FOOL ON THE HILL: An Investment Opinion
A couple of years ago, I reported on a large study that offered compelling evidence that individual investors can beat the market -- as long as they adopt a buy-and-hold approach. Undertaken by Brad Barber and Terrance Odean, business school professors at the University of California at Davis, that study showed how frequent trading simply kills investment performance, even before figuring in trading expenses and taxes. One reason is that investors tend to be overconfident, thinking they're smarter than they really are. The stocks they sell generally end up doing better than the stocks they buy as replacements.
That Barber/Odean study looked at trading activity in 78,000 discount brokerage accounts between February 1991 and December 1996. But a lot has changed about the brokerage industry in the last few years. Online brokers have now made it easier than ever before for investors to churn their entire portfolio with just a few mouse clicks. At the same time, though, commissions have fallen, as have trading costs associated with the spreads (the difference between the bid and the ask price). Moreover, the Web has offered individuals greatly improved access to investment information that's presumably made investors smarter. So one lingering question was whether the professors' work really spoke to the new investment scene.
The answer seems to be yes, a big yes based on a follow-up paper from the very same Barber and Odean. This new study deals with a subset of online investors who were part of the original dataset. So again, it's a somewhat dated sampling. Still, the most reasonable extrapolation from this work is that online stock trading is particularly perilous because it encourages exactly the kind of active trading that produces poor investment returns. Indeed, Barber and Odean show that merely switching from telephone-based trading through a broker to online trading via a computer can turn very good investors into bad investors. So this is a cautionary tale for just about everybody reading this.
The latest paper focuses on 1,607 individuals who switched from phone trading to online trading between 1991 and 1996. The professors match this pool of adventurers with a group of investors who had similar-sized accounts but who kept trading by phone. The folks who adopted online trading were more likely to be younger men with higher income and net worth. Also, some 80% of these mavericks reported having good or extensive experience compared to just 64% of the size-matched group.
These data points alone would suggest that the online traders were better, more sophisticated investors overall. And their performance record bears that out. Before going online, the average member of this target group produced gross returns that beat the market by 4.2% a year while outperforming their matched peers by 2.2% a year. Net of trading expenses, they still outpaced the market by 2.4% annually while besting their peers by 1.7%. The "market," for the purposes of this study, was a value-weighted index of stocks on the NYSE, Nasdaq, and American Exchange.
Online trading was perceived as an even riskier activity during the study period than it seems today. So it makes sense that it would have attracted a stronger crop of investors, folks who were confident enough to take more active control of their money and eager to reduce their commission costs. These investors had good reason to attribute their excellent performance to their own smarts rather than to pure luck. Moreover, online trading systems offered them access to more information while conveying an increased sense of control. Surely, it would make them even better.
However, more information does not necessarily improve an investor's ability to accurately forecast the market. And directly tapping in a trade rather than calling a broker doesn't by itself improve the outcome, though it may feel like it will. If anything, these advantages of online trading may simply foster illusions that feed an investor's overconfidence. These pioneer online traders, then, were particularly susceptible to trouble.
And trouble soon found this group of already super-confident investors. After going online, they traded "more actively, more speculatively, and less profitably" than they did before going online, according to the new study.
Before switching to online trading, the focus group averaged 70% annualized turnover of the stocks in their portfolio compared to 50% for the control group. (For comparison, the average mutual fund today turns over about 77% of its portfolio annually.) Yet, these online traders' turnover surged to a 120% annualized rate during the first month after switching. It eventually settled into a still remarkably high 96% annual churn rate. Most of these investors increased their trading activity after going online. However, this dramatic surge was driven by a small group (10% of the total) who proved particularly hyperactive traders, boosting their portfolio turnover by at least 109%.
Some 60% of the increased trading came from what Barber and Odean define as "speculative trading," which doubled for investors after they went online. Ironically, these pioneers had been relatively good speculators before switching to online accounts; they became comparatively lame speculators after doing so.
The bottom line according to Barber and Odean is that "online investors perform poorly" after making the switch to online trading. While their gross returns underperformed the market by an annualized 1.2%, their net returns after trading costs were far worse, underperforming the market by 3.5% a year. And this for a group that had been walloping the market!
The underlying explanation for these odd results is that these online investors' supercharged performance prior to going online had little if anything to do with their trading acumen. Barber and Odean track the returns these investors would have generated had they simply held for an entire year those portfolios they began the year with. They compare these theoretical buy-and-hold returns to these investors' actively traded accounts before going online. The professors found that the gross returns were essentially the same.
That is, stock selection alone accounted for these investors' market-beating returns before making the switch. More active online trading, then, only proved detrimental, leaving these investors with annualized returns 4% below what they would have attained by simply holding on to their original portfolios. As the professors conclude, "Trigger-happy traders are prone to shooting themselves in the foot."
Because it focuses on relatively early adopters, it's not clear how well the UC-Davis study applies to investors currently switching to online brokerage accounts. Overall, these newbies are probably less sophisticated investors, and so they may suffer from overconfidence, but with even less justification. Then again, they might suffer from less overconfidence because they have fewer reasons for believing they can easily beat the market. Also, as the popularity of online investing increases, it pulls in more of those relatively conservative investors that might have been found in Barber and Odean's control group.
There's also the chance that the net returns for frequent traders would look somewhat better today due to lower trading costs. In the Barber/Odean study from last year, the professors calculated that average round-trip commissions were 5% of the principal invested while the spread amounted to 1% of principal. So overall trading costs were very high. The new study figures that spreads fell from 1.13% to 0.86% for the online traders while commissions dropped from 3.27% to 2.51%. So overall, online trading costs amounted to 3.37% of principal. Commissions for most online investors today are probably somewhat lower, though they have now generally reached a plateau.
In any case, it's easy to be smug about how the Web has transformed investing for the good. Individual investors now enjoy nearly instant access to information once available only to professionals. Like-minded investors can share ideas via online message boards. Average people can trade for themselves, and at a fraction of what the "full-service" brokers used to charge. The Web has empowered us all to take greater, potentially better control of our finances.
Yet, the work of Barber and Odean suggests there's probably still a gap between the vast potential and the reality. Given that Forrester Research projects that there will be over 20 million online accounts controlling more than $3 trillion by 2003, there's good reason for us to want to close that gap. And the best way is simply to encourage investors to be owners rather than traders, to invest for the long term rather than the afternoon.