In December, Nomura’s Charlie McElligott recommended fading the Q4 reflation euphoria headed into January as part of a 1-month price reversal strategy.
Suffice to say that played out in dramatic fashion.
Part and parcel of that call was a view that the Q4 rise in yields and concurrent bear-steepening occasioned by ostensible improvements in the prospects for global growth would at least partially reverse in January – if for no other reason than the reflation trade had gotten ahead of itself just as a bullish seasonal for bonds came calling.
The coronavirus outbreak obviously helped push yields lower, but it’s important to note that the 2s10s (just to use an easy example) had already erased 10bps (at least) of Q4’s steepening before the virus scare set in.
Among other things, the return of the “duration infatuation” re-inverted the 3-month-10-year curve and pushed 30-year yields below 2% last week, as the rally accelerated into possibly spurious epidemic headlines.
It was starting to feel like August all over again, complete with convexity flows, although not to the same degree as late last summer. “This time around we think it important to focus on both banks and REITs as a potential source of hedging flows”, JPMorgan wrote, in a note dated Friday. “Though small in magnitude, based on their Q3 disclosure we estimate that REITs had a negative duration gap as of late last year with significant negative convexity net of hedges [suggesting] little room to weather a further rally and rates are now well below late -September levels”, the bank went on to say.
Based on JPMorgan’s model, banks’ gross exposure “has surpassed August levels”, and although “a decent fraction of prior risk delivery has likely been offset”, even an assumption that banks “were flat as of year-end” would entail having “roughly $275 million per bp more to hedge at current levels”.
And don’t forget the short-covering. Late last month, McElligott noted that spec shorts in bonds were getting “punished”. But that may have run its course, potentially removing another catalyst for the bond rally.
“The rally in USTs from a ‘flows’ perspective was largely on account a rolling ‘stop-out’ for Leveraged Fund shorts in USTs which peaked last week into month-end at the highs of the coronavirus freak-out, as we highlighted the whopping one-week 99.5th %ile ‘buy’ (to cover) from Specs across the UST futures spectrum from a ‘1w change in DV01’ perspective”, he writes, in a Tuesday note, recapping some of the points from Monday.
Charlie continues, writing that the scramble to cover “will likely ‘collapse under its own weight’ especially as negative convexity hedging flows too subside as we reverse off the rally peak”.
If you’re starting to get the impression that some see scope for yields to rebound and retrace at least a portion of the recent collapse, that would be the correct interpretation.
“At this point, a significant portion of the short-covering impetus is now removed and systematic positioning is already ‘long'”, McElligott remarks. Recall that on Monday evening, we noted CTAs are squarely long duration and, citing a separate JPMorgan note, flagged the prospect that mean reversion strats could begin to fade the bond rally.
Clearly, this setup is conducive to some “positive” macro catalyst (e.g., a “good” virus headline or some market-friendly political development) accelerating the nascent bearish impetus for bonds.
On Tuesday, those catalysts took the form of more liquidity from the PBoC, an emergency OPEC+ pow wow aimed at shoring up oil prices and the confusion in Iowa. McElligott, ever the keen observer and not one to eschew colloquialisms and humor, puts things in context. To wit, from his Tuesday missive:
Last night’s primary disaster in Iowa for the Dems (with some indications that Biden was tracking at 4th)—which is “trolling-ly” being spun on Twitter as “Trump Wins Iowa Primary,” and driving some of this current “risk-ON” trade overnight and havens again on their back-foot as Trump is viewed as higher probability to win against “hard left” candidate—is of interest, because it shows that it doesn’t take a ton to cause some outsized moves to reverse some of last month’s / last week’s “extremes”.
Does any of this mean we should expect to hear McElligott (or anyone else, for that matter) suddenly pivot back to a hard reflationist narrative centered around sharply higher yields on the back of, for example, the assumption of a “kitchen-sink-type” stimulus push out of Beijing?
In a word: No.
“Looking-out one year into end-2020, I am simply on the record as saying that there was no true catalyst for a rates selloff to the extent that many strategists and market participants were prognosticating”, Charlie says, before reiterating his own view, which he described in December as “more realistic”:
This current UST “cheapening from the prior extreme” is just another swing in the pendulum back from one end of the range (1.50) to a more stable state (1.60), and not a new regime of risk-taking beyond a tactical trade tied to positioning.