EconTalk |
Vernon Smith on Markets and Experimental Economics
May 21 2007

Vernon Smith, Professor of Economics at George Mason University and the 2002 Nobel Laureate in Economics, talks about experimental economics, markets, risk, behavioral economics and the evolution of his career.

Vernon Smith and James Otteson on Adam Smith
Vernon Smith and James Otteson talk with EconTalk host Russ Roberts about Adam Smith in front of a live audience at Ball State University. Topics discussed include Smith's view of human nature, the relevance of Smith for philosophy and economics...
Vernon Smith on Adam Smith and the Human Enterprise
Nobel Laureate Vernon L. Smith of Chapman University talks to EconTalk host Russ Roberts about how Adam Smith's book, The Theory of Moral Sentiments has enriched his understanding of human behavior. He contrasts Adam Smith's vision in Sentiments with the...
Explore audio transcript, further reading that will help you delve deeper into this week’s episode, and vigorous conversations in the form of our comments section below.


May 21 2007 at 7:53pm

Great interview. You can’t do much better than a free-market Nobel Prize winner.

It would be great to hear you interview Art DeVany about his research on movie box-office returns and possibly his evolutionary fitness approach.


May 22 2007 at 7:27pm

Definitely a great interview!

How about some topics like campaign finance regulation? There is a big Supreme Court case coming up. How about crime? What about this abortion and crime debate?

Jun 11 2007 at 1:18am

Great interview!

Comments are closed.


About this week's guest:

About ideas and people mentioned in this podcast:Books:


Web Pages:

Podcasts, Video, and Blogs:



