Vineer Bhansali and Lawrence Harris have written a scholarly paper on what they call the “extraordinary growth of short volatility strategies” since late 2010.
A lot of people and institutions seem to have come to the conclusion that spiking volatility is just that. A “spike” on a chart is definitionally a quick rise followed by a quick fall. A strategist with a respectably long or even medium long term horizon can outwait the rise. Thus, taking the short position is, as a headline at Seeking Alpha said last spring, “easy money.”
Bhansali and Harris are alarmed, even expressing the view that in the event of an unexpected central bank decision or other macro event the popularity of such strategies could prove disastrous, triggering “the next serious market crash.”
These authors observe that the simplest way to sell volatility involves the sale or purchase of options contracts, but that one can also take a position on vol by trading products defined by the indexes such as VIX. Or one can do the same thing in a more complex manner by adopting a risk parity strategy or selling delta-hedged straddles or otherwise harvesting the “risk premium.”
Central Banks
Central banks, in Bhansali and Harris’ phrasing, have “infinite time horizons” and given that it is natural that they are “the largest implicit volatility sellers in the market.” After and as a result of the global financial crisis in the second half of the first decade of the millennium “they made an implicit promise through their behavior that they will provide what many consider to be a perpetual put against a rapid sell-off in the markets.”
By suppressing volatility, central banks have also protected the (other) sellers of volatility, acting as a re-insurer in this field.
But changes in central bank policy could be taken by the market to be an implicit withdrawal of the implicit promise, and that could have very destabilizing consequences.
Bhansali and Harris are concerned about this because: the assets under management in volatility-contingent strategies is large; the assets on the short side of such strategies continue to grow; confidence continues to grow; the strategy is spreading among asset classes; all short vol strategies are similar; investors are not generally aware of the extent to which their strategies are correlated; mechanized trading is common; and participants continue to short despite “declining prices and obvious risks.”
As an old saying has it: be worried when everyone is greedy. These two authors are worried that everyone is now greed and is now expressing that greed in roughly the same way.
What is important to understand, Bhansali and Harris say, is that since any other strategies follow the VIX for purposes of risk allocation, gyrations in the VIX will affect the behavior of large institutional investors. They call this a classic tail wagging the dog scenario.
Worst Case Scenario
The mechanization of trading mentioned above, the reliance upon algorithms to an extent that largely cuts humans out of the loop once a high-level strategic decision has been made, increases the probability and the probable severity of a market crash. If humans don’t have the sense to worry about where this is headed, the algorithms certainly won’t.
Bhansali and Harris have in mind a worst case scenario that they spell out thus:
- A now unknown event causes index values to drop and/or VIX to rise;
- The delta of short options positions drops – shorts sell the market index in order to restore the delta neutrality of their own positions;
- Values fall, implied volatilities rise. Investors buy puts which means that put writers sell the underlying assets to hedge;
- Mechanical strategies for which VIX is a key parameter, including risk parity, volatility targeting, or trend following strategies, automatically move to reduce exposure in a replay of the 1987 portfolio insurance scenario;
- The feedback loop expands to include the risk parity traders, who sell more.
The paper provides further bullet points along the same lines but this should be enough to make the nature of the authors’ concern clear.
The paper has been available since last November, and will appear in the Financial Analysts Journal this coming November. Bhansali is the chief investment officer of Long Tail Alpha. Harris is a Professor, Marshall School of Business, University of Southern California.