by Michael Johnston on January 13, 2010 | Updated August 26, 2013
At the height of the leveraged ETF controversy in recent years has been the performance of these products when held for extended periods of time. Because leveraged ETFs operate with a daily investment objective, their returns over any period longer than a single trading session depends not only on the change in the related index, but on the path taken by that index over the period in question. If the S&P 500 increases by 2% for the month, there is no guarantee that the ProShares Ultra S&P 500 (SSO) will be up 4% over that time.
In order to amplify daily results, leveraged ETFs “reset” their exposure on a daily basis. If a bear 3x fund loses 3% in a day (i.e. the underlying index loses 1%), the exposure of the fund must be adjusted to maintain 3x daily leverage. It works the same way on the upside: if a 3x bull fund jumps 3% on the day, the fund must add additional exposure so that it can still provide the promised leverage the next day.
So it becomes clear that oscillating markets — where winning sessions are followed by losing sessions and vice versa — can erode the returns to leveraged ETFs over multiple sessions. In such markets, leveraged ETFs increase their exposure before a decline and decrease exposure before a gain. In 2008, financial markets experienced unprecedented volatility, leading to significant return erosion in leveraged ETFs.
Other Side Of The Coin
It is also important to note that daily compounding of returns can work for investors in certain markets. And it often does. Because exposure is increased after a winning session (assuming a bull fund), a streak of consecutive gains in a leveraged ETFs can begin to amplify results further.
This concept isn’t specific to leveraged ETFs by any means: it applies to everything from a savings account to a long only ETF product. If a stock gains 10% in two consecutive sessions, its cumulative increase in value over that period is not 20%, but 21%. Alternatively, investing in a leveraged ETF over time can be
Consider the following example in which an index rises 2% per day for ten consecutive sessions: