My research activities have shifted from traditional asset pricing and financial econometrics to various aspects of experimental finance. These include, but are not limited to: (i) the study of general equilibrium theory of asset pricing using large-scale, web-based experimental markets, with focus on the impact of cognitive biases on prices and allocations; (ii) the study of market dynamics (equilibration and stability); (iii) the development of econometric tools to analyze data from experimental markets; (iv) the development of flexible software to deploy platform-independent web-based markets; and (v) the exploration of the neuro-scientific foundations of perception of financial risks.
The purpose of experimental finance is to provide a solid scientific foundation for finance, from the individual (through behavioral and neuro-scientific observation) to the market (through price and volume recordings). While attention is paid to ecological relevance (are the risks of a type that most subjects would be familiar with?), the experimental settings reflect the level of abstraction that is needed to fruitfully study theory. For instance, many of the frictions such as taxes and transaction costs that are commonly found in the field are absent in my markets — just like they are absent in the most fundamental versions of the theory.
This approach has generated important contributions. My collaborators and I were the first to ever verify the basics of asset pricing theory, while at the same time uncovering a startling puzzle: while prices quickly converge to the right configuration, allocations do not. Follow-up experiments demonstrated that while the theory predicted only a tiny fraction of individual choices, it explained the very fraction that correlated across individuals, and hence, the one determining prices. Intrigued by the enormous behavioral heterogeneity in our experiments, we started a promising research program to understand the impact onto prices and allocations of cognitive biases as they naturally occur among large groups of individuals. Among other things, we are finding that market behavior (prices) does not equal the sum of its components (valuation by individuals).
With our experiments, we can answer questions that cannot possibly be answered with field data, such as: Do markets equilibrate? How do they equilibrate? Field studies interpret the data through the lens of equilibrium theory and cannot independently verify the appropriateness of this lens for lack of information. Unlike in physics, (market) equilibrium cannot be defended on more fundamental principles, and as such remains a concept that is to be verified empirically. In our experiments, we detected some striking regularities in the dynamics, which led to a formulation of a new theory of market equilibration, as well as follow-up experiments.
Recently, neuroscientists have involved me in research on the mechanics of risk perception in the human brain. From my point of view, the goal of this research is twofold: (i) to improve the positive and normative relevance of the decision theory underlying the theory of finance; (ii) to determine when and how humans change their perception of risk and reward in a — from an evolutionary point of view very novel — financial environment. Among other things, we have discovered that the brain decomposes gambles into expected return and risk, very much like in the modern theory of portfolio choice.