tend to lower prices. • If p<p* => excess demand will tend to raise prices. • If p=p* there is no force to change the price. • The market price (p*) determines the quantity to be traded (q*)

13. About experiment (1) • Variant of first recorded experiment in economics (Chamberlin, 1948) • Chamberlin found that market predictions don’t work! • I “predict” that the same will apply in our case (reasons will be explained).

14. Behaviour in experiment (1)

15. Behaviour in experiment (1)

16. About experiment (2) • Variant of experiment by Smith (1962) – Nobel laureate 2002. • Smith argued that Chamberlin’s experiment was too decentralized unlike real markets. • Experiment (1) was even more decentralized. In fact it was hardly a market.

17. About experiment (2) • Smith (1962) found that the predictions work remarkably well in a centralized trading institution (double-auction). • Experiment (2) was less centralized than Smith’s, but it should still have worked alright. Let’s see…

18. About experiment (2)

19. About experiment (2)

20. About experiment (2)

21. Behaviour in experiment (2)

22. Behaviour in experiment (2)

23. Summary of what we learned • Economic theory can help us predict market outcomes. • Creation of markets can lead to gains from trade, lower prices, increases in efficiency. • Competitive markets maximise efficiency. • Competitive markets work under very “disadvantageous” conditions • Few buyers and sellers (theory assumes that many more would be required) • Minimal informational requirements (i.e. each person only needs knows his/her value/cost)

24. Question 1 • Would our results change if instead of both buyers and sellers shouting prices we had only sellers announcing prices? • Would theoretical predictions change? • Would behavior change?

25. Question 2 • Why do you think we decided to pay you for your performance in only one of the experiments?

26. Question 3 • The optimization principle assumes that you will try to strike the best deal possible. That is, you would prefer a deal with a profit of $1 to no deal. • Would you accept this deal if you knew the other person was making a profit of $10? • How would this change the results?

27. Further reading Microeconomics: Varian, Hal. (1999) Intermediate Microeconomics – A Modern Approach, Norton & Co, New York. Experimental Economics & Institutions Kagel, J. and Roth, A. (1995) The Handbook of Experimental Economics, Princeton Univ. Press, New Jersey. Davis, D. and Holt, C. (1993) Experimental Economics, Princeton Univ. Press, New Jersey. Holt, Charles (1995) Industrial Organization: A Survey of Laboratory Research, in The Handbook of Experimental Economics, Princeton Univ. Press, New Jersey. (available on-line)