Experimental Studies of Antitrust Policy and Portfolio Selection

Abstract

Antitrust and collusion have been the subjects of many economic papers since the passage of the Sherman Act in 1890. However, research in these areas faces numerous challenges. Even when real-world issues are modeled, there are limits to theoretical predictions, as well as relevant legal issues regarding the use of data on cartel firms. Moreover, antitrust policy affects both firm decision-making and social welfare, so any proposed policy changes should be carefully implemented. Comparable challenges are present in the field of research on selecting portfolios of risky assets, where limitations on theoretical predictions about human decision-making and the significant impact of decisions on investments prevail. To address these practical problems, I uses experimental methods to study firms' collusive behavior under antitrust regulation and to compare two different preference theories for portfolio selection. In my first chapter, I explore the impact of strategic delegation between owners and managers on firms' compliance with antitrust policy. Decision-making authority is often delegated to managers in modern firms, and the salaries of executives, such as CEOs and top executives, are mainly based on profit and revenue, not solely on profit. This may result in different outcomes for firms' collusive behavior compared to what is predicted by the standard economic assumption of profit maximization. I compare the formation rates of cartels under both strategic delegation and profit maximization. The results indicate that while strategic delegation does not lead to a greater or lesser rate of cartel formation than profit maximization, it may result in a higher proportion of tacit collusion. In addition, antitrust policies have also influenced the collusive practices of firms, resulting in the formation of cartels that circumvent regulations. In the second chapter, the effects of leniency policies in the US and South Korea on the formation of cartels are analyzed. The US leniency policy offers complete exemption from fines to the first applicant only, while South Korea's policy allows subsequent applicants to receive partial fine exemption. This research focuses on the differences in the policies regarding (1) the size of fines imposed on cartels (High or Low) and (2) the number of applicants granted leniency (Single or Multiple (two)). According to the results, leniency policies that vary in terms of the amount of fines and number of leniency applicants do not have a significant effect on cartel formation, compared to a scenario where no leniency is granted. However, such policies increase the exposure of established cartels, leading to their eventual decline in cartel success. The chapter has been accepted for publication in the Review of Industrial Organization in 2023. The third chapter compares Expected Utility Theory (EU) and Quantile preference (QP) for portfolio selection. EU, commonly used in economics, involves maximizing the average payoffs from assets, while the QP theory maximizes a specific quantile of their payoff distribution. The experiment investigates which preference theory between EU and QP describes participants’ decisions more closely. The results indicate that generally, EU generally explains individuals’ optimal portfolio selections well but QP is a good alternative to EU in the cases where the first stochastic dominance between two assets does not exist. The chapter has been published in the Journal of Behavioral and Experimental Economics in 2022.