HALLOWEEN BILLS FEAR JOBS
EDU DDA Oct. 31, 2024
Summary: Suddenly the 4-week bill yield crashes by 11 bps today. Why? To figure that out, we have a look at more repo fails plus journey across the Pacific to see what the Japanese are up to. Not the BoJ, though the central bank does provide a useful contrast to this review. There are bills on both sides plus term-SOFR. In the end, however, the 4w bill today might be a bet on not just a bad October payroll report, it could be even more consequential than that.
An abrupt eleven-point drop in 4-week bills, more repo fails, plus term-SOFR. All of a sudden, there is lot going on at the front end of the curve. That’s already unusual given how at least mainstream sources are increasingly convinced the Fed is going to be moving more carefully with rate cuts. Perhaps even skipping a meeting.
Term-SOFR strongly says not a chance.
What about bills?
They seem to be in agreement though there is now clearly more going on there. In addition to the 4-week, the 3-month bill is being bought. The latter might be with SOFR in pricing Powell staying on track, though the former is right now behaving more like it had in the first half of September. That’s where repo fails are trending.
This, of course, means a whole lot having to do with US recession risks, so we have to begin in…Japan.
The Bank of Japan voted today to hold its interest rate steady at its current 25-bps level. Officials had been more cautious about the next one after what happened in early August. That reluctance was amplified by some well-founded concerns the rest of the global economy - singling out the US, specifically - might put a dent in Japan’s attempted recovery.
Following recent big name data, those reservations are being discarded. August’s volatility is already three months ago, plus recent stats on the US economy up to and including Q3 GDP have allayed many people’s concerned there would be an imminent collapse into full-blown American contraction (even though the vastly more important income data from today vehemently disagrees).
For the BoJ, however, it means the central bank is getting back on track for another rate hike. Speculation centers on either December or January, with markets betting on the latter as Economists and commentator-types thinking more the former. Governor Kazuo Ueda obviously isn’t going to make any commitments, though his comments today made it clear he’s targeting one or the other.
It's all about the yen.
“Our basic stance is that if our economic and price outlooks are realized, we’ll respond by raising rates,” Ueda said. It was also no longer necessary to say the bank had “time to mull” before making any decision to adjust policy as risks from the US economy had largely receded, he added.
“We’ve been looking at the downside risks to the US and overseas economies, but that fog is clearing somewhat,” Ueda said. “Needless to say, new risks could emerge depending on the policies coming from the new US president.”
As far as those US risks, that view is not shared widely by the marketplace, only one small corner, the one closest to the BoJ. JGB yields have moved up in tandem with those around the rest of the global bond market. Whether there is or was some pricing over another rate hike isn’t clear, it never is, though it is notable how the 10s and 5s have basically plateaued since the middle of this month. They’ve been going sideways at a level below where they’d traded on “hawkish” BoJ back in July - despite that previous 25-bps rate increase.
The closer you get to the front of the JGB curve, however, that’s where you get more of a probability of the next rate move from the BoJ. The 1s and 2s, for example, have retraced their October rallies from back when Ueda was still “cautious.” This retracement has brought the 1s to a new more-than-decade high while the 2s are now just short of one.
That puts the contrast between the front and the back into sharp focus. The short end is tracking what central bankers are thinking in the short run since officials have a unique ability to tune out and ignore building risks – such as when the BoJ did exactly that when it hiked into a building storm in late July.
JGBs were acting out a similar dichotomy then, too. ST rates were rising into the rate hike while the long end was rallying (modestly) with the rest of the world’s bonds as these US recession risks became too much to keep ignoring. Both collided at the end of July and to start August, not because the BoJ hiked rates rather due to the fact Japanese financial carry traders were fleeing those very US recession risks Ueda now says have “receded.”
The short vs. long in JGBs isn’t the only divergence in JGBs, either. There’s an even sharper contrast between short-term bonds and the 3-month J-bill. Yes, that means global US$ collateral.
At the same time the 1s and 2s are pricing the next Ueda hike, the 3m bill continues to yield next-to-nothing, nearly the full quarter-point under the current O/N rate let alone what the policy benchmark might be three months ahead after December or perhaps January. The 3m J-bill has barely moved at all from its low despite the 1s and 2s.
If you aren’t familiar with how that relates to US$ repo and collateral, you can either go here for the full background DDA version, or go here on YT for at least the basics. The shortest version is simply how when faced with collateral scarcity in US$ repo, Japanese carry traders will swap yen collateral for scarce USTs by rerouting their borrowing into FX and out of repo. Swaps allow them to effective post yen (the other side of the swap), which they have easy access to loads of it, instead of using USTs as collateral which, again, must be getting more difficult to easily and efficiently source.
It's more expensive in FX which is why repo is preferred, so this is not quite an emergency backup, though one the carry traders would prefer to avoid if at all possible especially during times like…apparently right now.
