Spitznagel: Not All Risk Mitigation Is Created Equal

[Daal]

He talks about using options to raise the CAGR of a portfolio, despite the fact that the options returns themselves have a 0% average return (no risk premium for buyers)

http://www.zerohedge.com/news/2017-11-21/spitznagel-warns-not-all-risk-mitigation-created-equal

"True risk mitigation shouldn't require financial engineering and leverage in order to both lower risk and raise CAGRs. After all, lower risk and higher CAGRs should go hand in hand! It is well known that steep portfolio losses crush long-run CAGRs. It just takes too long to recover from a much lower starting point: lose 50% and you need to make 100% to get back to even. I call this cost that transforms, in this case, a portfolio's +25% average arithmetic return into a 0% CAGR (and hence leaves the portfolio with zero profit) the "volatility tax:" it is a hidden, deceptive fee levied on investors by the negative compounding of the markets' swings. (The destructiveness of the volatility tax to a portfolio explains in a nutshell Warren Buffett's cardinal rule — "don't lose money.")."

 

[tommcginnis]

http://www.pionline.com/article/201...tary-not-all-risk-mitigation-is-created-equal

Same article, but without the need-to-shower that accompanies any visit to zerohedge.

(Snarky snark snark.)

Thanks for a great post/citation.

 

[truetype]

Surprise! The punchline is: "invest with me, not my competitors."

 

[sle]

Surprise! The punchline is: "invest with me, not my competitors."

Yeah, what else could it be?

The whole “volatility tax” idea is kinda interesting. Conceptually, if you would believe that 100% of any stock variance is explained by the general market you should be able to short high beta stocks, go long low beta stocks and lock in the excess mean reversion that occurs during the volatile times. In practice, you find that you bleed more on idiosyncratic moves and that beta of high vol is not that high.

 

[sss12]

Yeah, what else could it be?

The whole “volatility tax” idea is kinda interesting. Conceptually, if you would believe that 100% of any stock variance is explained by the general market you should be able to short high beta stocks, go long low beta stocks and lock in the excess mean reversion that occurs during the volatile times. In practice, you find that you bleed more on idiosyncratic moves and that beta of high vol is not that high.

@sle in addition to your point, how do you keep the delta of the different positions in check while waiting for this reversion/volatility ? I guess some could delta hedge but that is way above my pay grade.

 

[sle]

@sle in addition to your point, how do you keep the delta of the different positions in check while waiting for this reversion/volatility ? I guess some could delta hedge but that is way above my pay grade.

Well, in an ideal world where all of the stock performance is explained by the broad market, you would be steadily rebalancing this portfolio.

 

[Daal]

The whole “volatility tax” idea is kinda interesting

I believe Spitznagel is being a little misleading here. He finds that lower volatility improves geometric returns but he is not measuring actual dollars earned (which is more of a function of arithmetic returns). Its possible to have improved CAGRs without a single extra dollar earned.

This paper explains it well in the context of risk reduction from diversification
http://www.bfjlaward.com/pdf/25968/65-76_Chambers_JPM_0719.pdf

This portfolio has the same expected return (arithmetic mean) but varying geometric means (CAGRs) depending how much volatility it has:

SDs drop and Geometric returns improve as you add diversification, but yet, the investor won't make a single extra cent from this. This is not to say that he will be making a mistake by diversifying but to talk about improved CAGRs while you are making less money "What at first appears to gratuitously lower the arithmetic return of the portfolio (and drag on the portfolio as a line item in 9 out of 10 years) turns out to be a CAGR boon." seems ridiculous.

What I think he is trying to get at is that there is an improvement in risk adjusted returns, which is likely to be true. But that improvement should not be measured by the CAGR, this gives a FALSE impression that returns will improve, which they likely won't. It should be measured by metrics like the Sortino, MAR or Calmar ratios. But given the Spitz is in the hedge/fee business, its just more appealing to him to give that false impression of return boosting from options buying

Anyone got any comments on whether am I wrong on this?

 

[globalarbtrader]

I believe Spitznagel is being a little misleading here. He finds that lower volatility improves geometric returns but he is not measuring actual dollars earned (which is more of a function of arithmetic returns). Its possible to have improved CAGRs without a single extra dollar earned.

This paper explains it well in the context of risk reduction from diversification
http://www.bfjlaward.com/pdf/25968/65-76_Chambers_JPM_0719.pdf

This portfolio has the same expected return (arithmetic mean) but varying geometric means (CAGRs) depending how much volatility it has:

View attachment 179762

SDs drop and Geometric returns improve as you add diversification, but yet, the investor won't make a single extra cent from this. This is not to say that he will be making a mistake by diversifying but to talk about improved CAGRs while you are making less money "What at first appears to gratuitously lower the arithmetic return of the portfolio (and drag on the portfolio as a line item in 9 out of 10 years) turns out to be a CAGR boon." seems ridiculous.

What I think he is trying to get at is that there is an improvement in risk adjusted returns, which is likely to be true. But that improvement should not be measured by the CAGR, this gives a FALSE impression that returns will improve, which they likely won't. It should be measured by metrics like the Sortino, MAR or Calmar ratios. But given the Spitz is in the hedge/fee business, its just more appealing to him to give that false impression of return boosting from options buying

Anyone got any comments on whether am I wrong on this?

I wrote about this on my blog here.

Short answer is I think it does make sense to use CAGR / geometric returns if the target is your median expected wealth at the end of the investing period. So you have to be thinking in terms of probability expectations and also think that using the median is the best kind of average. As I talked about in my second book thinking like this means you'll end up allocating slightly less to equities than you would if you were just maximising arithmetic mean.

However on the specifics of this article I'm not sure it's safe to assume that options buying is net zero cost (in arithmetic return space). Although I think it's plausible to do financial engineering that lowers the arithmetic return but raises the geometric return, I'm not sure there exists 'zero cost' crash insurance - the market makes you pay dearly for buying OTM puts, and even smarter tail hedging funds should expect to cost you (arithmetically) over time.

Trend following managed futures arguably come closest to this ideal (because they buy the optionality synthetically so don't pay the volatility premium) and may even have a slightly positive return over time, but unlike OTM puts they aren't guaranteed to provide perfect crash protection.

GAT

 

[srinir]

I wrote about this on my blog here.

Trend following managed futures arguably come closest to this ideal (because they buy the optionality synthetically so don't pay the volatility premium) and may even have a slightly positive return over time, but unlike OTM puts they aren't guaranteed to provide perfect crash protection.

GAT

I think trend following as a diversifier over the long term, but during event risk it might even hurt investors big time if they are positioned in long risk assets and short bonds.

Just off topic: It was a good blog article. You mentioned in that article Sharpe Ratio is calculated using geometric mean. It should be calculated by using arithmetic mean.

Sharpe himself uses arithmetic average
https://web.stanford.edu/~wfsharpe/art/sr/SR.htm#fn1
another from M*
https://gladmainnew.morningstar.com/directhelp/Methodology_StDev_Sharpe.pdf

@Daal Thanks for the link. It was a very good article, clarified somethings for me.

 

[sle]

However on the specifics of this article I'm not sure it's safe to assume that options buying is net zero cost (in arithmetic return space). Although I think it's plausible to do financial engineering that lowers the arithmetic return but raises the geometric return, I'm not sure there exists 'zero cost' crash insurance - the market makes you pay dearly for buying OTM puts, and even smarter tail hedging funds should expect to cost you (arithmetically) over time.

Tail funds usually make one out of two premises - either that some form of tail risk premium is underpriced or that you can find relative risk premium combination that provides them with tail protection while paying little to no carry. My experience is that the former is usually a complete fiction whipped up to raise money for said funds (heavy bleeding is almost always the result). The latter is hard because tail events are rare by definition, so whatever risk premium combination you come up with might not perform as well in the next crisis.