CXO Advisory Group reviews a paper asking whether individual investors can beat the market. Contrary to the EMH, the authers find that individual investor's profits are persistent in both underpreformance and out performance:
We document strong persistence in the performance of trades of individual investors. The correlation of the risk-adjusted performance of an individual across sample periods is about 10 percent. Investors classified in the top performance decile in the first half of our sample subsequently outperform those in the bottom decile by about 8 percent per year. Strategies long in firms purchased by previously successful investors and short in firms purchased by previously unsuccessful investors earn abnormal returns of 5 basis points per day. These returns are not confined to small stocks nor to stocks in which the investors are likely to have inside information. Our results suggest that skillful individual investors exploit market inefficiencies to earn abnormal profits, above and beyond any profits available from well-known strategies based upon size, value, or momentum.
The authors ask if the data can be used to trade (Note: profits are before transaction costs):
Using the one-week holding period, the short horizon strategy earns abnormal
returns of 5 basis points per day, or 13.7 percent per annum. Again, the results are
robust to removing small stocks and trades in stocks an account has traded previously.
The strategy is also assessed using a one-day and a one-month holding period. Using
the one-day holding period, the return is 7 basis points per day, whereas the daily
return of one-month holding period is indistinguishable from zero, suggesting that
many of the profitable trades have horizons of less than a month.Our long horizon portfolio strategy mimics the portfolio holdings of previously successful traders and shorts the positions of previously unsuccessful traders. Like
the short horizon strategy, this long horizon strategy earns statistically significant
abnormal returns of 4 to 5 basis points per day. Moreover, because holding periods
are matched to those of the trades that are being mimicked, this strategy achieves
the outperformance with far less turnover.
The authors conclude by asking the obvious:
An interesting further question is whether large brokerage firms are aware of the value of the information contained in their customers’ trades.
Can ubiquitous free trading be far behind? After all, if brokerages can potentially profit from their customer's trade data, shouldn't they encourage it?
Comments