Market participants are suffering from (or maybe “relishing in” is better) a veritable “VIX-ation.”
Why is vol so low? What’s the role of VIX ETPs in all of this? Are there structural factors at play that will effectively keep vol suppressed indefinitely? Are central banks indirectly suppressing vol and if so, can’t that theoretically go on forever? What happened last Wednesday? What happens if a sudden vol spike causes VIX ETPs to rebalance?
Note that last question. Those inclined to think one (or three) steps ahead might very well posit what we described last week as “the market’s nightmare scenario.” To wit, from “Investigating The Market’s ‘Nightmare Scenario’“:
One wonders, for instance, what would happen if rebalancing by VIX ETPs after a sudden vol spike exacerbated said vol spike which in turn forced vol control funds and CTAs to deleverage all at once. And then there’s the risk parity component.
One person who has some some answers to a lot of the questions posed above is Deutsche Bank’s Rocky Fishman who, in addition to having a pretty awesome name, writes some really good notes.
Maybe it’s just us, but it seems like Rocky has been busier than usual lately. His latest is called “VIXation: VIX Markets In Depth.”
In it, he begins with a quick Q&A before moving quickly to one of his favorite topics: the feedback loop created by inverse and levered ETPs and more specifically, just how much in VIX futs they’d need to buy in a rising vol environment. As Fishman reminds us (and he loves to remind us about this), “low VIX futures levels make an N-point vol spike look like a huge percentage.” That, as we and several other outlets have noted recently, is a pretty important part of the equation (more here).
Read more below from Rocky’s latest.
Via Deutsche Bank
How did the VIX get so low? Did that indicate complacency? Will we see single-digit VIX levels again?
SPX realized vol was the #1 driver. By far the most important reason the VIX hit a multi-decade low was that realized vol hit a multi-decade low. Exponentially-weighted metrics of realized vol were dragged down by both the very low recent realized vol and the extended period since there had been a period of high volatility. Fundamental explanations for the low index volatility include low correlation between SPX sectors, steady economic growth, and low interest rates. The most important technical contributor was gamma: we believe that short-dated option-selling strategies left dealers with a large long gamma position, causing their delta-hedging to dampen SPX moves.
Contributing factors included falling implied/realized ratios, post-French elections monetization, the new VIX methodology, and skew. Beyond realized volatility, several other factors contributed. The implied/realized volatility ratio fell following French elections – partially driven by monetization of tactical option positions put on before the elections. In 2014, the VIX changed its methodology to include weekly options instead of only monthly options; the old methodology would have resulted in somewhat higher VIX lows, with a low point of 10.5. Low skew also helped pressure the VIX, as the VIX calculation includes a high weight on out of the money put options.
We believe this year’s VIX low is behind us. It’s hard to say we will never see a single-digit VIX level, especially in a high vol-of-vol environment that can see markets rebound and volatility fall very sharply in the aftermath of an event. That said, we believe this year’s low in SPX implied and realized vol is behind us, and expect at least somewhat higher volatility for the coming months.
Spot VIX Level: Last week’s VIX spike driven by rising (but low) realized vol, high vol risk premium
The VIX is historically low again. Wednesday’s selloff, driven by the turmoil in Washington, was not extreme (-1.8%) but drove the VIX up by 5 points – a larger-than-usual response to the spot SPX move with uncertainty expected to continue. The fundamental worry that appears to be driving vol is concern that business-friendly Republican legislative priorities are at risk if much of the short period between now and congressional mid-term election campaigning gets consumed by the current controversies.
Expanding vol risk premium, high skew helped the VIX rise. Even after Wednesday’s selloff, SPX realized vol is quite low, and in single digits by just about any short-term metric. (For “feedback loop” watchers we note that this low RV leaves typical vol control funds fully allocated even though vol is higher than it was.) Implied/realized ratio growth, and the impact of much higher SPX skew both contributed to the VIX’s growth.
VXST doubled. VXST 9-day VIX index hit its first readings below 9 and above 20 in months on back-to-back days Tuesday and Wednesday, as the abrupt change in risk sentiment drove the price of SPX weekly options sharply higher. The rush to short-term protection reflects the just-in-time hedging strategy that has become prevalent in recent years. As markets rebounded, the VXST returned to the 10’s by the end of the week.
VIX ETPs: Large short, levered ETPs would have substantial vega to buy on a further spike after buying almost $50mm vega Wednesday
Vol spike drove up the vega outstanding by ETPs. The short products’ short-covering and the levered long ETPs’ adding to long positions increased the ETP market’s vega outstanding from a relatively low level (i.e. not very long) to a historically average level. SVXY inflows following the event indicated ETP investors were “fading” this spike; historically we have seen increased interest in short VIX ETPs when the VIX has risen.
Big VIX futures short? ETPs offset the short futures position. Data reported to the CFTC shows the “leveraged fund” community with a large short VIX futures position, which coupled with rising short interest raises concern that there are large short VIX future positions that could need to be covered in a severe shock. We believe that the short futures positions in the CFTC data, which closely track the ETPs’ long position, are typically held as part of multi-leg strategies that are often long tail risk and would therefore not be covered quickly in an event. We see the close linkage between the CFTC data and the VIX ETPs’ position as evidence that VIX ETPs are predominantly created by dealers to be sold to end users, rather than created to be loaned out; if the shares were primarily create-to-lend, dealers would not need offsetting futures positions.
VIX ETPs are a larger-than-usual feedback loop in markets. Inverse and levered VIX ETPs’ need to buy VIX futures when vol is rising and sell it when vol is falling creates a feedback loop in vol that can lead to high vol-of-vol. Currently, the combination of low VIX futures levels (making an N-point vol spike look like a huge percentage), large short ETPs, and large levered ETPs leaves over $70mm vega to buy on a hypothetical 5-vol spike in the VIX futures curve – lower than was the case prior to Wednesday but still high. VIX futures (and secondary ETP trading) volumes have been strong, so the market impact of a large amount of VIX futures to buy may not be dramatic.
Short ETPs have had post-event inflows. The SVXY has drawn significant cash inflows in the couple of days following Wednesday’s selloff, helping its AUM rebound.
Vol of Vol: Washington turmoil drives implied vol-of-vol to spike levels despite low YTD realized
Uncertainty about uncertainty. Like so much that has happened in the last few months, last week’s events leave true uncertainty about uncertainty: should we be concerned? Do these latest developments from Washington put the SPX’s valuation at risk? It’s not surprising, then, that the VVIX and other metrics of vol-of-vol never did not hit historic lows this year and have now risen to high levels.
Vol’s downside driving high vol-of-vol. Fundamentally, the reason we have high VIX option prices is that assuming this selloff becomes an isolated event (from the market’s perspective), realized vol conditions could return to their exceptional lethargy of the past few weeks. If that happens, any outright long vol positions can be costly, so the optionality provided by VIX calls can be valuable.