“Crazy, raging rates moves” were all the… err… rage this week as dramatic front-end action and plunging bond yields grabbed headlines in the wake of the Fed’s “dovish surprise” (an encore of sorts following January’s “epic” pivot) and amid proliferating global growth jitters.
Previously moribund rates vol. woke up after sliding to record lows, hedging activity and convexity flows amplified things and, as much as it pains me to say this, it is at least possible to suggest that Stephen Moore’s demands (on behalf of Donald Trump) for an “immediate” 50bp rate cut were gasoline on the fire, not because anybody takes Moore seriously and not because anybody necessarily believes his path to the Fed is going to be an easy one (or that anybody will care what he says if he ever gets there), but rather because his comments suggest that the White House isn’t done when it comes to putting pressure on Jerome Powell. On Friday, Larry Kudlow echoed Moore’s calls for rate cuts and Trump himself took to Twitter to lambast Powell anew.
In any case, we documented all of this in real-time throughout the week, no easy task considering everything else that was going on.
The manic moves in rates weren’t digested particularly well by equities. Although the S&P ended the week with a solid gain (closing the book on the best quarter for US stocks since 2009), “risk-assets [were] spooked by the instability seen in the rates-trade”, Nomura’s Charlie McElligott wrote Wednesday, around the time mainstream financial media outlets started picking up on the swap spreads story and charts of the spike in Merrill’s MOVE index were serving as clickbait manna from heaven.
Here, for those who missed it, is the “best” version of that chart, and we say “best” because this one has it all – it’s the 5-day RoC, it’s got the standard deviation lines on it and it comes pre-packaged with a big, bold Charlie McElligott header framed in a red box:
Humor aside, this was a big story this week (it was the story, depending on who you are) and on Friday evening, Deutsche Bank’s Aleksandar Kocic was out with a new note explaining what happened.
Kocic’s explainer is (easily) the best take on this we’ve seen. It can be broken down into multiple parts and, therefore, is amenable to several posts. So, that’s what we’re going to do – break this down into multiple posts, with the first one (this one) confined to his technical discussion of this week’s gamma spike, which he describes as having “a distinct déjà vu vibe.”
“It always happens [this] way [and] no matter how hard we try to remember, when the markets calm down, everyone seems to repeat the same steps”, he says, calling it a combination of factual constraint and an obligation to move irrespective of consequences that forces a certain repetitive behavior.”
So, it’s Sachzwang and Zugzwang – a factual constraint residing in the nature of things that leaves no choice but to perpetuate the existing conditions and a requirement to act, despite knowing that doing so could be a bad idea.
For Kocic, this week’s action is more important for what it says about the future order of things than it is for anything else. “Although we see the last week’s rally as a fade with no significant fundamental baggage, it carries certain importance because of its paradigmatic value – it sets a template for future episodes of calm markets and subsequent breakout of the range”, he writes.
Next, he lays out exactly what happened. There’s no utility in trying to paraphrase this. In fact, doing so would almost certainly leave something lost in translation, so here is the extended block quote from Kocic’s piece which details the dynamics:
There are three key participants that can be bundled in the following loosely designated groups in terms of their positioning and action. Yield enhancers/systematic gamma sellers: The core position is short straddles (e.g. 1M10Y). Hedge funds: Short straddles and long low-strike receivers as insurance (e.g. 6M2Y as a hedge against emergency Fed cut). Dealers: Long ATM gamma and short low-strike receivers. In this episode, gamma sellers were just catalysts. At inception, they were short gamma and neutral delta. When the rally started after the FOMC meeting, they become less short gamma, so no need to cover gamma, but they became short delta – the receiver side of the straddle they sold is in the money — so they had to receive (in a rally!), which accelerated the rates move which added more volatility. Hedge funds, which were also gamma sellers, reached out for the tail risk protection in terms of low-strike receivers, mostly related to possible emergency rate cuts or repricing of the recession. Those were typically low-delta structures in the upper left corner.
The dealers took the other side of the gamma shorts – they developed a long position in short-dated straddles and, in order to flatten their gamma exposure, they had to short about an order of magnitude more (in notional amount) of lowstrike receivers. So, when rates rally began, the dealers started developing gamma deficit: they became shorter gamma on their long straddles position as well as on the low-strike receivers as rates approached those strikes. At the same time, short covering of duration of yield enhancers and systematic gamma sellers accelerated the rally and made dealers’ exposure even more uncomfortable.
In our estimate, there was a net deficit of about $15bn 1M10Y equivalents during the gamma spike, responsible for a 20bp rise in short-dated vol.
So, that’s the point-by-point. But Kocic doesn’t leave it there. Rather, he proceeds to put it in the context of metastability, which is a fixture of things – “structural metastability”, as it were.
The notion that stability breeds instability isn’t new, of course, and Kocic has written voluminously on how stability can be destabilizing.
Read excerpts from Kocic’s previous work on metastability
This week, he explains the action in rates in the context of metastability – predictably, his exposition is great.
“Generally, the market environment and decomposition dictates convexity distribution across different strikes”, he writes. “However, in calm markets, this distribution tends to be barbell, which in itself carries the seed of instability.”
Cue the following visual, which Kocic calls “the gamma see-saw”, where block size is indicative of notionals and the distance from the weight to the fulcrum is the risk. Longs are on the right, shorts on the left.
As you can see, Kocic essentially breaks this down into three parts, and abstracting from the specifics (i.e., from how the figure relates to the breakdown as detailed in the lengthy excerpt above) the overarching message is that as periods of serenity get longer, the short gamma position gets larger and thereby more precarious.
“At some point, more cautions players will start buying insurance against outsized moves in terms of low-delta OTM structures – this is the second stage of position refinement or diversification”, Kocic continues, describing what happens when the situation depicted in the first configuration starts to become tenuous in the eyes of some market participants. “To match a long ATM gamma, dealers have to sell an order of magnitude more OTM structures (in terms of notional amount).”
Here’s what happens next:
As the market moves away from ATM and towards the insurance strike where dealers are short convexity, the fulcrum moves to the right and the weight of the short convexity leg becomes exposed. This is the beginning of the third, short covering, stage. Dealers’ books become negatively convex and the see-saw tips towards the side of short convexity exposure (second layer). Short covering, typically in terms of buying more ATM convexity, restores the balance (third layer).
Kocic proceeds to contrast that state of affairs with the way things used to be when MBS hedging was the dominate market mode. The point, generally speaking, is that while vol. used to be higher, stability was, well, more stable – or “more robust”, as he puts it.
The “insurance mode” – as illustrated above – is defined by periods of tranquility and placidity shattered by episodic vol. spikes which expose the inherent underlying fragility of the tenuous equilibrium. That’s what we got this week.