Is Negative Convexity Risk ‘Flashing Red’?

This week’s trade-related turmoil served to bring a familiar set of risks back to the fore.

Last year, market participants got a crash course (figuratively and literally) in many of the dynamics that the likes of JPMorgan’s Marko Kolanovic have been warning about for years. From the February 2018 VIX ETN extinction event to bouts of apparent CTA deleveraging to the “accelerant” effect of dealer hedging to the perils of low liquidity, investors learned the hard way that modern markets aren’t always kind and can be particularly (and indiscriminately) cruel in a pinch.

On Sunday evening, following Donald Trump’s Twitter announcement that tariffs on $200 billion in Chinese goods would more than double on Friday, Nomura’s Charlie McElligott assessed the risks associated with possible profit-taking from asset managers, CTA de-leveraging on an assumed break below key trigger levels and a possible negative gamma flip.

Fast forward to Tuesday and US stocks had joined some of their global counterparts in logging steep declines amid an inversion in the VIX curve and some apparent de-risking across both discretionary and systematic investors.

On Wednesday morning, McElligott noted that we’d need to see “a few days of ‘sticky vol’ pulling trailing realized higher” for the vol.-targeting crowd to deleverage meaningfully. According to Wells Fargo’s Pravit Chintawongvanich, vol.-targeters likely had to trim their equity exposure by roughly $10 billion thanks to Tuesday’s move.

Read more: Nomura’s McElligott On VVIX, More Mechanical De-Risking And ‘Pure Minsky’

On Thursday, Barclays’ Maneesh Deshpande quantifies the risk of further de-risking. “[The] recent increase in equity volatility is likely to drive further systematic selling from Volatility Control Funds”, he writes, while reminding you that Barclays’ estimate of AUM for this universe is ~$355 billion.

As regular readers are aware, Barclays uses the S&P Daily Risk Control 10% index as a benchmark. “Going into the prior sell-offs in 2018, the allocation to equities was at 100%,  but until the past week the allocation according to the benchmark strategy had only reached ~87% to equities”, he continues, adding that the scope of the current selloff will likely see that trimmed to 80%.

Ultimately, Deshpande expects some ~$25 billion more worth of selling from these funds as their allocation is pared.

(Barclays)

On the bright side (and you can be sure this will be left out by anyone of a doomsday persuasion who covers Deshpande’s Thursday note), Barclays reminds you that relative to 2018’s deleveraging episodes, $25 billion in selling pressure isn’t really that daunting. Volatility control funds sold some $150 billion during those episodes.

Deshpande proceeds to sound a similarly hopeful tone about the flip in dealer option gamma positioning. “In context of previous risk-off episodes in 2018 the dealer short gamma ($30Bn to hedge of 1% move) is not at extreme levels yet”, he writes.

(Barclays)

For anyone who needed another reminder that you cannot take VIX futures speculator positioning at face value, Barclays reiterates what Nomura and Goldman (and anybody else with any sense) have pounded the table on recently, noting that “speculators positions going into last week’s sell-off was the most negative in history [but] a nuanced view [is] that these short VIX futures speculators are supplying vol to the long VIX ETP investors and hence these positions are offset by the dramatic increase in the VIX ETP AUM.”

That said, leveraged VIX ETP risk (i.e., the fabled rebalance risk that too many market participants didn’t understand until it was too late last year) still exists. “One significant source of risk in VIX ETP complex is from leveraged VIX ETPs which have to rebalance to maintain the target exposure”, Deshpande continues, before tempering his assessment by reminding folks that “while the AUM in the leveraged VIX ETPs is now almost half the peak AUM prior to Feb 2018 crash, the leveraged EOD flows is significantly lower as SVXY and UVXY lowered their target leverage to -0.5x and 1.5x respectively.”

Finally, he reiterates that poor liquidity (i.e., a lack of market depth) has become something of a fixture, which is obviously dangerous given that, as JPM’s Kolanovic wrote in January, the relationship between liquidity and volatility “is very strong and nonlinear e.g., market depth declines exponentially with the VIX.”

(JPMorgan)

In his Thursday note, Deshpande provides the following handy table which summarizes negative convexity risk for markets at a time when one wrong move on the trade front (i.e., one miscalculation) could start tipping dominoes.

(Barclays)

That gives you a sense of where things stand in terms of the key dynamics that played havoc with markets during last year’s episodic bouts of tumult. As you can see, Deshpande isn’t totally sanguine, but he’s by no means sounding the alarm. The title of his note sums it up best: “Negative convexity risk not flashing red…yet”.