Leveraged ETFs

Recently Dr Eric Falkenstein published a post about strategy involving shorting a pair of leveraged ETFs. People observe that since both leveraged and inverse leveraged ETFs tend to underperform, shorting both is a relatively market-neutral way to capitalize on this negative drift. According to the author the strategy generates 14% return with standard deviation of 12%. Image from falkenblog:

 

Unfortunately the strategy does not quite work as well as expected - see readers comments at the end of post. The author notes "I am ignoring the short rebate, which for these may have been highly negative for some of these, but on average these have pretty meager short rates" which is just not true. The main hindrance to profit is borrowing costs - leveraged and inverse ETFs are usually in demand to short, and require a premium (negative rebate), and occasionally are impossible to locate. According to someone who was running such strategy it worked well in 08, but in 09 became flat, and was closed in 10.

While trying to understand leveraged ETFs, and particularly the underlying mathematics I came up with more questions than answers. Two excellent resources I came across is Jian (Stanley) Zhang article with Marco Avellaneda, and Eric Forgy's blog.