On Thursday afternoon, following an ugly 30-year auction that sent yields surging just hours after the hottest MoM core CPI reading since January 2018 seemingly undercut Jerome Powell’s contention that inflation pressures weren’t likely to materialize anytime soon (despite piling rate cuts atop near-record-low unemployment and tariffs), we wrote the following:
It doesn’t do anybody any good to wax hyperbolic/hysteric about this, but it is worth noting that a violent selloff in bond land or some kind of vicious bear steepening episode likely wouldn’t be digested well.
Of course, waxing hyperbolic/hysteric is something different from warning that something is possible. The former is the purview of tabloids, while the latter is just prudent risk management.
This week, long-end yields rose pretty sharply, on a combination of profit-taking after a stellar run for bonds and a nascent reflation narrative taking hold as markets ponder the prospect that central banks might be some semblance of successful in engineering growth and inflation.
The good news this week was that equities hit new highs even as bonds sold off. That opens the door to a scenario where positive economic data and a rosier outlook for the economy can be greeted with cheers by risk assets instead of the jeers that prevail when the perverse “good news is bad news” dynamic is still firmly entrenched. As noted on Friday evening, the fact that central banks have now pre-committed to easing further helps to mitigate fears that better data will mean less accommodation.
The risk, though, is that yields rise too far, too fast. That would be a decidedly unfortunate development at a time when duration risk is the highest it’s ever been, as investors reach for yield in a world increasingly devoid of it.
As Bloomberg warned late last month, the current setup (i.e., record high duration) means a mere one-percentage-point uptick in yields would translate into a $2.4 trillion loss. With “long USTs” the most crowded trade on the planet (according to BofA’s latest Global Fund Manager survey), trend followers riding the wave and risk parity seemingly all-in (and levered, by definition), the ingredients are there. As we put it on Thursday, “with no shortage of kindling scattered about, all we’re waiting on is a struck match”.
JPMorgan’s Nikolaos Panigirtzoglou talked at length about this two Fridays ago, noting that the setup “points to significant risk of an abrupt bond reversal going forward”. The chart on the right shows CTAs benefiting from the bond rally, while the visual on the left is just bond vol. “creeping up” as global yields decline.
Bloomberg did a little post on this as did at least one other platform. It’s worth noting that while the financial media often juxtaposes Panigirtzoglou’s views with those of Marko Kolanovic (the implication is usually that Nikos is bearish and Marko is bullish), the reality of that dynamic is simply that Panigirtzoglou, based in London, writes about a variety of topics in his weekly notes, while Marko’s team is responsible for equities and deriving the bank’s official price targets. Sometimes there are differences, but once a month or so, they’ll get together and synch things up for a global asset allocation piece.
In any event, Panigirtzoglou flagged the obvious in the piece mentioned above, which is that “any abrupt selloff [in bonds] is unlikely to be favorable to equities given the support that equity markets received from the strong rally in bond markets in recent months”.
All of this conjures memories of VaR shocks/tantrums past, including the 2015 bund tantrum.
Well, in a note dated Friday evening, Goldman says the likelihood of such an event is lower than it was then. After noting that their positioning metrics (including a CFTC-based measure and a shorter-term options indicator) have “been squarely on the side of the rally in duration”, for most of the year, the bank admits that “both positioning metrics are now somewhat more neutral, suggesting that positioning is less likely to be the strong tailwind to lower yields that it was in the first half of 2019”.
That said, Goldman says neither measure is consistent with the setup that presaged 2015’s episode.
“By the same token, however, these positioning metrics do not suggest that longs are crowded to the point of putting the bond market at significant risk of a large sell-off, which is in contrast to early 2015, when an unwinding of extreme positioning contributed to the sharp spike in yields”, the bank writes, referencing the visual below and maintaining that in their view, “risks remain balanced towards a rally in the second half of the year, but the extent is likely to be more limited relative to the first half”.
Obviously, Goldman’s suggestion that bonds will rally in H2 is colored by their recent kitchen-sink cuts to year-end yield forecasts, but the point here is simply to flag another, less dire take on the prospects for a sharp rise in yields.
Crucially, none of the above is to say that the current setup isn’t conducive to a “shock” bond selloff. Indeed, coming full circle to what we said at the outset, we’re the furthest thing from sanguine, and anyone who frequents these pages knows we’ve pointed out the risks of crowded positioning in bonds and rates on too many occasions to count.
But laying out the risks is something different from screaming “fire” in a crowded theatre, and when it comes to the “duration infatuation”/”bond love affair”, the auditorium is packed to capacity. Given that, the financial media would do well to avoid accidentally prompting a stampede to the exits.
You’re also encouraged to note that while post-crisis monetary policy has of course contributed to a higher incidence of VaR shocks (by virtue of suppressing volatility and thereby encouraging the levering up of exposure), the fact that we have names for them (e.g., “the bund tantrum”) is itself evidence that they are still relatively rare. You don’t name something that happens in the normal course of business.
Finally, to reiterate a point made above, don’t forget that at least some of the (for now contained) selloff in bonds is probably just down to profit taking after the best month since the infamous June 2016 bond rally.