An anti-ETF blog post by Terry Smith of Tullet Prebon got a lot of attention Friday from several MSM sites. While it might make some investors think twice about using the product, I think this really needs to be dissected to be put in its proper perspective and utility.
Point number one was that "some ETFs do not hold physical assets of the sort they seek to track. They are 'synthetic' and hold derivatives." Smith is based in London, and this is more prevalent in ETFs listed in Europe. There are, of course, ETFs in the US that use derivatives, and so obviously face counter-party and other risks mentioned by Smith.
It does not take a lot expertise to understand that the more complicated a product is under the hood, the more risks there are that have nothing to do with the exposure being sought. If ProShares were to have some sort of problem with its counter parties, then SPXU is going to have a problem regardless of what the S&P 500 is doing. This should not be news to anyone, and if this possibility is unacceptable to you, then you would simply avoid the funds.
Smith's next issue was "ETFs do NOT always match the underlying in the way people expect. Because of daily rebalancing and compounding, you can own a leveraged long ETF and lose money over [a] period when the market goes up but during which there are some sharp falls." We covered this here many times when this was a front-burner issue in late 2008.
My own observation is that the narrower the segment covered by an inverse or levered fund, the less predictable the fund will be. The broader levered and inverse funds have "worked" over some longer periods of time. Look at a YTD chart comparing client holding SDS and SPY. It worked most of this year. It may have stopped working for the last month or so (you can decide for yourself). This all serves to tell us what we should already know, in the future it will work over longer terms, except for the times that it doesn't.
Seriously, there is no way to know whether these will work over the longer term; it all depends on the combination of up and down days. This should not be news, and like the point above, anyone for whom this is unacceptable would simply not use these types of products. The reason we use it (in moderation) is that if the stock market goes down a lot today, then I have full faith that SDS will go up twice the amount of the decline.
Number three was the old chestnut about short interest being larger than the number of authorized shares. The example from before was the SPDR Retail ETF (XRT). IndexUniverse addressed this worry very effectively; only authorized participants can create and redeem shares. Massive short positions have to be closed by trading out of them.
Smith's last point was that fund companies make money trading the derivatives that underlie the funds. If we broaden this to include stocks (which is, of course, incorrect, as the fund companies don't create shares, APs do that) then the issue becomes whether the fund does a good enough job tracking the stock basket it is supposed to track. You obviously need to decide that for yourself.
One point I have made repeatedly about ETFs is that they are merely access. It is up to the end user to decide what the best access is for them for each exposure they want in their portfolio. It is not logical that an ETF can be the best tool for all times and for every single exposure. And for some segments, there are no ETFs. Someone wanting Canadian banks or Scandinavian banks must use individual stocks, unless they want very diluted exposure from a broader fund.
ETFs have flaws, as do individual stocks, as do broad-based funds. A huge building block for portfolio success is understanding the potential drawbacks of what you own.