Shortly after the May 6 "flash crash" on Wall Street, our exchange-traded fund analyst team started getting e-mails from disgruntled ETF investors. Some of these investors, it turns out, had put in place stop-loss orders instructing their brokers to dump their ETF shares if the price of their securities dropped below a given level. So when the market went into freefall, these ETF investors' brokerage firms automatically jettisoned their shares. What were temporary losses for many investors turned into real losses for the unlucky people with stop losses.
My colleague John Gabriel railed against the illusory safety net of stop-loss orders in a recent article. He argued that using stop-loss orders could be a sign that an investor doesn't have sufficient conviction in his or her picks. John also noted that on highly volatile market days like May 6, technical factors could mutate the results of a stop-loss order, leading to unintended consequences.
That column prompted many useful comments. Some of you agreed with John that stop-loss orders don't make sense for most investors, while a few others noted that you had used stop-loss orders to good effect before (or wish you had).
My view is that if you're a long-term investor, it's a big mistake to get bogged down in performance-based rules that dictate when you buy and sell. For that reason, I'd argue against not just stop-loss orders but any investment discipline that uses price as a key determinant of when you buy and sell.
Trader or Investor?
Where you come down on this issue says a lot about whether you're a trader or a long-term holder of stocks. If you're the former, you might use mechanistic sell rules like stop-loss orders to help you stay abreast of what could turn out to be momentum trends unfolding in your holdings or in the market as a whole. I'm not saying that's necessarily advisable, particularly when attempting to gauge the whole market's future direction (or that of a given market segment). True, there's a body of academic research pointing to the persistence of momentum trends in the market (i.e., when performance of an asset is heading in one direction, either on the upside or downside, it tends to keep going that way for a while). But it's worth noting that very few professional investors have had consistent success with market-timing strategies of any kind--momentum or otherwise. That, in turn, casts doubt on smaller investors' ability to play the game, particularly if they have to rely on less sophisticated trading systems than many pros have at their fingertips.
And if you're a long-term investor in stocks, the use of mechanistic selling rules like stop-loss orders should be off the table. Ditto for other self-imposed selling rules, whether automated or not, that hinge on a security's price, performance, or ranking. I cringe when I hear investors (or, worse yet, some financial advisers) tout that they'll sell any fund that drops by 10% or has a category ranking in the bottom half of its group.
While such formulae help create a sense of discipline and order in a chaotic market, in reality they're a formula for poor investing results, helping ensure that an investor sells low. (Unfortunately, many investors combine such rules with similarly mechanistic "buy" formulas--for example, only considering holdings that land in the top 25% of their peer groups. That helps cinch the "buy high" side of the equation, too.)
The disclaimer isn't just boilerplate: Past performance isn't particularly predictive of how an investment is apt to behave over long periods of time. If it were, investing well would be simple. Instead, Morningstar's research into fund-performance patterns has generally found past performance to be a fairly weak predictor of how it's apt to behave in the future.
A Symptom Rather Than a Cure?
Using mechanical, performance-based sell rules, as John suggests in his piece, may be a sign that you don't have adequate conviction in your holdings. If you've done your homework on a stock or fund from a bottom-up, fundamental standpoint and nothing intrinsic has changed since then, you should be prepared to buy it hand over fist when it's going down. As a famous fund manager once said to me, "If I liked a suit when it was full-price, I should like it a lot more when it's half off."
And perhaps more important, if you feel the need to use mechanistic, performance-based selling rules to protect yourself from big losses, it could be a sign of a mismatch between your asset allocation and your time horizon. If you're holding stocks in an account where you can't afford to lose much money, you're in the wrong asset class. Stop-loss orders--or any other performance-based sell discipline--aren't the answer.
Crafting an Investment Discipline
That's not to say that investors shouldn't have a well-articulated buy and sell discipline, ideally part of their investment policy statements. They should. Having such a policy is key to ensuring that you don't get buffeted around by market winds. But buy/sell parameters in a policy statement should be centered around fundamental factors such as management changes, expense ratio increases, or, for stock investors, meaningful changes in a company's operations or its valuation. Such a discipline takes more time, and requires a more nuanced evaluation, than one that is based strictly on an investment's past performance. But it also makes better investing sense.
