Daniel Dorn

Trading as Entertainment

Among 1,000 German brokerage clients for whom both survey responses and actual trading records are available, investors who report enjoying investing or gambling turn over their portfolio at twice the rate of their peers. Including entertainment attributes as additional explanatory variables in cross-sectional regressions of portfolio turnover on objective investor attributes more than doubles the fraction of the total variation of portfolio turnover that can be explained. The results are robust to controlling for gender and proxies for overconfidence constructed from survey responses. Nonpecuniary benefits of trading thus appear to offer a straightforward explanation of the “excessive trading puzzle.”

Trading as Gambling

This paper offers evidence from three different samples consistent with investors substituting between playing the lottery and gambling in financial markets. In the United States, increases in the jackpots of the multistate lotteries Powerball and Mega Millions are associated with significant reductions in small trade participation in the stock market. California-based discount brokerage clients and German discount brokerage clients are significantly less likely to trade during weeks with larger lottery prizes in the California and German lotteries, respectively. Variation in lottery prizes affects speculative trading in more lottery-like securities such as individual stocks and options, but not trading in bonds and mutual funds. Trading that is likely associated with long-term savings motives, such as trading in retirement accounts, does not respond to lottery jackpots, either. The negative relation between trading activity and jackpots is stronger for individuals who are more likely to play the lottery. This paper was accepted by Brad Barber, finance.

Investors with Too Many Options

During the last decade, markets for covered warrants (bank-issued options) have flourished in Europe and Asia. In these markets, investors often face a choice between many instruments that differ only slightly from each other. Based on retail trades in call options on the German DAX index, this paper documents substantial price dispersion across securities that are close substitutes. Moreover, investors generally fail to identify attractively priced options. The results suggest that the observed product proliferation imposes a substantial search cost on investors even though the products are homogenous and their pricing is well understood. The search cost is estimated to average 1% of the amount invested, the same order of magnitude as the average spread. JEL Classification: G11, G13, D83

Rational Disposition Effects

The paper examines the tendency to sell winners and hold on to losers in a dynamic noisy rational expectations equilibrium with informed and uninformed investors. The key feature of the model is that the information asymmetry between investors varies over time. Besides demonstrating that the disposition effect is not intrinsically at odds with rational behavior, the model makes two novel predictions. First, disposition effects among uninformed investors should weaken after events that reduce information asymmetry. Second, disposition effects among uninformed investors should be weaker in persistent winners and persistent losers. The data, transactions of 30,000 clients at a German broker between 1995 and 2000, are consistent with these predictions.

Student-Managed Portfolios: Wisdom of Independent Crowds?

This study uses a student-managed portfolio as a novel setting to examine the effect of crowd wisdom on investment decisions. Crowd wisdom is measured by peer voting at Drexel University’s Dragon Fund, where student managers vote to accept or reject each other’s recommendations. Based on a sample of 202 stock pitches between 2008 and 2016, the main finding is that accepted stocks significantly outperform rejected stocks. For example, accepted stocks returned 3.0% in excess of their corresponding sector benchmark during the 12 weeks following the recommendation, significantly higher than the negative 2.9% excess returns of rejected stocks. The results are robust to outliers and different evaluation horizons, are not due to size and book-to-market tilts, and cannot be explained by professional observers whose expertise might influence student votes.