New Volatility Regime
The equity market hasn’t really decided what it wants to do. However, we are now very clearly in a new volatility regime. From the start of 2017 through to the end of this January, realized volatility was 6.9%. In February and March, it was 23.6%.
Below I show the S&P 500 from the start of 2017 until the end of this March.
Things have changed.
But what is particularly interesting to me is how exactly things have changed, because that is a little unusual.
Generally periods of high volatility have one major cause: the Asian crash, LTCM blowup, the dot-com bubble or the housing credit crisis. But this year we have had two distinctly different periods.
The start of the turmoil was on Friday, February 2nd, when the S&P 500 dropped 2.1% and the VIX rallied 28.5% from 13.47 to 17.31. This was a large move: the 34th largest in history. But Monday the 5th was truly exceptional. The S&P 500 dropped 4.2%, but the VIX rallied 115.6%. This was the largest VIX move ever, and nearly twice the previous record of 64% (and in that case the move was a comparative blip from 11.15 to 18.31). As the VIX was only implying a daily move of about 1%, the S&P 500 return was a genuinely large event. However, the response of the VIX was well out of line with an underling move of that size. Talton's regression model linking the S&P 500 returns to the VIX returns predicted that a 4.2% drop in the equity index would correspond to a 12.8% move in the VIX. The actual VIX increase was 9 times the expected amount.
On the next day the VIX had its largest range ever, with a low of 22.42 and a high of 50.3 (the 70th high value recorded and recall that this was only a few weeks after sub ten VIX values).
This totally unprecedented tsunami is illustrated below.
And the brunt of this more was felt in the ETN space. It is often the case that dramatic market turmoil is linked to new financial products. Portfolio insurance is often blamed for exacerbating the 1987 crash. Credit derivatives were the cause of the 2008 financial crisis. It looks like volatility ETNs can be heavily implicated in February’s volatility spike.
Short volatility products had become more popular in 2016 and 2017 because the realized volatility of the S&P 500 index was very low and the contango decay was very high. Open interest increased enormously and resulted in crowding in these products prior to the crash in February 2018. In VXX alone, short interest increased nearly 1300% from the end of 2013 to the end of 2017.
It is obvious that a 100% rally in the VIX 30-day future would drive a short ETN to zero, however conditions for termination (named “acceleration” in the prospectus) don’t even require this.
From the prospectus,
“...an Acceleration Event includes any event that adversely affects our ability to hedge or our rights in connection with the ETNs, including, but not limited to, if the Intraday Indicative Value is equal to or less than 20% of the prior day's Closing Indicative Value.”
In this eventuality,
“...you will receive a cash payment in an amount (the "Accelerated Redemption Amount") equal to the Closing Indicative Value on the Accelerated Valuation Date.”
It is important to stress that XIV is an ETN not an ETF. An ETF owns the stocks, bonds or commodities that make up the portfolio, while the ETN is merely a note that pays the return on the portfolio. Whether and how the issuer hedges their obligation it up to them. This mean that there is no direct way to create an arbitrage between the ETN and its fair value. This lead to severe dislocation between the fair value of XIV and its price on Monday afternoon. By the close, the one-month future had risen by 45%, yet XIV had only dropped by 15%. Directly after the stock market close the VIX futures spiked to an increase of 100% on the day, which triggered the acceleration event in XIV.
This after-hours jump was not due to any nefarious manipulation. Around the close, ETNs re-balance their exposures to the VIX. So, on this day, short VIX ETNs needed to buy futures to reduce their exposure and long VIX ETNs needed to buy VIX to increase exposure. This severe buying pressure created a large imbalance and drove the price higher. Even if the actual product issuers were hedged with swaps, the counterparties to those agreements would have needed to hedge. This dislocation could have happened in the past and it is unclear, why in this particular case, the hedgers so dramatically under-estimated their hedging needs. Class lawsuits are already being prepared against the issuers.
February was an endogenous market event. What has followed has been driven by external factors.
The first earnings season had mixed results.
The Mueller investigation is intensifying.
Tariffs and other restrictions on free trade have been applied, driving commodity volatility which has spilled over into the equity space.
The Syrian civil war, while not unusual in itself, has led to the first direct conflict between the US and USSR since the Cuban missile crisis.
And just as 2017 saw low volatility in all asset classes, 2018 volatility is high in everything.
Sadly the cause of the February volatility wasn't clear at the time. In general it is safer to sell into an endogenous event than into a period of political and economic uncertainty.
Talton did well in February. But, as always with hindsight, things could have been better.