The key letter in an ETF is the F. It’s a Fund, with exactly the same meaning of the word as applied to a mutual fund. It’s an allocation to a basket of securities with some sort of common attribute or factor that you want represented in your overall portfolio, not a fractional piece of an asset that you want to directly own. Yes, unlike a mutual fund you CAN buy and sell an ETF just like a single name stock, but that doesn’t mean you SHOULD. Like so many things in our modern world, the exchange traded nature of the ETF is a benefit for the few (Market Makers and The Sell Side) that has been sold falsely as a benefit for the many (Investors). It’s not a benefit for Investors. On the contrary, it’s a detriment. Investors who would never in a million years consider trading in and out of a mutual fund do it all the time with an exchange traded fund, and as a result their thoughtful ETF allocation becomes just another chip in the stock market casino. This isn’t a feature. It’s a bug.

What we saw last Monday morning was a specific manifestation of the behavioral fallacy of a category error, one that cost a lot of Investors a lot of money. Investors routinely put stop-loss orders on their ETFs. Why? Because … you know, this is what Great Investors do. They let their winners run and they limit their losses. Everyone knows this. It’s part of our accepted mythology, the Common Knowledge of investing. But here’s the truth. If you’re an Investor with a capital I (as opposed to a

Trader with a capital T), there’s no good reason to put a stop-loss on an ETF or any other allocation instrument. I know. Crazy. And I’m sure I’ll get 100 irate unsubscribe notices from true-believing Investors for this heresy. So be it.

Think of it this way … what is the meaning of an allocation? Answer: it’s a return stream with a certain set of qualities that for whatever reason – maybe diversification, maybe sheer greed, maybe something else – you believe that your portfolio should possess. Now ask yourself this: what does price have to do with this meaning of an allocation? Answer: very little, at least in and of itself. Are those return stream qualities that you prize in your portfolio significantly altered just because the per-share price of a representation of this return stream is now just below some arbitrary price line that you set? Of course not. More generally, those return stream qualities can only be understood … should only be understood … in the context of what else is in your portfolio. I’m not saying that the price of this desired return stream means nothing. I’m saying that it means nothing in and of itself. An allocation has contingent meaning, not absolute meaning, and it should be evaluated on its relative merits, including price. There’s nothing contingent about a stop-loss order. It’s entirely specific to that security … I want it at this price and I don’t want it at that price, and that’s not the right way to think about an allocation

Eoin Treacy's view

The whole question of stops has hit a nerve as market participants assess what went wrong with their strategies on August 24th. Perhaps the greatest challenge in interacting with any market is understanding your own personality before taking any action. Many people find the experience of buying comparatively easy and for a whole range of reasons the same people tend to find selling harder. It really is time well spent deciding what would need to happen to prove your hypothesis wrong that would justify selling before purchasing any instrument. This is as true of ETFs as it is of an individual stock, bond or commodity. 

For an investor I believe it is important to ask is whether one is placing a stop to protect against downside risk or because the decision to sell is difficult and is being outsourced to a stop strategy. The tax consequences alone, of selling an investment position that may have been held for a considerable time, are enough to give one pause about using stops. However that does not mean we should eschew scripting what an ending might look like and remaining vigilant for top formation development characteristics. 

If you are an investor and the ETF represents an allocation which fulfils a purpose within an overall portfolio strategy such as decreasing overall risk, improving beta or even generating alpha it really is questionable whether one should have a stop. This also assumes that an investor is unleveraged and willing to sit through the occasional medium-term correction which could see the portfolio fall in value by 20% or more. Rationally that is to be expected in the course of a lengthy investment time horizon.     

This article from the Wall Street Journal highlights the extent to which retail investors are engaging with leveraged ETFs; exemplifying just how much of a casino mentality exists within what is a highly diversified ETF market. If knowing yourself is the first step then knowing the product you are buying has to be the second. Every individual market has idiosyncrasies that relate to asset class but also the consistency of trends. For example one can expect more volatility in some instruments than others. The problem with ETFs is that the range of products is so varied that there is a temptation to lump them all together. This is particularly true for less experienced participants. Additionally, the practice of reading prospectuses has never been very popular and appears to be even less appreciated as more people are exposed to the ease of trading electronically.  

The best time to consider selling or reducing a position is before a large downdraft not after such an event. Whether one decides to use a stop to sell or we use a mental note to action the decision once a set of conditions are met will be dependent on the complexion of the market. Considering the potential for extreme volatility in ETFs the wisdom of expecting stops to perform as expected has to be questioned. As discussed earlier this week (August 28th, August 31st), using an option strategy or a buy order strategy to benefit from occasional bouts of extreme volatility is a viable course of action.