On the surface, stop-loss and stop-limit orders seem like great ideas. All one needs to do is simply place the protective order, and if the market ever falls to the price indicated on the stop-loss or stop-limit, the order is executed and that will be that.
That theory conflicts with the reality of how these orders play out, though, and ignores temporary market “failures” like the ones that occurred on Aug. 24 or the flash crash on May 6, 2010. [Ed. Note: In a possible response to the Aug. 24 flash crash, NYSE announced in mid-November that it would eliminate stop and good till canceled orders in February 2016.]
Forgetting market mechanics and function for just a moment, both crashes came and went in a matter of minutes — each literally less than one hour. Markets whooshed down, countless stop orders were executed, and then markets whooshed right back up, arguably resulting in at least a semi-permanent impairment of capital.
While stop-loss orders were intended to offer a measure of protection, they in fact made it worse on both occasions. The automated nature of these orders cannot recognize these types of events for what they are, and these events do not represent fundamental reasons to sell. For example, a stock that closed two consecutive days at $60, but spent an hour in between trading below $50 due to some perceived malfunction in the market, should not be sold.
Such a scenario can be avoided when market participants take a more active role in overseeing a portfolio. This is not about active management versus passive management, but rather about more active engagement with the portfolio regardless of an investing philosophy.
Other less dramatic drawbacks with stop-loss orders include what happens after the stop order is elected. Some practitioners may advocate for using a 10% buffer for stop orders, but the point at which the order executes could be the low for the stock before climbing much higher. An obvious flaw in 10% stop orders is that a 10% buffer for a lottery-ticket biotech stock is much different from one for a mega-cap soda company. Arguably, a biotech stock would be given more leeway than a mega-cap consumer staples company.
Another drawback is that the stock in question is likely the top choice for a given exposure. Why should that top choice be substituted for one that is only a second or third choice?
Market mechanics and the shockingly large, yet temporary, drawdowns during such extreme events show that ETFs will trade even with imperfect information. In this context, imperfect information includes fast market trading conditions, stocks that could be mispriced or, in the case of the Aug. 24 drawdown, index constituents that did not open in a timely fashion. If this is going to remain an issue, it is likely more susceptible with smaller cap stocks, which helps illustrate why small-cap ETFs experienced especially large declines on that day.
These issues matter, and they impact ETF prices because they are connected with market-making activity that allows the market to function. If underlying assets cannot be properly priced for any of the aforementioned reasons, then the likelihood increases that ETFs will not be properly priced either. This is all the more reason to not use stop-loss orders for ETFs in addition to individual issues.
Advisors should not necessarily become buyers during these temporary panics. Regardless of which side of the market an ETF trade is on, the market can price on imperfect information that potentially creates surprises in both directions. It's unlikely that the typical advisor's strategy for clients includes speculating on such uncommon and extreme market behavior. Furthermore, August's event reinforces the importance of using limit orders whenever placing an ETF trade.
--- Read "ETFs Are Not Just for Trading" on ThinkAdvisor.