After the closing bell sounded on another “worst since 1987” session for US equities, I quoted Nomura’s Charlie McElligott describing the current state of affairs.
“We continue to exist in a rolling ‘VaR-down’/’de-grossing into cash’ state, which means further knock-on deterioration in market depth and liquidity from MM’s as well, going hand-in-hand with gappier price action as bid-ask dramatically widens on much smaller size”, Charlie wrote, in his Monday missive.
This has been the regime for weeks. Following February expiry, the vaunted “gamma pin” lost quite a bit of its power. Equities were unshackled. The dynamic which had kept stocks in a range, tamping down volatility and encouraging a steady grind higher against a “QE stasis”, Goldilocks macro backdrop faded, and the dominos started to tip as the virus news worsened.
Once the COVID-19 panic got things moving in the wrong direction and dealers’ gamma profile flipped to the “dark side” (as it were), hedging flows exacerbated price moves. After that, CTA deleveraging began as spot careened through key levels.
As trailing realized was pulled higher, the vol.-targeting crowd (which came into the selloff with the highest exposure since January 2018) began to pare their exposure. That deleveraging executed “passively” (if you will) in the market until the vol.-targeting universe had pared some $150 billion over the course of a month. By midway through last week, there was very little left for that crowd to sell.
In the final act, the risk parity kraken was released.
By the time Thursday’s selloff – which seemed really bad at the time, only to be upstaged just two sessions later on another “Black Monday” – was in the works, RP’s footprints were showing up. The next day, the scope of the deleveraging was revealed.
Nomura’s model estimated aggregated gross exposure was cut “from 450% (61st %ile since ’11) down to 305% (2.9%ile) on the session”. This was the breakdown, according to the bank’s model, as of Friday morning:
“[The] risk parity gross-down shock from last week shows that even after the Friday bounce, estimated exposure remains ‘gashed-down’ to only the 20th %ile since 2011 [with] equities gross exposure just 0.3 %ile; bonds 13.3 %ile; credit 35.4 %ile; and commodities 51.8 %ile”, McElligott wrote Monday.
“The proliferation of VaR sensitive investors, such as hedge funds, mutual fund managers, risk parity funds, variable annuity funds, banks, dealers and market makers raise the sensitivity of markets to self- reinforcing volatility-induced selling”, JPMorgan’s Nikolaos Panigirtzoglou notes, in a piece that essentially reinforces and otherwise recaps the dynamics outlined on innumerable occasions by Panigirtzoglou’s famous (in market circles anyway) colleague Marko Kolanovic.
It’s the dreaded “liquidity-volatility-flows” feedback loop that I refer to habitually in these pages. As I wrote over the weekend, the risk parity unwind was just another manifestation of the same “we’re all momentum traders now” dynamic I’ve been so keen on emphasizing for the last couple of years and laid out so powerfully earlier this month by McElligott.
Read more: We’re All Momentum Traders Now
These VaR shocks are catastrophic after a decade during which suppressed cross-asset volatility allowed for ever more leverage to be deployed across various asset classes. Some model-based, systematic deleveraging is faster-moving than others (e.g., trend/momentum strats impact the market rapidly, while vol.-targeting and risk parity deleverage on a delay as trailing realized gets pulled higher), but the bottom line is that people are beginning to understand more and more that this ostensible quant-ish “arcana” not only matters, but is in fact the only thing that matters on certain days.
Here’s JPMorgan’s Panigirtzoglou with a quick recap of the basics and how it can all end up going awry:
These investors set limits against potential losses in their trading operations by calculating Value-at-Risk metrics. Value-at-Risk (VaR) is a statistical measure that investors use to quantify the expected loss, over a specified horizon and at a certain confidence level, in normal markets. Historical return distributions and historical market volatility measures are often used in VaR calculations given the difficulty in forecasting volatility. This in turn induces investors to raise the size of their trading positions in a low volatility environment, making them vulnerable to a subsequent volatility shock. When the volatility shock arrives, VaR sensitive investors cut their positions as the Value-at-Risk exceeded their limits and stop losses are triggered. This volatility induced position cutting becomes self-reinforcing until asset prices reach a level that induces the participation of VaR-insensitive investors, such as pension funds, insurance companies, SWFs, endowments, households or corporates themselves via share buybacks.
That’s the long and short of it (no pun intended), and in Panigirtzoglou’s view, last week was more about those dynamics than it was escalating COVID-19 fears. That’s not to say that the coronavirus epidemic isn’t everywhere and always the proximate cause of the market’s fundamental consternation. It’s just to underscore the extent to which these self-feeding loops can take on a life of their own after something comes along and tips the first domino.
“Last week we saw the biggest VaR shock since the Lehman crisis”, Panigirtzoglou writes, referencing the following visual, which is just volatility across a handful of key assets.
Panigirtzoglou then reiterates that “this huge rise in volatility likely induced position reduction by VaR sensitive investors, not only asset managers but also dealers and market makers, as they breached their VaR limits”.
The problem, as alluded to above, is that volatility is inversely correlated to market depth. As vol. rises, market depth becomes more and more impaired. The more impaired it is, the larger the price swings, and the higher still is volatility. That’s the “closing of the doom loop”, as it were.
In the final straw, market makers pull back, bid-asks widen out, market depth completely collapses, and volatility simply rises in what might as well be a straight line.
“Effectively [last] week we saw a new phase characterized by a vicious and self-reinforcing circle between volatility, liquidity and VaR related selling”, Panigirtzoglou says, summarizing. “VaR sensitive investors reeling from the spike in volatility are admittedly unlikely to return soon”.
And so, it’s left to the fundamental/discretionary crowd and VaR-insensitive investors (JPMorgan again lists pension funds, insurance companies, SWFs, endowments, households or large companies via share buybacks) to catch this falling knife.