For one thing, we keep getting signals that collateral in US$ repo is not as fluid as it really should be given the amount of bills being constantly issued (and that’s another matter that isn’t appreciated nearly enough; the rising trend in collateral problems at this level of bills). The 3m J-bill therefore has tracked this reversal in repo fails, so already a pretty solid connection strongly implying this continues to be the case.
According to the latest data on fails released a few hours ago by FRBNY (this comes from the weekly primary dealer survey, so this isn’t an exhaustive tally of fails or collateral, the survey only captures a portion of the repo market from what is reported to the Fed by specifically primary dealers therefore leaving particularly offshore doing who knows what; we make an assumption that if fails, for example, rise for primary dealers there is very likely a similar or at least some degree of difficulty across all the rest of repo) fails moved back above $200 billion combined during the week of October 23.
Outside quarter-end weeks, fails hadn’t been more than $190 billion in any others since last December. This confirms the rising trend in fails is indeed ongoing.
Now, all of a sudden, the 4-week T-bill rate just collapses today – no hyperbole, either. According to the US Treasury’s own calculation for its “investment” yield, it went from 4.87% yesterday, at the low end of its recent short run range, to just 4.76% after today’s trading. Why?
There is no month-end fail problem, collateral or bill shortage, at least there never really has been before. None of the other bill maturities did anything close to that. The only one which has traded similarly is the benchmark 3-month bill, but that was only three bps today and down around nine bps total from last Friday.
Some of this reduction in bill yields is definitely related to the next set of Fed rate cuts. Mainstream sources may be talking about even the FOMC pausing at its next meeting in a week and half, the money markets are seeing just the opposite, cuts coming at both November and December. Not only are bills moving down (like Japan, there is a difference between the short and long ends of the UST curve), term-SOFR is on a direct line to lower policy rates.
For a reminder, term-SOFR here is the CME’s calculation for the implied term structure for the O/N SOFR rate, which is not the same as term-SOFR futures – though, somewhat confusingly, term-SOFR is derived from term-SOFR futures. In fact, I had been planning on doing a deeper dive on term-SOFR today until the 4-week bill stole the focus and made it more about collateral than strictly that “term structure.” We’ll revisit this other topic at a later date.
One-month term-SOFR is now down to 4.67% as of yesterday (the calculation is always a day behind). It hasn’t deviated at all since beginning to pick up the Fed pivot back in mid-August. The rate altered its track right away after the FOMC meeting and announcement in September meaning it has been zeroed on a twenty-five in November this entire time despite the last payroll report, GDP and everything else.
Three-month term-SOFR only had the one hitch related to the September payroll report. It was enough to push the rate up, but the tenor has been unwinding that backup throughout the balance of this month more like the one-month rate. Three-month is the one signaling a rate cut in December, only temporarily rethinking it and not by all that much.
Together with the bills, term-SOFR is not buying Fed regrets cutting let alone rethinking more of them. Maybe the data has looked good in the mainstream, but, in reality, it hasn’t been good especially where it counts (incomes confirming labor data on hiring, hours, etc.) At the very least, the market says it is soft enough to keep the FOMC erring on that side of macro caution.
I’d have put the 4w T-bill into that category, too, until today. This big drop is more like what had happened in the first half of September when its yield dropped so much faster and farther, even getting under term-SOFR for a few days, as collateral got to be a bigger problem heading into the September bottleneck than was appreciated at the time outside repo and FX therefore Japan’s bill.
That last one is the key, I believe, to collateral. They’re the ones stirring up the system which, if I’m correct, means US recession risk, too. We’ve seen several indications that Japan financials don’t have the patience nor the positive outlook Ueda does over US macro potential. The more they don’t like that potential, the more they unwind and end creating these difficulties in collateral and funding as a byproduct.
To sum up: the Bank of Japan says US risks have receded therefore rate hikes are back on the table for Japan in the near future. ST bond rates are trading that way but LT yields are not. The latter is another signal that the next in the global rally may be getting close, which would be the opposite of Ueda.
Collateral absolutely is, too. UST scarcity is tied to still-rising recession fears, therefore more unwinding of carry trading, the effects of which are still being picked up in repo fails, the 3m J-bill and now, all of a sudden, an unusually strong move by the 4-week US T-bill.
On top of all that, term-SOFR is steady about upcoming Fed rate cuts, which would be less likely if Ueda was right, too.
Everything outside the BoJ and the very short end of the JGB bond curve is thinking US risks have receded. When included the collateral across the oceans, there is also an indication of fragility especially in the face of how many bills are constantly issued.
In terms of macro data, the markets are positioned like incomes not GDP or September payrolls. We’ll see tomorrow about the October jobs report, though there is a very good possibility it will already be negative given hurricanes and Boeing.
Is there more negative coming with tomorrow’s report beside those one-offs?
That just may be why the 4-week did what it did. For all the “good” data on the US, the funding/collateral markets weren’t willing to just wait and find out. Not a lot of faith, in other words.
