“Relentless” is probably too strong, so I’ll use “persistent” instead.

Given the persistent rise in US yields over the past two months, and the extent to which many observers view the long-end selloff as a function of supply concerns and related reservations about the allegedly parlous state of America’s fiscal house, there’s now some debate about bonds’ capacity to rally in a slowdown.

Certainly, such concerns are warranted, or will at least be excused given investors’ rather harrowing experience in 2022, when bonds not only failed to diversify an equity bear market, but were in fact mired in a bear market themselves, and as such were a source of portfolio volatility. Indeed, far from being a hedge, bonds were the sponsor of the market “event.”

Sympathetic though I am to anyone with 2022 PTSD, it’s unlikely, in my opinion, that bonds won’t rally if the US economy stumbles. Treasurys are on track for an unprecedented third straight annual loss, and with apologies, this situation isn’t amenable to a gambler’s fallacy analysis. US Treasurys can’t log negative returns every year in perpetuity. That won’t work. The world would explode. Not literally, but figuratively. I suppose it’s possible that US debt could log small negative returns consistently for a prolonged period of time, but an escalatory selloff à la Liz Truss’s gilt meltdown is a total non-starter. Indeed, that episode was itself a non-starter (the BoE stepped in).

It’s not just about a Fed “put” for Treasurys. It’s also about muscle memory. A negative NFP print, whenever it comes along, will be accompanied by a bond rally, unless bonds have already rallied by then due to, for example, other evidence of a looming slowdown. If not, then supply concerns are real. And we’d be in real trouble. If the day ever comes when there are serious questions (as opposed to unserious ones) about sponsorship for the US long-end, then, ironically, we’ll quickly have bigger problems than bonds.

But some people are going to ask the question. Witness Jefferies in the latest installment of GREED & Fear. To wit:

The debt ceiling is a useful construct for one political party to use to criticize the other. Still, as the recently agreed fudge shows, there is a certain convenience for both sides of the political aisle to forget about it altogether given the painful arithmetic consequences of the current double whammy of rising Treasury bond yields and the relative short maturity of Treasury debt. On this point, it is worth highlighting again that American corporates have done a much better job terming out their debt than the US Treasury. US annualized net interest payments as a percentage of federal government revenues have risen from 8.3% in April 2022 to 14.2% in August 2023, the highest level since August 1998.

The above is why it remains tempting to see the continuing weakness in Treasury bonds, in terms of relatively elevated yields, as evidence of supply concerns entering the market. But the acid test of this will only come if Treasury bond prices fail to rally on bearish economic data, such as clear evidence of a real weakening in the labor market.

There it is: The suggestion that bonds might not rally in the event the US labor market rolls over in earnest.

Again, I don’t believe that’s likely, but I’m keenly interested in the debate. As conveyed in “Fade The Fed?” and also in “Higher For Longer = Hard Landing,” this seems an opportune time to fade the bond bear consensus to the extent it is actually consensus.

If a hard landing’s coming, it’s coming relatively soon. I regularly lampoon the “infallible” yield curve recession narrative for the cliché that it is, but derision aside, the three-month, 10-year curve is at least as reliable as any other recession canary. Granted that’s a low bar, and I don’t think it’s especially honest to call something a “predictor” when its claims to prescience are subject to a lot of caveats about “varying lag times,” but the bottom line is this: If you accept the notion that the three-month, 10-year curve has never failed to predict a recession, then it’s either in the process of being nine for nine or wrong for the first time.

Note that I’m not saying anything about the shape of the curve here. The more policy-sensitive the sector, the harder the rally once there’s enough accumulated recession evidence to make rate cuts a foregone conclusion — the new dot plot suggests twos can only rally so much, but if the data were to evolve such that those dots are viewed by markets as hopelessly antiquated, then the door’s wide open to a larger bullish correction. Maybe that moment is six months out. Maybe it’s next month. I have no idea. I also have no real conviction regarding how much further long-end yields can rise in the event the data holds up. All of that to say I don’t pretend to have a view on the timing of the inevitable bull steepener (as recession approaches) or the scope of any interim bear steepening (if the data holds up and the long-end continues to reprice akin to the August/September selloff).

What I do know (or think I know, anyway) is that supply concerns won’t outweigh the muscle memory associated with long-end rallies on demonstrably bad economic data. It’s very difficult for me to imagine market participants adopting a view that says, “Well, payrolls are shrinking now, and ISM services looks like it might be poised to enter a serious contraction and the consumer looks like she might be finally cracking, but I’m not going to buy bonds because the supply/demand picture is more bearish than usual.” Even if you held that view, you’d be worried that other investors aren’t as rational as you and might succumb to the temptation to buy the bond dip ahead of the recession. And that’s to say nothing of the algos, which are doubtlessly programmed to cover shorts in rates and pile into any and all dovish expressions at the first sign of recessionary Bloomberg red heds.

The better question, I think, is how long-duration equities and other rate-sensitive stocks perform in the event yields do fall on bad data. I included a quote to that effect in the latest Weekly+. “The breakout of US bond yields to new highs nixed the bullish breakout of equity leaders to new highs, i.e. tech’s Magnificent 7, semiconductors, homebuilders, broker-dealers and industrials [which] all had rallied back close to 2021 highs on ‘rates have peaked’ optimism,” BofA’s Michael Hartnett wrote this week. “The most important ‘tell’ in Q4 and into Q1 will be how the ‘leaders’ react once yields turn lower,” he added. “If lower yields = resumption of big XHB, SOX upside, it’s ‘Bull5000,’ but if lower yields can’t generate upside, it’s sell the last rate hike and back to ‘Bear4000.'”