‘It’s Akin To The Exit Door Getting Smaller As More People Try To Run Out’: A Simple Explainer On Volatility-Liquidity Reflexivity

… conditions can change quickly, especially with diminished liquidity.

That’s from Goldman’s Rocky Fishman, who this week noted that the carry cost of longer-term S&P hedges is now the lowest since 2007.

Much has been made of the collapse in cross-asset vol. that’s unfolded alongside 2019’s “everything rally”. For a fleeting moment, previously moribund rates vol. woke up late last month, but that episode notwithstanding, this year has witnessed one of the largest cross-asset vol. collapses since Draghi’s “whatever it takes” moment.

Read more

From Doomsday To The Largest Cross-Asset Vol. Collapse Since ‘Whatever It Takes’

Ironically, it’s possible to argue that one reason why the bounce off the December 24 lows (i.e., the rebound off the “He can’t putt!” Christmas Eve massacre) was so dramatic is due to the same dearth of liquidity that made Q4 such a harrowing rollercoaster ride.

Obviously, the Fed’s dovish pivot and the “me-too” movement (sorry) it triggered among other global central banks was the proximate cause of improving sentiment, but up until April, you could easily have argued that the fundamental backdrop (e.g., political entropy in the US, global growth concerns, etc.) was largely unchanged from December.

“Given liquidity, it is plausible that just short covering, buybacks, dealers’ gamma hedging, and some limited re-leveraging drove the entire recovery”, JPMorgan’s Marko Kolanovic wrote on March 21.

The following visual compares the Q4 rout to other post-crisis selloffs and the inset shows market depth for each episode. “The average liquidity during the last selloff and recovery was less than 1/3 of the liquidity in previous episodes and about half of the worst liquidity drawdowns of the last decade”, Kolanovic went on to note.

(JPMorgan)

As Goldman’s Fishman wrote this week, liquidity has seemingly stopped improving in 2019, despite the rally pressing onward and upward. Note the headers on the following visuals:

(Goldman)

One of the most dangerous aspects of the current environment is the extent to which the relationship between volatility and liquidity is self-feeding, self-referential and otherwise reflexive. When you throw in vol.-sensitive systematic strategies and market making, you end up with what Kolanovic has variously described as the “‘Liquidity-Volatility-Flows Feedback Loop’”.

If you haven’t read that linked post, it’s a must, and in their Q2 volatility outlook, SocGen is out underscoring the point.

One section is called “Liquidity: A horror story”, and it begins with the bank quickly recapping what happened in Q4. “Periods of high volatility see an especially low level of liquidity — perhaps because of volatility being a parameter for how much liquidity to provide”, they write, adding that “what was truly different between 2017 and 2018 is that the 10-day realized vol was 8x that from a year ago.”

(SocGen, CME)

Considering what happened last year in February, October and December, and knowing what we know about the extent to which volatility and liquidity are inextricably bound up with one another in what, during a pinch, can turn into an exceptionally pernicious loop, it’s probably worth asking what the situation might look like if things got really bad.

As SocGen puts it, “it is worth reflecting on what the book depth could be in a sell-off of the magnitude seen during previous recessions, and how investors need to tweak their calculations over the time needed to exit large positions in their portfolios”.

Right. After reminding everyone that while liquidity for smaller sized trades has nearly recovered from last year’s worst days, liquidity for large sized trades is still impaired, the bank delivers a short critique that touches on everything said above. This is well worth the 2 minutes it will take you to peruse it and we’ll present it without further editorializing.

Via SocGen

Liquidity providers today quite possibly have volatility feedback into how much liquidity they provide, and if this is indeed the case, the reflexivity in markets ratchets up by multiple orders. An old-school investor’s reaction to this reflexive non-linearity can vary from mild discomfort to an existential crisis. Almost every investor is taught to be cautious about exiting a crowded trade. Having volatility feedback to liquidity is akin to a situation where the exit doors in a movie theatre get smaller and smaller as more people try to pass through it. As some stage, one must start to counterweigh the door against the utility that one derives from watching the movie. What are the options available to our hypothetical moviegoer? She must either decide not to go to the movie (stay out of the markets), go there with a plan to spend a few nights in the movie theatre in case the door shuts before she can exit (invest long term and be ready to own the market through the troughs) or take up a seat very close to the door (invest tactically and in relatively small size). Being prudent and sitting out while everyone else seems to be enjoying the best movie since 1991 is a difficult exercise in self-discipline. (Of course, a portfolio manager losing assets due to underperformance is a much more serious affair than missing out on a movie.) On the other hand, if everyone tries to bid for the seats close to the exit, they will quickly become expensive and the moviegoer will have to pay a premium to stay close to the door (high convexity in option pricing), which is not good for the pocketbook. Moviegoers with large families do not have this option anyway — they will have to take a seat somewhere in the middle to accommodate all the family members and will need to make up their mind in advance as to whether the risk of a contracting door is worth taking. At the very least, the investor needs to be aware that there is a contracting door and do the maths accordingly.