Experimental economics — a fast-growing branch of economics that involves the creation of a microeconomic environment in a laboratory — is being widely used at the John M. Olin School of Business. Applications of the exerimental research have multiplied, spanning many industries and producing results that have impacted everything from how airlines price their tickets to how companies manage their employees. Ronald R. King, Ph.D., is helping to lead the burgeoning new area of research at the business school.
In her first year of the Ph.D. program at Olin, Shawn Davis, Ph.D. ’02, knew she wanted to focus her research on some aspect of auditing but was unsure as to what methodology to employ. When she assumed the role of teaching assistant, under the guidance of King, the answer presented itself.
“I found Ron’s publications and working papers fascinating,” recalls Davis, a visiting assistant professor of accounting at Georgia State University, where she teaches information systems auditing and assurance. “His contagious enthusiasm for research in experimental economics rubbed off.”
For her thesis, Davis designed an experiment that investigated the extent to which public disclosure of auditor materiality thresholds affects both investors’ perception of the auditor’s report and market behavior. Her hypotheses were based on the economic theory of market efficiency and the cognitive psychology theory of overconfidence and under-confidence.
Her findings suggest that disclosing auditor materiality thresholds increases the accuracy of investors’ perceptions of the auditor’s report by reducing investors’ overconfidence. In short, experimental markets are more efficient when auditor materiality thresholds are disclosed as investors price-protect themselves more completely with the information. The research holds implications for current policy debates on full disclosure as well as other auditing measures regarding investment decisions—all-important issues in the post-Enron era.
Ronald R. King, Ph.D.
“The use of experimental markets has become prevalent in business research as a way to test economic and psychological theories under controlled conditions,” said King, Myron Northrop Professor of Accounting, who investigates how legal institutions affect the production and use of information. “Importantly, tying subjects’ pay to their decisions/actions adds validity to the tests of economic theories.”
An example of research conducted by King, along with co-authors Nicholas Dopuch, Hubert C. and Dorothy R. Moog Professor of Accounting, and Rachel Schwartz, associate professor of accounting, is an assessment of the effect of the SEC’s recent rule that requires firms to disclose the amount of non-audit fees paid to their auditors. The new disclosure is intended to inform investors of auditors’ incentives to compromise their independence. The experimental method allows the authors to control what information is observable and what must be inferred from other sources. The results indicate that subjects are prone to “belief persistence,” where people retain erroneous beliefs even after they observe discrediting evidence. This suggests that the disclosures of non-audit fees should be interpreted in the light of other variables related to auditor independence.
The laboratory has proved a popular and efficacious venue for the study of economics. Viewed as a pedagogic alternative to the traditional lecture, the experimental approach has been embraced by faculty and student alike as a vehicle to learn by doing. The method allows students to experience economic theory as applied to specific settings that bring ensuing discussion to life. “For example, pairs of students participate in ‘prisoners’ dilemma’ games where each of the two players has incentives to ‘cheat,’ yet both would be better off with cooperation,” King said. “This situation adds salience to the frictions of settings that have both the possibility of competition and cooperation.”
At Olin, the experiments are conducted in the Taylor Experimental Laboratory. In the Taylor Lab, students are seated at computers where they receive information, make decisions, and interact with other subjects as instructed by the experiment’s designers. Examples of experimental environments constructed in the lab include three-party litigation, negotiation models, contract incentives, the effect of enforceable law on contracts, and information use under uncertainty.
Achieving star status
Experimental economics has developed from the 1960s, when economics experiments were first conducted and reviewed, many in the area of accounting, to the present-day practice of top business schools recruiting experimental economists to serve on faculties. Applications of the research have multiplied, spanning many industries and producing results that have impacted how airlines price their tickets to how companies manage their employees.
For example, in the mid 1990s, the Federal Communications Commission sought experimental economists for help with designing its auction for airwaves for wireless service. More recently, research conducted at Georgia State University concluded that the kite-flying prices of electricity in California could be reduced if consumers could alter electricity access at certain times in return for price discounts.
Major corporations such as IBM and Hewlett-Packard have fallen into step, establishing their own labs in which they use experimental markets to forecast sales and profits, among other activities.
Such experimental approaches achieved star status in October 2002, when economist Vernon Smith of George Mason University received the Nobel Prize for establishing laboratory experiments as an essential tool in empirical economic analysis. (He shared the prize with Daniel Kahneman of Princeton, whose pioneering work regarded integrating psychological research into economics.)
Smith, who served as King’s thesis advisor at the University of Arizona, visited Washington University just days after he was notified of the award to participate in an interdisciplinary conference on trust and reciprocity in experimental economics. Sponsored by the Olin School and its Center for Research in Economics and Strategy; the Weidenbaum Center on the Economy, Government, and Public Policy; and the Center for Interdisciplinary Studies at the School of Law, the conference featured presentations by psychologists, political scientists, anthropologists, and economists.
“The interaction among the social sciences was revealing,” noted Bill Bottom, professor of organizational behavior at Olin, who researches how people use information, evaluate risks, and make decisions. Bottom, whose published work relies primarily on experimental methods, helped to organize the conference. Among the presenters: anthropologist Jean Ensminger of the California Institute of Technology, who spoke on how market forces shape cultural norms in small-scale societies, and Smith, who explained the impact of neural functioning on individual behavior in exchange relationships.
The dynamic between economic- and psychological-based theories is intensifying. “Economics has forced psychologists to think more carefully about how institutions like markets, voting, and organizations shape individual beliefs and actions,” Bottom said. “Psychologists have forced economists to think more carefully about how biology, cognitive limitations, and emotions influence the way individuals deal with complexity. The conference at Olin illustrated this mutual influence.”
Also evident is the increasing influence of experimental economics on the investigation of policy proposals. Ideas can be tested in the lab at a fraction of the cost they might incur being hastily implemented and then failing to deliver on their promise. Moreover, King noted that no archival data exist to test a proposal to determine the potential benefits and/or the unintended consequences.
Conversely, critics of experimentalist literature see limitations in the lab. Some view experiments as too artificial to reflect genuine, i.e., real-world situations and results. Others note that subjects may play into behaviors expected by the experimenters or be swayed by contrived incentives that may skew rational choices.
One thing is certain: Experimental economics has made valuable contributions to the understanding of markets, institutions, and human behavior, lending new meaning to the notion of playing by the rules of the game.