Printer Friendly

Teaching tools: a public goods experiment for the classroom.


George Stigler wrote in the American Economic Review twenty-seven years ago, that the typical student in an introductory college economics course would "memorize a few facts, diagrams, and policy recommendations, and ten years as untutored in economics as the day he entered the class" [1963, 657]. And later, in the Journal of Economic Education, Stigler argued that "economics belongs in everyone's education once we have learned how to teach it" [1970, 80].

As we enter the nineties, academic economists continue groping for methods and techniques that will improve the way in which principles courses are taught. The proliferation of commercially available materials in the form of textbooks, workbooks, software, telecourses (e.g., Economics USA), and transparencies provides evidence of the increasing interest and expenditure aimed at improving the teaching of economics. However, there exists at least one very promising teaching technique that continues to see relatively limited action in the typical economics classroom. While experimental economic methods have become quite familiar to academic economists as a legitimate and powerful research tool, their usefulness for the economics classroom has not yet been generally recognized.(1)

In this article, I describe a simple, yet powerful, public goods experiment for use in microeconomics or public finance courses. Since it is likely that many instructors in economics are reluctant to use experimental methods without sufficient preparation,(2) enough detail on experiment administration and suggested post-experiment discussion is provided to assist even the novice experimenter with getting started. Instructors interested in additional information on experimental economics or experiments for the classroom should consult the references listed in the concluding section.


A theoretical concept that introductory students often find hard to understand is the vertical summation of individual demand curves to obtain the market demand for a public good. The typical student also fails to fully appreciate how a free rider can cause underinvestment in public goods. The experiment outlined below provides a simple, yet powerful, technique to teach not only the vertical addition and the free rider problems, but also an ideal way to construct a prisoner's dilemma game and discuss the resulting underinvestment in the public good.

Students are placed into N teams, where a "team" may be as few as one person. Each team is endowed with T tokens and must choose, in each of the decision-making rounds, whether to "invest" these tokens in the "private account" or the "group account." The private account will pay the team $0.30 per token invested each round. The group account will earn, for the entire "society," $0.50 per token invested each round. From the group account, each team will then receive ($0.50) [G.sub.i]/N where [G.sub.i] equals the total number of tokens invested in the group account in round i by all N teams. Note that regardless of how many tokens any single team places into the group account, all N teams receive the same payout from the group account. Teams must invest all T tokens in each round and tokens cannot carry over from one round to the next. Finally, earnings from previous rounds are not allowed to be reinvested in the future.


Assume a representative class of sixty students.(4) Assume the instructor divides the class into twenty teams (N = 20) of three students each. Each team is endowed with T = 25 tokens per round. The teams read the "INSTRUCTIONS" overhead transparency while the instructor distributes to each team: (i) the "TEAM RECORD SHEET," and (ii) the "TEAM DECISION BALLOTS" (one for each round).(5) For each round, once all the team ballots have been submitted, the instructor sums the tokens invested in the group account and reports this number and the dollar payout per team to the entire class. Each team then completes the TEAM RECORD SHEET entering its 5 percent share from the group account and its $0.30 per token from the private account investment earnings.

Of course, unless the instructor is independently wealthy or receives a grant, the students will not receive payments in cash for playing the game. Therefore, in order to motivate the students, the instructor may choose to offer extra credit points (e.g., a maximum of five points) based on actual team performance as a percentage of maximum potential performance.(6) Given the numbers used in this example, and assuming four rounds are completed, the maximum potential payout is $50 per team.(7) Thus, a team that earns $20 (or 40 percent of the maximum potential) would be awarded two of the five possible extra credit points per team member.

Experiment Variation

One enjoyable and worthwhile modification of the experiment is to allow the teams to "collude" after a few rounds. Once the teams become frustrated with the relatively low dollar payouts being achieved in the early rounds, they may request that communication across teams be permitted. This is the perfect opportunity for the instructor to allow the teams to form a cartel. The instructor departs the room for a few moments during which the students are free to discuss whatever they want. Upon the instructor's return, the teams are then given the opportunity to complete their ballots privately.


Although the outcome will vary depending on the exact specification of the relevant parameters, instructors can anticipate results that are roughly consistent with Figure 1. The number of tokens invested in the group account (public good) will typically be 30 to 60 percent of the total possible in the first round and decline each round until collusion is permitted (period 4 in our example). Although collusion should increase investment in the public good, a significant underinvestment is still likely.

The instructor can explain the vertical summation of demand by reference to the structure of this experiment. In this particular example, the marginal social benefit (MSB) to investment in the public good is $0.50 per token. Since the opportunity cost of investing in the public good is the foregone earnings from the private account, the marginal private cost (MPC) of investment in the public good is $0.30 (which is equal to the marginal social cost, MSC). Therefore, as shown in Figure 2, since the MSB curve always exceeds the MSC, the socially optimal investment in the public good is 500 tokens (i.e., 20 teams x 25 tokens per team) per round.

The private market, however, fails to provide the optimal amount of the public good due to the inability of teams to exclude others from the benefits. Since teams must share the earnings from the group account equally, the marginal private benefit (MPB) to any team is $0.025 (i.e., $0.50/20). By including the MPB curve in Figure 3, the private optimality can be shown to be zero, and the underinvestment in the group account explained.

Once the construction of Figure 3 is complete, the instructor can derive the market demand curve for the public good by vertically summing the willingness to pay by all twenty teams. As an extension of the classroom experiment, instructors may want to assign a homework problem in which MSB declines so that students can construct a normal downward-sloping demand curve.

