Does house money exacerbate price bubbles? We compare house money asset market experiments with an earned money treatment where initial portfolios are constructed from a real effort task. Bubbles occur; however, trading volumes and earnings dispersion are significantly higher with house money. We investigate the role of cognitive ability in accounting for the differences in earnings distribution across treatments by using the cognitive reflection test (CRT). Low CRT subjects earned less than high CRT subjects. Low CRT subjects were net purchasers (sellers) of shares when the price was above (below) fundamental value. The opposite was true for high CRT subjects.

Using laboratory experiments, we provide evidence on three factors influencing trader performance: fluid intelligence, cognitive reflection, and theory of mind (ToM). Fluid intelligence provides traders with computational skills necessary to draw a statistical inference. Cognitive reflection helps traders avoid behavioral biases and thereby extract signals from market orders and update their prior beliefs accordingly. ToM describes the degree to which traders correctly assess the informational content of orders. We show that cognitive reflection and ToM are complementary because traders benefit from understanding signals’ quality only if they are capable of processing these signals.

Our study provides the first causal test of classical and behavioral asset pricing models that incorporate skewness pricing. In line with these models, our experimental markets show that skewness is systematically priced. Our findings also reveal that positively skewed assets available in small supply exhibit negative expected returns, which is consistent with prospect theory, but not with expected utility models. Furthermore, in line with the mechanism underlying prospect theory, we show that the negative returns of the positively skewed asset are most pronounced during market sessions where traders overweight the low probability of receiving a large payoff.