The Exchange Traded Fund (ETF) market has exploded in recent years to total over $600bn in the US alone, offering both long and short exposure to most asset classes. It is now used by institutional as much as individual investors to hedge risk and rapidly change portfolio exposure. They are a very useful tool to create a sophisticated and flexible portfolio at low cost, but I've often warned of their significant tracking error, particularly for inverse and leveraged ETFs, as many investors have discovered to their cost this year. Short ETFs have a non-linear relationship to their index; as an ETF loses value, its exposure also decreases. The reason for this curved decay line, is that short ETFs, unlike their underlying indices, are rebalanced on a daily basis. At the most basic level, the return of a leveraged fund is path dependent so that a rise and subsequent fall in the underlying market generates a radically different return to a fall and subsequent rise over the same period, even if the underlying asset move is the same in both cases. Given the path dependance, the more volatile the return pattern over time, the higher the ETF's tracking error. (A second issue is the rollover costs associated with popular 'vanilla' commodity funds like USO or UNG and their relative size to the physical market,  producing wide tracking error to the underlying futures)

The charts below simulate this effect to demonstrate the relationship between asset volatility and tracking error. Leveraged ETFs seek returns that are between +300% and -300% of the return of an index or other benchmark (target) for a single day only. Due to the compounding of daily returns, returns over periods other than one day will likely differ in amount and possibly direction from the target return for the same period. Many retail investors have misunderstood the interplay between correlation and leverage in these products, which are frankly suitable only for aggressive day trading strategies, and that's not a game I'd feel confident playing. At 14% monthly volatility, returns are tightly grouped....

But at say 40%, the dispersion of ETF returns versus underlying explodes...

Leveraged funds experienced net outflows of $1.85 billion in July, according to investment-research firm Morningstar. This comes after they took in a net $11.4 billion through the first half of the year and accounted for almost 30% of the total ETF inflows in some months. The Financial Industry Regulatory Authority (FINRA) in June issued a notice reminding brokers and advisers that the instruments are complex and typically unsuitable for individual investors who plan to hold them longer than one trading session. Some brokers like UBS have placed constraints on sales of these ETFs or stopped them outright in the wake of that notice, and not before time. There are a variety of low- risk option strategies available to hedge long risk or speculate on extreme market movements, with finite and transparent downside. While plain vanilla sector and country ETFs remain a core part of any portfolio (although commodity funds should be used with care when steep contango applies), inverse leveraged funds should have no part at all, ever.