Past studies into behaviors influencing asset bubbles have focused on single-horizon assets and investors who trade over a common period. The authors test how overlapping sets of investors who enter and exit the same asset market at different times affect price levels. Controlling for several known conditions that are associated with asset bubbles in the literature, the authors show how new cash and inexperienced investors entering the market can inflate prices higher than their fundamental values. This result occurs despite training and complete market transparency. More experienced generations of investors better account for fundamental value.

Previous researchers analyzing bubbles in laboratory settings have uncovered several insights. The most notable are that experience over three trading sessions reduces irrational pricing (as long as fundamental value is reasonably discoverable); that relegating dividends to a single, period-end payout reduces the impact of new liquidity on sentiment; that futures markets provide useful information about true value; and that bubbles seem to be driven by high cash-to-asset ratios, lack of futures market information, and inexperienced traders.

The authors incorporate these insights into their experimental design and focus their research on a few factors they choose to isolate. Future researchers could develop similar targeted studies that merely adjust one or more of the parameters of interest.

According to the authors, even when traders know the true fundamental value of an asset, such factors as inexperience and additional liquidity cause prices to exceed that value under certain circumstances. They further demonstrate that bubbles can recur for the same asset as new traders enter the market. Prices spike when each new generation begins trading and decline when a prior generation exits, resulting in a clear two-bubble pattern.

Although training can help reduce behavioral biases, only real experience seems to enable investors to avoid irrationality mistakes. In fact, the results show that participants tend to “anchor” their price forecasts in previously observed prices even when those forecasts are not tied to fundamental value. The more experienced generations, however, rely more heavily on fundamental value when estimating prices.

The authors model their study on previous research by assuming an asset with finite future dividend payments and constantly declining fundamental value. Their study is unique because it allows “generations” of investors to enter and exit the market in an overlapping fashion. To reduce known sources of bias, the authors train subjects in assessing fundamental value and check for comprehension. During the study, the subjects see the full order book representing the market and keep an eye on the current fundamental value graphically. Accurate price estimates between trading periods result in financial rewards. The authors allow dividend payments throughout the study so that cash flows can affect liquidity.

The study includes four iterations of 18 participants, who are separated into three “generations” of 6 participants each. The first generation receives cash and shares, but the second and third generations receive only cash to increase the market’s cash-to-share ratio. The three generations trade across 25 periods and overlap for two ranges of periods in the middle of the study. Thus, each generation faces a different problem of how to value assets (e.g., the third generation cannot sell to a new generation and knows the asset will only be worth the value of the final dividend when the study ends).

This study does not perfectly capture how broad, liquid markets actually operate. Most inexperienced traders probably do not bring sufficient liquidity to the market to affect a publicly traded asset’s market price significantly. Traders do not usually have the benefit of a fully transparent trading book and a chart showing true fundamental value. Real asset prices incorporate growth expectations and sentiment as well as discounted cash flows, so fundamental value remains somewhat subjective.

Nevertheless, the authors highlight how individual investors behave irrationally, how new liquidity can encourage overpaying for assets, and how trading experience helps reduce suboptimal investment decisions.