In recent years, exchange traded funds (ETF) have grown massively popular, and for many good reasons. ETFs have some attractive characteristics and provide individual investors an easy way to gain sector or broad market exposure. However, ETFs don't always work the way you may expect judging by their names. Many people who look at the returns of an ETF, compared to its respective index, get confused when things don't seem to add up. Investors should know the following factors when investing using ETFs. (Learn more in Ultra ETFs: The Ultimate Investment?)
Tutorial: ETF Investing
Does Leveraging Work?
Leveraged ETFs feature some of the most glaring naming mistakes when investing over an extended period. These ETFs generally come with the names "Ultra Long" or "Ultra Short," and if you look into their descriptions they promise two to three times the returns of a respective index, which they do - sort of. Let's look at a few examples of how ETFs don't always work the way you would expect.
The Ultra S&P 500 ProShares is an ETF designed to return twice the S&P 500. Leveraged ETFs boost results, not by actually borrowing money, but by using a combination of swaps and other derivatives. However, the effect is the same, and if the S&P 500 returns 1%, the SSO should return about 2%. But let's look at an actual example. During the first half of 2009, the S&P 500 rose about 1.8%. If the SSO had worked, you would expect a 3.6% return. In reality, the SSO went down from $26.27 to $26.14. Instead of returning 3.6%, the ETF was essentially flat.
It's even more troubling when you look at the SSO along with its counterpart, the UltraShort S&P 500 ProShares which is designed to return twice the opposite of the S&P 500. Over the 12 months ending June 30, 2009, the S&P 500 was down nearly 30%. The SSO behaved pretty well and was down about 60%, as you would expect. The SDS, however, was down about 20%, when it should be expected to be up 60%! (Learn more in Dissecting Leveraged ETF Returns.)
Are SPDRs More Predictable?
Leveraged ETFs aren't the only ones with problems. S&P 500 Depository Receipts, also known as SPDRs, seek to exactly mimic the S&P 500. Over the 12 months ending June 30, 2009, SPY shares were down just under 30%, which is about right. Its counterpart, however, Short S&P 500 ProShares, which seeks to move exactly the opposite, was curiously down around 4% during the period.
So now that we've looked at a few examples of how ETFs don't always do what they are supposed to do, let's examine why. ETFs are really designed and marketed to track the daily movements of a corresponding index. You may ask yourself why that would matter, since if it tracks its index properly each day, it should work over any extended period of time. That is not the case. The compounding effects of daily returns will actually throw off the math, and can do so in a very drastic way.
For example, if the S&P 500 moves down 5%, the SSO should move down 10%. If we assume a share price of $10, the SSO should be down to about $9 after the first day. On the second day if the S&P 500 moves up 5%, over the two days the S&P 500 return will be -0.25%. An unaware investor would think the SSO should be down 0.5%. The 10% increase on day two will bring shares up from $9-9.90, and the SSO will, in reality, be down by 1%. It decreases a full four times the decline of the S&P 500. Typically, you will find that the more volatile the benchmark (S&P 500 in this example) for a leveraged ETF, the more value the ETF will lose over time, even if the benchmark ends up flat or had a 0% return at the end of the year. If the benchmark moved up and down drastically along the way, you may end up losing a significant percentage of the value of the ETF if you bought and held it.
As for the problems with normal ETFs, the math works similarly, but on a smaller scale. If you start at an index value of 100 and it drops 5% on the first day, a long ETF with a starting value of $100 will drop 5% to $95, while the short ETF also at $100 will increase 5% to $105. If on the next day, the index rebounds by increasing 5%, the short ETF will drop 5% to $99.75 [105 x (1-0.05)], and the long ETF will rise 5% to $99.75 [95 x (1+0.05)]. Both end up at the same value, and both have dropped 0.25%, when they are supposed to be inverses of one another. Taking this over more extended periods can cause even more accentuated problems. In volatile markets, like those seen in 2008 and 2009, even unleveraged ETFs can show significant discrepancies. (See Investment Strategies For Volatile Markets to learn how to adapt your strategy based on market conditions.)
We used the S&P 500, and its corresponding ETFs, as the basis for all of the examples above because they are some of the most visible and heavily traded ETFs, but there are similar ETFs designed for other indexes and sectors, where all the same rules apply. Long, unleveraged ETFs will generally behave as you would expect when comparing with its index. Levered and short ETFs, however, can look like they have significant differences, especially when an unknowing investor buys one as a long-term investment. ETFs can be useful for short-term plays, but many are not the best choice as long-term plays.
Companies that sponsor ETFs outline the issues that investors will likely experience if holding the ETF over an extended period of time. Many ETFs are generally designed for short-term (daily) plays on an index or sector, and should be used that way. Most, except long unleveraged ETFs, will not work as you may expect over a long period of time, especially in volatile markets.
For additional reading on ETFs, take a look at Mutual Fund Or ETF: Which Is Right For You?