Podcast Episode Highlights
0:38What is experimental economics? While Perdue U., 1955, teaching Principles of Economics. What is connection between what people actually do in markets and the theory of supply of demand as taught? Two stories to explain market equilibrium: 1. if equilibrium is reached, it's because people in the market have complete information; or 2. alternative is that there have to be a large number of buyers and sellers who all have to be price-takers. Not satisfactory. January 1956 tried first classroom experiment: made half buyers and half sellers, gave buyers secret values (demand schedule) and sellers secret costs (supply schedule). Buyers profit from exchange (value exceeds price they pay). Two-sided auction: buyers announced bids, sellers announced asking prices. Contracts on acceptance. Some information but incomplete information available to each. Market converged to price and volume represented by theoretical intersection of supply and demand curves--competitive equilibrium. "This is some kind of magic!" Experiment was first day of class, students knew nothing about theoretical economics. What it was showing is that people are quite good at discovering prices and quantities that clear a market even if their individual goals are only private benefit.
8:29But: If you asked the students if there was any kind of model, there was widespread denial that any model could have predicted that price and volume. Write equilibrium price and volume in sealed envelope. Also didn't perceive that they were making as much money as they possibly could (which is implication of theoretical equilibrium. John Nash: each person is doing best he can for himself given the behavior of everyone else). Don't have perception of gains from trade and welfare-maximizing benefits of trade, nor understanding that they are doing very well for themselves. Also satisfied Alfred Marshall's insights into equilibrium: total gains maximized, no forgone benefits to group as a whole, surplus is maximized. Tried more experiments. Turned out that even 4-6 people were enough. Textbooks today, though, all still say that you need either perfect information or infinitely many people. Even with only a few people, sellers compete with sellers, buyers with buyers, implying it is hard to collude even with only a few people in the market. How sensitive are the markets to the trading rules? Posted offer markets: sellers post take- or leave-it offers each period? Those markets have more inertia, don't adjust as rapidly as two-sided auctions, but still get there. First paper on this published 1962. Rejected earlier by Journal of Political Economy (JPE), U. of Chicago. "If you believe in markets you don't necessarily need evidence." George Stigler eventually became a fan of experimental economics. Harry Johnson took over as new editor and ultimately published the revised paper. Johnson: "I haven't been at this job very long and I've learned a lot. I've discovered that you have to keep evaluating everybody, including yourself. I have to confess that I was one of the original referees (that was negative on the paper), but you've convinced me." Smith: "Harry Johnson was the last editor of the JPE who read everything that came in."
20:01Interdependence. This was a simple market. By early 1980s we started doing experiments where subjects interacted through computers. Became possible to do more complex experiments. Around 1982, amount a buyer was willing to pay for commodity A depended on price of B. Double-auction trading solved that problem also, convergence to equilibrium in about 7-8 trading periods. Four non-linear equations mathematically, but subjects didn't need to know this. They only tried to do well for themselves individually. Invisible hand. Humans solve problems through institutions like auctions. English auction (traces to Babylonians): buyers put up fixed quantities for sale. Modern double auction evolved from that. Failed institutions are not recorded. Traditions, trading rules in markets, economic institutions. Underneath is where you find the specialization that exchange makes possible. Adam Smith, division of labor is determined by the extent of the market. "People can't find specialties that take advantage of their own unique characteristics unless there is a means whereby what they earn in that specialty can be used to buy the products of others." People anciently had to discover property rights, trade, etc.; can't have any of these without the other. They are simultaneous, interactive. Wealth is created. Hard to explain to students.
28:41Trying to create experiments to see if people can discover both specialization and exchange. Bart Wilson, two years ago. Don't even use the words trade or prices. People in chat room can use words, talk to each other. Create conditions where they can triple their earnings if they discover specialization and exchange. In given period, they decide how much Blue and how much Red to produce. Get payoff function such that Red and Blue are complements, have to be consumed in fixed proportion, but the proportions differ by subject. If one specializes in Red and the other Blue but don't trade, they make nothing. But if they discover both specialization and exchange they make three times as much as if they just engage in home production. Economies of size 2, 4, and 8. Interestingly, some people are curious and explore and discover specialization and exchange; some never discover it (autarchy, self-sufficiency); some discover it late (developing). Three kinds of economies. Can study property rights, enforceable and non-enforceable; innovation, change. Not everyone gets to equilibrium (maximum potential) in this kind of world. Will probably be able to show that wealth creation is not independent of the distribution of wealth in this kind of world. Those who discover this do very well for themselves and for the economy, others. Welfare theorems (in textbooks) do not have dynamics, innovation, discovery.
35:13Which economists have influenced you most? 1949-51, teacher Richard Howey, U. of Kansas, development of economic thought. Take a problem you are interested in and then learn all the things you need to know in order to be good at that. Model Smith has applied in his work. At Harvard, Wassily Leontief, who had a certain amount of skepticism about economics and a great sense of humor. After two weeks of studying utility theory, student raised his hand and asked, "Professor, what is utility good for?" Leontief responded "It's good for teaching." Important to keep asking what it's good for. People relate to the examples Russ uses in class. Smith tries to do the same with experiments, to make it possible to relate it to their personal experience. Austrian school, Böhm-Bawerk, work on the theory of value, nice market examples, almost experimental, horse market with small numbers of market participants. Intuitive approach. Hayek, 1945, "Use of Knowledge in Society," impressed but "you can't understand it by reading it once." 1937 paper, "Economics and Knowledge." Bruce Caldwell has traced Hayek's ideas back through 1930s. Innovative line of thinking. "Institutions are information-generating devices." People incorporate the information into an adjustment of their behavior over time. May not be one price, things are moving all the time, but useful exercise to start with things not moving but information allows them to learn over time; then let supply and demand be unstationary, change over time. Movement in right direction. Can supply lots of experimental examples available for Hayek's ideas, which of themselves do not contain many examples. Socialism v. capitalism may have motivated Hayek's work, but lots of wisdom on other ideas. Hayek: "The proper study of social science is the study of that which is not." Characterizes laboratory, experimental economics, where you can study trading rules and institutions that nobody has every observed, which can help you understand which ones have survived.
46:47The word "markets" can mean many things. Farmer's market, NY Stock Exchange. What Hayek meant by markets was something quite amorphous and difficult to put your finger on. Smith quote:
Markets are about recognizing that information is dispersed in all social systems and that the problem of society is to find, devise, and discover institutions that incentivize and enable people to make the right decisions without anyone having to tell them what to do.
Vast, seemingly chaotic but actually ordered events. Variation: Markets as information aggregators. Vernon Smith's Nobel Prize 2002 shared with Daniel Kahneman, psychologist, for behavioral economics, which questions idea that people are rational. Experimental economics interfaces with that, ecological rationality. How does your work interface with behavioral economics? That literature is interested in decision-making under uncertainty, from expected-utility theory: people are not consistently good expected utility maximizers. Missing: how do people adapt to uncertainty not as isolated individuals but together? For example, we trade not only in markets but we trade favors. Roy Radner and co-authors, economics of survival, modeling individual as desiring to survive, different from utility-maximizer, firm. Turns out to not be equivalent to expected profit-maximizing. Risk preference, risk aversion. Critical level of wealth below which you will act as if you prefer risk. You're probably going to go broke anyway, but when poor enough there is a chance to get out of the trap if you take big risk. Experiments with a survivalist might look inconsistent but are explained by survival model. Smith: "I feel that a lot of that work [behavioral economics] is really not linked up with the kind of economic traditions that humans have developed that have ended up, enabled them to create new forms of wealth through specialization and markets." Nobel film interview. Danny is asked what he thinks about the result that humans are not very competent. "Yes, but that can't really be true because look how we've survived." Behavioral economics may not be a proper picture of what humans are or have become unless it is linked up with exchange.
58:21What areas of public policy are most ripe for improvement based on the experiments you've done? Electricity, airport landing slots examples. Markets have an important role to play in improving electric power systems--non-U.S. experience. In U.S. we have a transition from regulated to non-regulated, which is different from foreign experience, which had government-owned electric utilities transitioning to non-regulated. No congestion pricing used at airports to allocate landing slots. Water markets, network market, is another area to consider self-regulating properties. Australia. Yet another kind of market: Government uses incentives akin to markets to try to steer resources, but they often seem to forget or be unable to discover the sources of stability that occur in unregulated markets. California deregulation of electricity was really a change to a highly regulated stylized market. Can the things you create in the lab capture enough complexity to base recommendations on it? Market design can be very flawed when government tries to capture it by a system of rules.

More EconTalk Episodes