Edward L. Millner

An experimental investigation of efficient rent-seeking

ConclusionsThe debate about the measurement of the amount of rent dissipated in rent-seeking activities is unable to be answered by observations from the naturally occurring environment. Current theoretical investigations have indicated that a number of factors will affect the amount of rent that can be assumed to be dissipated by rent-seeking. The theory of efficient rent-seeking predicts that the number of rent-seekers, their risk posture, their ability to enter and exit the rent-seeking process, and the manner in which the probability of receiving a rent is determined will affect the amount of rent that is dissipated.To derive some indication of the appropriateness of the predictions from the model, we report a laboratory investigation of rent-seeking activity. The experiments involved two participants and were based on Tullocks analysis of rent-seeking as a lottery. The empirical results indicate that the manner in which the probability is determined does matter in rent-seeking activities. As predicted by the theory, participants in experiments where the exponent in the probability function equalled three dissipated significantly more of a lottery prize than participants in experiments where the exponent equalled one. Therefore, our results lend support to Tullocks argument that social welfare is protected by adopting institutions which lower the exponent if such an adoption can be effected without adverse effects on the efficiency of the selection process.We view our experiments as a first attempt to use laboratory methods to examine rent-seeking. Much work remains. For example, our results indicate average dissipation rates in excess of that predicted by the Cournot-Nash solution when the exponent equals one and below that predicted by either the Cournot-Nash or a precommitment strategy when the exponent equals three. An obvious question is why? Future experimental work should address the extent to which consistent-conjectures outcomes, focal point solutions, and end-game strategies cause the observed divergence from the predicted values. Also, future laboratory experiments are needed to address other issues raised by the theory such as the effects of changing the number of rent-seekers and of allowing rent-seekers to enter and exit the rent-seeking process. These types of experiments will help refine how we estimate the welfare loss created by rent-seeking activities.

Risk aversion and rent-seeking: An extension and some experimental evidence

Summary and concluding remarksTheoretical investigations indicate that the risk attitudes of individuals will effect the amount of rent that can be assumed to be dissipated by rent-seeking activities. Following this line of investigation we extend Hillman and Katzs work to a small numbers case and demonstrate that the degree to which a monopoly rent is dissipated is dependent upon the structure of the risk attitudes of two risk averse individuals.Our earlier laboratory results were evaluated with respect to the risk neutral Cournot-Nash predictions. However, given the uncertainty present in the rent-seeking experiment, our ability to reject these risk neutral predictions may not be, in fact, a failure for the model but a result of risk aversion. In laboratory experiments in which we control for the relative risk attitudes of individual agents, we show that risk aversion matters. The relative risk aversion of individuals affects the level of rent-seeking activity and the extent to which rents are dissipated. In our experiments, the relatively less risk averse individuals dissipated relatively more rent.In future work, it would be instructive to conduct a set of laboratory experiments in which a reliable control for risk neutrality is introduced. This would allow for a more precise test of the risk neutral Cournot-Nash predictions.

Rebates, Matches, and Consumer Behavior

An experiment conducted to examine the effects of different discount formats on consumer purchases is reported. Participants made a series of purchase decisions for chocolate bars given (1) “rebates” from the listed price, (2) “matching” quantities of chocolates for each bar purchased, and (3) simple price reductions. Contrary to standard theoretical predictions, and consistent with results in the context of charitable contributions by Eckel and Grossman (2003), we find that participants purchase significantly more chocolate bars under a “matching” sales format than under a comparable “rebate” format. Inattention to the net consequences of decisions, as well as some “rebate aversion,” explain the preference for matching discounts.

Contestability in Real-Time Experimental Flow Markets

This article reports the results from laboratory markets using a new trading institution in which sellers post offers to sell flows of output in real time. We use the new institution to examine the outcomes in natural monopoly experimental markets. No. stable price outcomes were observed in markets contestable by two firms, a finding that is consistent with the contestable markets hypothesis, but in conflict with assertions commonly made by those advancing the theorys role in policy formation. The average efficiency of markets contestable by two firms, while greater than that of monopoly markets protected from entry, was significantly less than that associated with the sustainable equilibrium.

A qualitative response model of student performance on a standardized test

SummaryA national sample of high school students who took the Test of Economic Literacy was analyzed by the use of probit analysis and OLS to determine what factors influence high levels of performance on an achievement test. The statistically significant factors were combined SAT scores, the teachers course work in the subject matter, whether the teacher has a high degree of enthusiasm toward the teaching of the subject, and, for students who scored over 30 on the TEL, whether the student does more than two hours of homework a night.

Output and welfare effects of optimal price discrimination in markets segmented at the initiative of the seller

Abstract This paper investigates the output and welfare effects of optimal price discrimination in markets segmented at the initiative of the seller. A market with n segments and prices is compared to a noncooperative Cournot-Nash oligopoly with n competitors. Output and welfare are greater than (less than, equal to) Cournot-Nash levels when the demand function is concave (convex, linear) from below.

A re-examination of Harrison's experimental test for risk aversion

Abstract This paper examines several aspects of Harrisons (1986) test for the risk attitudes of experimental subjects. We identify specific refinements to Harrisons test that must be adopted in order to establish correctly a subjects risk posture.

On the comparative statics of the dominant-firm model

This paper examines the effect of a change in demand upon a dominant firms share of the market. The direction of change in market share is not readily determined in the general form of the model. It depends upon the values of a number of parameters. In the linear and log-linear forms of the model, however, the number of crucial parameters reduces to two. The market share of the dominant firm varies directly (inversely) with demand when the y-intercept of its marginal cost curve is greater (less) than the y-intercept of the supply curve of the fringe in the linear model and, in the log-linear model, when the elasticity of marginal cost with respect to output is less (greater) for the dominant firm than it is for the fringe.

Comparable worth and rent seeking

Sex-based wage discrimination has attracted wide attention among economists and jurists alike. Decisions in cases brought under the US Equal Pay Act of 1963 and the Civil Rights Act of 1964 establish that it is illegal to pay women less than men when both perform identical work. However, women continue to earn significantly less than men on average. In part, the earnings differential reflects the fact women tend to be crowded into relatively low-paying occupations. To address the alleged discrimination that occurs when men and women perform different jobs, it has been suggested that workers be paid according to the “comparable worth” of their jobs. Under this proposal women would be entitled to payments equivalent to men when their jobs require equal degrees of skill, training, experience, responsibility, and effort. A number of federal and state court cases in the US have sought, and continue to seek, interpretations of existing statutes to require equal pay for jobs of comparable worth. Extensive political activity is now underway at the state level seeking to enact new comparable worth legislation. Efforts to implement the comparable worth principle are undoubtedly motivated, in part, by considerations of equity and fairness. However, activities that promote the comparable worth doctrine impose both costs and benefits on diverse economic groups and individuals. This paper has two purposes. First, we use economic analysis to investigate the net welfare effects of seeking to implement the comparable worth principle. Second, we examine the economic implications of the analysis to determine how activities that promote the comparable worth doctrine will affect the current legal status of comparable worth at the federal, state and local levels. We show that implementing comparable worth policies would create artificial economic rents in the labor market. Efforts to promote the comparable worth