If game theory is included in the course, this is an excellent opportunity to construct a prisoner's dilemma payoff matrix for a simplified version of our experiment. Abstracting from the complexities that arise with an n-person prisoner's dilemma game, assume a two-person (or two-team) game with the other parameters of the above example experiment remaining the same. Assuming that the teams invest either in all private (P) or in all group (G) account investments, then the resulting payoff matrix is depicted in Figure 4. The Pareto optimal solution is for both teams to invest in the public good earning a total of $12.50 per team each round, or $50 each team during a four-round game. The dominant strategy, however, is to invest in the private account, thus resulting in a single round total earnings of $7.50 per team. Using this payoff matrix, the instructor can discuss the actual experiment results with reference to such issues as repeated games and end-game problems.


Experimental economics is a relatively new tool that gives economists the ability to test their assumptions and theories in ways that were previously not available. Although experimental economics is still a young field, it has already succeeded in demonstrating its value for the science of economics. Additionally, experimental research can provide a significant positive externality for the teacher. A pool of interesting and relevant experiments covering a wide range of behavior is now available for teaching introductory and intermediate-level microeconomics.

Many experiments have been devised to test economic theories. Hoffman and Spitzer [1982] develop an experiment of Coase's theorem. Isaac and Plott [1981] and Bishop [1986a] discuss experiments appropriate for demonstrating cartel behavior. In addition to the above references, the interested reader should consult Bailey [1989], Bishop [1986b], Hoffman and Spitzer [1985], Plott [1982], and Taylor [1988] for a further discussion of experimental economics.



This is an experiment in the economics of group decision making. The class will be divided into _____ "teams" (each team being composed of one or more members.) The experiment will occur over a sequence of _____ decision-making rounds. At the start of every round, each team will be endowed with _ "tokens." In each round, every team must "invest" the endowed tokens in either a "Private Account" or in a "Group Account." Each team in the class has a Private Account, however there is only one Group account for the entire class.

Each team will "earn" $0.30 for each token that is placed in the Private Account. Each token placed in the Group Account will generate earnings of $0.50 for the entire class. The earnings from the Group Account will be divided equally among all teams, regardless of the number of tokens that any individual team places in the Group Account. [Figures 1 to 4 Omitted]

(1)Of the forty academic economists who attended the Seminar on Laboratory Experiments for Undergraduate Instruction in Economics at the University of Arizona, only a few were familiar with the use of experimental economics in the classroom. Also, in light of the rapid expansion in the production of textbook peripherals, it is somewhat surprising that, to the author's knowledge, separate experimental packages are not yet available. (2)The vast majority of the faculty present at the Seminar on Laboratory Experiments expressed a particularly strong interest in discussing experiment details before conducting any experimentation at their respective institutions. (3)The basic structure of this experiment is similar to the one reported by Isaac and Walker [1988]. (4)A class size of sixty is representative of the national average for an introductory course reported by Siegfried and Wilkinson [1982, 136]. (5)An example of the Instructions transparency and experiment handouts are included in the Appendix. (6)It is important that the teams are motivated to be "earnings maximizers." If the instructor wants to create a true prisoner's dilemma situation, then the common-interest Pareto optimal outcome must be induced. (7)This number is the Pareto optimal outcome derived as follows: (20 teams x $0.50 per token x 25 tokens x 4 rounds) / 20 teams = $50. The maximum potential payout is actually greater than $50. If one team could completely free ride on the other nineteen teams, then they could earn $55. Using $55 as the base in the percentage calculation for extra credit points may, however, be somewhat "conservative."


Bailey, Ronald. "The Economists' New Guinea Pigs." Forbes, 13 November 1989, 148-52. Bishop, Jerry E. "All for One...One for All? Don't Bet on It." The Wall Street Journal, 4 December

1986a, 37. ______. "Lab Experiments Test Old Economic Rules, Raising New Questions." The Wall Street Journal,

25 November 1986b, 1. Hoffman, Elizabeth and Matthew L. Spitzer. "The Coase Theorem: Some Experimental Tests." Journal of

Law and Economics, April 1982, 73-98. ______. "Experimental Law and Economics: An Introduction." Columbia Law Review, June 1985, 991-1036. Isaac, R. Mark and Charles R. Plott. "The Opportunity for Conspiracy in Restraint of Trade: An

Experimental Study." Journal of Economic Behavior and Organization, 4(2), 1981, 1-30. Isaac, R. Mark and James M. Walker. "Group Size Effects in Public Goods Provision: The Voluntary

Contribution Mechanism." The Quarterly Journal of Economics, February 1988, 179-99. Plott, Charles R. "Industrial Organization Theory and Experimental Economics." Journal of Economic

Literature, December 1982, 1485-527. Stigler, George. "Elementary Economic Education." American Economic Review, May 1963, 653-59. ______. "The Case, If Any, for Economic Education." Journal of Economic Education, Spring 1970,

77-84. Taylor, Herb. "Experimental Economics: Putting Markets Under the Microscope." Federal Reserve Bank

of Philadelphia Business Review, March/April 1988, 15-25.

JOHN R. BROCK Associate Professor of Economics, U.S. Air Force Academy, Colorado. The author is grateful to Donald Wells, Arlington Williams and the participants of the University of Arizona and National Science Foundation jointly sponsored Seminar on Laboratory Experiments for Undergraduate Instruction in Economics, May 1989, for stimulating interest in this subject. Robert Moore and an anonymous referee provided helpful suggestions.
COPYRIGHT 1991 Western Economic Association International
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1991 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Brock, John R.
Publication:Economic Inquiry
Date:Apr 1, 1991
Previous Article:A comment on competition and bidding behavior: some evidence from the rice market.
Next Article:Why doesn't society minimize central bank secrecy?

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters