Thursday, December 20, 2018

AQR's Cliff Asness: "Once More, Without Feeling "

We'll use Mr. Asness' tweet as the intro:
From AQR, Dec. 19:
I know that, unfortunately, a large fraction of my blogs fall into the category of “viewed properly this isn’t that big a deal.” This one, about this year’s overall U.S. stock market return, is another. Also unfortunate, is how much we need commentary like this (not just mine!). This world is pretty much designed to convince us that we’re always at DEFCON 1, when 5 is the mode and 4.5 the mean. 1

There are some interesting, and perhaps even actually extreme, things going on this year (e.g., breadth of losses across many investments and investment styles). But the overall U.S. stock market is not one of them. However, if you, for instance, watch cable business news too much and absorb the headlines (e.g., “it’s the worst December since the Great Depression!”, “did you see the POINT drop!”) you’d be forgiven for thinking the S&P 500 is on a death march. So, while perhaps repetitive in theme and pedantic, those of us repeatedly pointing out the near constant exaggerations are, I hope, doing something useful.

This post is really simple and short. I’m just going to look at daily returns on the S&P 500 and compare 2018 year-to-date (YTD) to similar length periods over history. 2
 
The YTD annualized daily volatility of the S&P 500 3 as of December 17, 2018 is 16.5%. That puts it at the 66th percentile over history going back to 1928. 4 So, sure it’s above median, but if that’s your standard for harrowing, you’ve led a rather sheltered life. It’s 29% of the maximum we’ve ever seen over the same length of time and, to make sure the maximum wasn’t a freak occurrence, it’s 43% of the 95th percentile rolling volatility ever observed. That just ain’t a very big number.
If we look at actual returns, 5 2018 YTD is at the 24th percentile versus comparable history. Below average, sure, but could you imagine the blaring headline “Freakish year that only happens one quarter of the time!”

Looking at the drawdown (comparing the endpoint of December 17, 2018 to the high point within 2018), it’s at the 22nd percentile. Meaning it’s a worse drawdown than experienced over 78% of the similar length periods (using drawdowns from the peak within 50 week look-back periods here again to make it an apples-to-apples comparison). Again, that’s a mildly bad year versus history. But, again, that’s also nothing to write home about (or, frankly, to write about at all!).

This isn’t exhaustive. There may be some measure or scale that shows 2018 to be truly exceptionally difficult. 6 For instance, as mentioned above, broadening things globally, and to other asset classes like bonds and commodities, and perhaps even to known systematic strategies, likely shows this to be a more surprisingly bad year. Breadth is likely a bigger outlier than depth. 7....
...MORE