Jan

2

Today I attended a lunch presentation with pension funds as the target audience. They defined risk as volatility and wanted reduce risk while maintaining much of the return. It was said that buying puts reduced return too much for most fund managers. The strategy presented was to reduce equity exposure from 100% to 50% and invest 50% in a low risk asset (short term bonds), at the same time sell both OTM calls and puts. They presented a back test of 10 years where the strategy outperformed index slightly while having a lower volatility (they outperformed during the 2008 crash and vol looked to be lower all along). I'd think they expose themselves tail risk by selling OTM puts, so was surprised they outperformed during the GFC and that they came out ahead. I still think they make it 'look' good during 'normal' markets but will get killed performance wise during sufficiently high upside and downside volatility–so I really think it is somewhat of an intellectual fraud to call this a 'low risk equity exposure' for pension funds.

Alex Castaldo responds: 

I did not attend the presentation mentioned, but I am familiar with this kind of option selling strategy. One of the simplest is the PUTW (or PUTSM) strategy whose results are updated daily on the CBOE web site.

In the attached chart I compare it's total return since 1/2007 to the SPY total return (the S&P 500). Starting both strategies at an arbitrary level of 923, we see that PUTW falls to 690 in early 2009 (a 25% drop), while SPY falls much further to 503 (a 45% drop). What I find particularly interesting is how well PUTW holds up in the first half of 2008: while the stock market is going down PUTW manages to be steady or slightly up because it is selling puts at a high implied vol; only when the stock market begins to sell off very sharply after 9/30/2008 does PUTW also drop.

But even more interesting is what happens in March 2009 and after: the SPY begins to climb faster than the PUTW, slightly faster at first but markedly so after September 2012 and soon thereafter SPY passes PUTW. At the end of December 2016 SPY is at 1798 while PUTW is at 1668.

My conclusions are:

(1) The PUTW strategy has a lower volatility than SPY, both in terms of a lower drawdown in 2008 and a generally smoother path throughout the period (std dev of 11.5% per year versus 15.2% for SPY). The claims made during the presentation are believable. There is no intellectual fraud here.

(2) Everything has drawbacks as well as benefits. The drawback of PUTW is not that it will lose heavily in the future during periods of enormous volatility, but the opposite: that it will underperform during prolonged bullish periods for the market and probably over any sufficiently long period (long enough for the implied vol to adjust to whatever the situation might be and for the law of large numbers to take effect). So there is nothing magical here, as Dr. Zussman would say just the familiar tradeoff between volatility and return.

(3) The correlation between SPY and PUTW monthly returns is 0.85 (beta is 0.65) so PUTW is not all that different from SPY in terms of sources of risk (it is not a very good diversifier for stock market risk).

(4) The performance of PUTW is not theoretical or proprietary or reserved only for pension funds; it is explained on the CBOE web site (roughly speaking: sell 1 month ATM puts fully collateralized with cash) and since February 2016 there has been an ETF (also called PUTW) that implements it. So far it is small (30 million in assets). It has a 0.38% expense ratio, and so far has been tracking the CBOE version with accuracy that is quite respectable, and in line with expectations.

(5) Yes, you can reduce risk by selling Vega. Or increase risk by selling Vega, it is all in the proportions and how you do it.


Comments

Name

Email

Website

Speak your mind

Archives

Resources & Links

Search