Wednesday, September 12, 2012

Failure of LVOL Highlight Theoretical Flaw

Russell Investmests recently announced is shutting down its ETFs due to lack of demand: 25 funds with only $300MM in assets. A signature fund was LVOL, which actually was one of its more popular funds with $70MM, and was primed to pick up the hot low volatility investing trend.

It took a factor approach to the low volatility focus, first creating a 'volatility factor proxy', and then choosing 100-200 stocks from the Wilshire 1000 that load up on this factor. As the 'volatility factor proxy' selects stocks itself, this approach is too convoluted, and investors rightfully don't trust black boxes, which given weasel word language about momentum and beta. Their LVOL overview page is completely unhelpful, with just about every tab containing either the same pie chart of sector weightings, or regulatory boilerplate and irrelevant data. Indeed, the regulatory language clearly gets in the way of communication, highlighting that financial regulations are generally the opposite of helpful.

 In contrast, SPLV says on its home page
The Fund will invest at least 90% of its total assets in common stocks that comprise the Index. The Index is compiled, maintained and calculated by Standard & Poor's and consists of the 100 stocks from the S&P 500 Index with the lowest realized volatility over the past 12 months.
It's clear what they are doing, and they aren't trading too much (eg, monthly as with LVOL). If you spent time learning advanced financial theory, you would learn that LVOL did it right, SPLV did it wrong, because factor loadings are the measure of risk, not mere characteristics like volatility. That the whole field has a big flaw is a theme of my new book( available in paperback and Kindle!).

 Too much schooling is a bad thing. Recently I was working with my publisher, a subsidiary of Amazon, and their structure was clearly created by an MBA: I had a team with specialists who handled very narrow tasks but somehow were never the right person to comprehend any of my specific questions (eg, β should be different than b?). They all spoke as if they were reading from a script, and often I would find their unhelpful verbal answers on their website's FAQs. The fact the actual typesetters were in India, and I could never talk to these people, I'm sure made sense to the top guys at Amazon, who are traditional Harvard-educated efficiency experts. I have a feeling monks from 12th century Ireland were more accurate and faster.

 It's bad enough grad school costs too much, but it makes you less efficient.


Mercury said...

That would totally never happen in a government or insurance cartel dominated healthcare system with a vanishing small number of privately practicing doctors. Not ever.

Anonymous said...

Good Stuff. I just ordered your book on kindle and am looking forward to reading it. I will let you know what I think of it when I am done.