This site is part of the Informa Connect Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them. Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 3099067.

Quant Finance
search
Buyside

Non-fundamental market sell-off and “volatility feedback loops”: a market impact perspective

Posted by on 19 February 2018
Share this article

A few days after a sudden spike in volatility sparked a stock-market crash, market participants are left to ponder the wreckage of the sell-off and the mysterious dynamics that caused it. 

All the more than the VIX averaged 11 through 2017 – the lowest since 1990 (with a short VIX futures strategy P&L at +150% in 2017, making money every single month of the year 2017), in the context of solid fundamental factors (continued global growth, solid-to positive earnings, falling unemployment), easy monetary policy and share buybacks. January’s stock-market euphoria and strong returns likely prompted pensions to reduce risk after a strong start to the year, setting the stage for a grind higher in realized equity volatility that would hit a crescendo.

The recent market sell-off has stemmed from 3 major drivers:

  • A demand from leveraged VIX exchanged-traded products (“ETPs”) which was the trigger of the jump in volatility.
  • The selling of Equities by “managed volatility funds” post the jump in volatility, whose Asset Under Management increased greatly over the past decade.
  • The hedging flows for the trillion dollar S&P 500 index put options and for the trillion dollar Variable Annuities.

Aymeric Kalife

Demand from leveraged VIX exchanged-traded products (“ETPs”) was the trigger of the jump in volatility (the VIX jumped 20 points to 37 in one day then as high as 50 the following day).

  • Market technicals were flashing warning signs, where positioning in “short volatility” had become extremely crowded. As a result the rebalancing needs of VIX ETPs following a spike in volatility was near all time highs, implying that short VIX ETPs had more vega to buy on a given vol spike than ever before, creating the potential for an outsized increase in volatility should the S&P sell off sharply: a 3-point spike in VIX futures would drive VIX ETP issuers to buy $110mm vega - double the highest ever seen before 2017.
  • Only $3.6 billion transactions in reversing 2 specific short VIX strategies have triggered jump in volatility - a tiny fraction of the roughly $2 trillion estimated to be linked to short-volatility strategies and of the $23 trillion in market value of S&P 500 companies.
  • As volatility-related products buy VIX futures in order to rebalance ahead of their 4:15 p.m. daily deadline to calculate the value of their underlying assets -- they effectively pushed up the price of the contracts and eventually the index in order to rebalance positions quickly to avoid unhedged overnight risk (ETN issuers) or excessive tracking error (ETF issuers).

Once realised volatility rose, various forms of systematic volatility control funds added outflows, initiating “volatility feedback loops”. 

  • Since the Brexit low realised volatility left volatility control funds with well below “volatility targets”, implying high Equity allocation. The sudden selloff brought then realised volatility to “volatility targets”– implying to decrease their leverage to Equities.  Actually “volatility control products” sell equities when volatility is rising and buy equities when volatility is falling, translating into market “volatility feedback loops” that exacerbate both selloffs and rallies. It is one key driver of the repeated pattern of sharp selloffs followed by consistent rebounds as experienced.
  • $500 billion of assets are tied to those “volatility control funds” that target a given level of volatility -- two-thirds of which are traded by algorithms, while 275bn$ are held in variable annuity (VA) Subaccounts.
  • A volatility spike from a low starting level (as experienced) drives the most aggressive de-allocations - particularly if the starting vol level is just low enough to be at the maximum allowable equity allocation. The strategy does most of its selling of equities on a handful of trading days : a “volatility-control fund” with a 12 percent target, 70 percent average exposure to stocks would sell equity futures totaling 15 percent of their assets, while a sudden 4% global equity selloff would drive $20bln of selling by volatility control funds.
  • Then it can take weeks to replace positions that have been sold, because realized volatility slowly "forgets" a big market move in the aftermath of a significant selloff.

Similar to 24th August 2015, the largest and quickest punch still came from “feedback effects” of hedging flows for the trillion dollar+ S&P 500 index put options and for the Variable Annuities.

  • “Hedging Feedback effects” mostly stemmed from the buyback of significant short call options consistent with the rallye, as illustrated by the massive long gamma positions held by market makers before the market crash. The hedging feedback effects were all the more impactful than ~50% trading concentrated on shorter dated options and close to the money (30% in strikes within 2% of spot and less than 2 weeks to expiry).
  • “Hedging Feedback effects” were further strengthened from the significant short puts positions held from market makers (30-60% short option convexity imbalance) or embedded within Variable Annuities, who then delta-hedged with the index, i.e. sold more index as the market dropped or bought more as the market rallied, which exacerbated the initial move and realized volatility.
  • Total risk exposure may have amounted up to 40% of the total market gamma. This was reinforced with varswaps bought back by investors, where market makers sold vanilla options (particularly OTM puts) to hedge their long varswaps and then delta-hedged those vanilla options using cash equities, pushing further down the Equity market as it slumped, specifically at the market close as the variance swap payoff is calculated based on the closing levels of the index.

Still, the potential mitigating actions to reduce the short-term impact may lie in the following:

  • Setting the hedging volatility assumption consistent with the hedging feedback effects, higher than the Black&Scholes or the average realized volatility that reach limits in case of such market turmoil, in favor of a varswap reference which  encapsulates “convexity” risk without being too much expensive.
  • Setting the “capped volatility fund” consistent with robust stress tests scenarios, specifically mean-reverting rangy extreme scenarios (repetitive up & down ones during several subsequent weeks or even months), or less reactive to short term volatility.
  • Complement the delta- hedging by a 20% portion of long-term options / varswaps in order to compensate of the negative P&L.

Fortunately, while for equities this looks like a 2015 type of crisis, other asset classes disagree: the moves in the VIX and the delta moves in cash equities have been driven specifically by equity-vol products ONLY. This is because there is a big macro/fundamental difference between now and the August 2015 market crisis that dealt with an EM crisis (e.g. China), crisis of credit spreads, collapse in commodity prices, and weak global growth, i.e. legitimate fears of a global recession. Now, the situation is exactly the opposite: global growth is very strong, US corporate earnings are at record highs (and continue to be revised higher), commodities have stabilized, and the USD is weak. Regarding hedge fund positioning and liquidations, substantial de-risking already took place as illustrated by significant reduction in equity beta of all hedge funds.

QuantMinds International banner

Professor Aymeric Kalife is the Head of Savings, AXA Group Risk Management, and an Associate Professor in Finance at Paris Dauphine University. He will be speaking at QuantMinds International, May 14-18, on dealing with persistently low interest rate environments. 

Share this article

Sign up for Quant Finance email updates

keyboard_arrow_down