Abstract:
Multiplicative growth processes that are subject to random shocks often have an asymmetric distribution of outcomes. In a series of incentivized laboratory experiments we show that a large majority of participants either strongly underestimate the asymmetry or ignore it completely. Participants misperceive the outcome distribution's spread to be too narrow-band and they estimate the median and the mode to lie too close to the distribution's center, failing to account for the compound nature of average growth. The observed biases are measured irrespective to risk preferences and they appear under a variety of conditions. The biases are largely consistent with a behavioral model in which geometric growth is confused with linear growth. This confusion is a possible driver of investors' difficulties with real-world financial products like leveraged ETFs and retirement savings plans.