“The theory of economics does not furnish a body of settled conclusions immediately applicable to policy. It is a method rather than a doctrine, an apparatus of the mind, a technique of thinking, which helps its possessor to draw correct conclusions.”
– John Maynard Keynes –
“State Dependent Price Setting Rules Under Implicit Thresholds: An Experiment”
Joint with J. LeBlanc, A. Civelli and C. Deck [pdf] – Journal of Economics Dynamics and Control
Abstract: How firms make their pricing decisions is a fundamental question of macroeconomics. We use a laboratory experiment to examine individual choices in a price updating task that provide insight into how well state dependent models reflect behavior. We find that in general subjects behave as if they recognize the importance of a state dependent pricing strategy, but they are unable to ascertain this threshold with precision and they also exhibit a substantial degree of time dependence. As a result, they update prices too frequently, and perform statistically significantly fewer real effort profit-generating tasks than theoretically optimal under full state dependence, which results in statistically significantly lower profits as well.
“The Effects of Alcohol Use on Economic Decision Making”
Joint with C. Deck, S. Jahedi, L. Ham [pdf] – Forthcoming – Southern Economic Journal
Abstract: In a controlled laboratory experiment, we study the causal effect of alcohol on economic decision making. A treatment group was given a dose of alcohol designed to target a blood alcohol concentration (BAC) of 0.08 while the BAC of those in the control group remained 0.00. We investigate the behavior of control and treatment groups in the following types of tasks: math, uncertainty, overconfidence, strategic games, food choice, anchoring, and altruism. Our results indicate that alcohol consumption has little systematic effect on economic behavior, at least for the BAC level considered. Further, there is little evidence that alcohol differentially impacts the choices of male and female subjects.
Working & Work in Progress Papers
“Overconfidence and (Over)Trading: The Effect of Feedback on Trading Behavior” [pdf]
Abstract: The existing literature argues that people overestimate the accuracy of their information which then leads to high trading volume and lower profits in financial markets. This paper reports a laboratory experiment designed to confirm this conjecture and to identify how feedback about the accuracy of information can limit the negative impact of overconfidence. An experiment, similar to Devas et al. (2009), is designed to study trading behavior when signals depend on the performance on an overconfidence task. In this experiment, overconfidence leads to higher trading volume in the absence of feedback, but this effect disappears when participants are given feedback. Contrary to previous research, there is only weak evidence of overconfidence negatively affecting trading profits. Markets with heterogeneous accuracy of information, due to the performance on the overconfidence task, aggregate information at similar levels to markets where the accuracy is homogeneous and the distribution of information is common knowledge. Lastly, no gender differences are found on the level of overconfidence and there is only weak evidence of an effect of gender on trading volume and profits.
“The Impact of Overconfidence in Acquiring Information”
Joint with D. Forbes
Abstract: Overconfidence has been shown to produce excess market entry, increased volatility in financial markets, and excessive investments. In this paper we offer a possible mechanism through which overconfidence can lead to such consequences. We argue that overconfident people incur a greater intrinsic cost when they ask for help and as a consequence ask for less help. To test this idea a two-part experiment is employed. In the first part subjects complete a simple counting task which is used to calculate two measures of overconfidence, overestimation and overplacement. The second part of the experiment allows subjects to ask for free help that improves the chances of getting the prize. The results show that both overconfidence measures are negatively correlated with the amount of help that subjects seek, consistent with overconfidence leading to bad market outcomes.
“The Role of Overconfidence in the Insurance Markets”
Abstract: (coming soon)