HOW LOW CAN ALL THESE CURVES GO?
EDU DDA Sept. 22, 2025
Summary: Waffles are back on the US rates menu, but they don’t seem connected to the slight back up in Treasury note yields. Instead, bills keep plowing lower. Anyway, most people simply want to know how low rates might end up going. We have a few ways of sketching a rough outline, starting with swaps then applying another new record low in CtG. In the real economy, getting to really low future ST rates was always going to be about Beveridge, though maybe not the one right now you were thinking.
The short run outlook for interest rates is somewhat backward in the respect ST rates keep heading lower while Treasury note yields rose modestly for the fourth straight day. In fact, after hitting 4.01% back on the 11th, the 10-year CMT is fourteen basis points higher. The 2s have backed up to 3.61%, rising a dozen basis points from their recent low on September 8.
While the rise in rates could be the return of waffling, more likely we’re seeing September emerge as the month draws toward its close. If the markets thought there was any real backsliding on rate cuts among the FOMC’s members, we’d see Treasury bill yields following the notes. Instead, the bills are bid especially at auctions where institutions want even more of the instruments and are bidding higher prices to make sure they get them.
Today the usual 3m and 6m sales were heavily bid revealing cash market expectations despite a roster of Fed speakers speaking on their favored topic, the “inflation” only they see as a possible problem.
The ongoing moderate curve divergence does raise some questions. One that I get frequently across all platforms is how low market rates might be able to go when given enough time. Really, what everyone wants to know is how low the 10-year yield will get. Either that, or how far the Fed will cut. While predictions are impossible, we do have some ways of estimating future probabilities and setting some parameters, starting with the swap market.
Continued eurodollar tightening (as discussed Friday) is one means of forcing rates downward, either directly as policymakers have to respond to financial “volatility” or indirectly from the consequences put on the real economy. On the macro side, sliding further down the Beveridge Curve in the US is one means, but there are other troubled labor markets around the world, starting over in China, that would lead to the same place.
China, it appears, is finally seeing the other side of its deflation. Trying to correct for “oversupply” necessarily means cutting back on supply which therefore also means lowering production volumes and by extension the need for workers.
The one outcome the Chinese have been trying to avoid before now is showing up in more data and evidence. As it does, that employment deflation puts even more downward pressure on global bonds, or exactly what we’re seeing in indications like copper-to-gold and interest rate swaps as both keep pointing in exactly the direction everywhere is going.
Short run outlook
Today’s 3m and 6m auctions produced even lower rates across-the-board. For the benchmark 3-month maturity, the high rate – the calculated yield which clears the whole $82 billion sold – slid to 3.86%, with a low (a gauge of demand intensity; the lower the low yield goes, the more auction bidders are saying they will stand ready to pay higher prices if they have to in order to ensure they grab their share of the bills being sold) at 3.78%.
Those are down roughly four basis points each from last week’s auction, meaning despite the back up in note yields noted above it hasn’t shifted the cash market outlook for the short run. The final 3m auction for last month produced a clearing rate of 4.10%, meaning a fifteen-basis point further decline with the August payroll report, CPI, retail sales then the September FOMC in between.
These results aren’t projecting an imminent acceleration into the Pringles, by any means, instead another steady and ongoing contribution to the path toward lower rates overall.
Not for lack of trying on the part of the waffling gang. Already this week we’ve heard from several Fed speakers, each trying to bring “inflation” back into the conversation (in what increasingly appears to be a transparent attempt to not have to admit being wrong about it).
The most head-scratching one was Cleveland’s Beth Hammack. She used both “robust” and “overheating” regarding the US labor market, citing a low number of layoffs and a low unemployment rate. She then puzzlingly added, “I think that we should be very cautious in removing monetary policy restriction. It worries me that if we remove that restriction from the economy, things could start overheating again.”
It doesn’t matter that the economy never once was, at any point, anywhere close to overheating. If you account for all workers and not just those the BLS considers officially part of the labor force, the unemployment rate would be north of 6% already (when I account for labor force dropouts adjusting the current unemployment rate that’s only in the near-term context of the past two years rather than looking at where the participation rate is in relation to before 2020).
THE ADJUSTED RATE WOULD BE MUCH HIGHER IF WE WERE ACCOUNTING FOR LF DROPOUTS USING THE PRE-2020 PARTICIPATION RATE
This is the kind of thinking which infects Economics; should the Fed lower the fed funds target rate a couple more times suddenly the labor market which already shows jobs have been lost this summer would somehow pop right back up and soar to the point Phillips Curve inflation ignites. No one bothers to ask – or demand an explanation – of just how that would happen.
Given so much that’s wrong in that train of thought, it’s easy to see why it isn’t taken too seriously in the markets outside of where it might create doubts about rate cutting. In this case, there doesn’t appear to be even that much going on.
St. Louis’s Alberto Musalem was the day’s other waffler. This is nothing new as it was Musalem who I pointed out all the way back at the end of March when introducing my breakfast terminology for all this. Rather than argue against rate cuts outright, like Hammack he worries that more than a few more rate cuts might make them “overly accommodative” and we wouldn’t want that:
I supported the 25-basis-point reduction in the FOMC’s policy rate last week as a precautionary move intended to support the labor market at full employment and against further weakening. However, I believe there is limited room for easing further without policy becoming overly accommodative.
Two of the Fed’s biggest waffles aren’t against rate cuts, they merely prefer more “caution” which simply means neither of them is quite ready to admit defeat on inflation.
So it is easy for the bill market to overlook both as more noise. Given the ongoing drop to ST bill rates that raises the chances the short run rise in note yields from the 2s out to the 10s really is nothing more than the September effect. Compared to last year’s September, this one is preceding with far less enthusiasm, to put it kindly. To me, that’s the market looking at the shape of the economy 2025 vs. 2024 and seeing the downside being far more advanced (and obvious, outside of the macro tourism industry) with the upside potential entirely absent (see: copper-to-gold).
How low long run
Where does the 10-year bull steepening stop? How about the 2s? That’s the question(s) on everyone’s mind. If not how many more rate cuts from the Fed, with or without Jay Powell, then seeking some estimation for the terminal low on the market curve.
I’ve pointed out many times before the longer-run relationship between swap rates and Treasury yields, one that is already connected given that swaps are pricing ST rates (and reflect expectations about the shape of forward ST rate curves). Even so, I’m not suggesting there is some mechanical relationship from one to the other, there are way too many variables for that – including the Fed standing in the way as it has been the last three-plus years.
However, the swap market defines the longer run background for the financial system and economy into which the Treasury market will be trading and transacting (and FOMC policymakers will be waffling and cutting anyway). And since swaps are grounded in ST rates (SOFR), taking account of curve steepening during the bull steepening process there is every reason to see a rough range translated from the swap market and spreads.
That’s why over the past twenty years there’s been a relatively close intermediate and certainly long run correlation between longer-dated swap spreads and Treasury rates.
In other words, even if the Fed is forced to go back to zero that stretches just how steep the yield curve can get and gives us something of an upper boundary for our loose estimates. Going back to early 2009, to draw on an example, the difference between ST rates and the 10-year was at most just under 400 bps and only for about ten months during the most recovery-like period of “green shoots” in the second half of 2009 and the first few months of 2010 before Euro$ #2 started to show up. From that point forward, the steepness was at most 300 bps and the times when it was were more reflationary (on the upswing) than not.
That’s one thing copper-to-gold at record lows is doing – eliminating the reflationary cases and upside from the next stage. In terms of the yield curve, that argues for shallower than steep.
In other words, if we thought the swap curve was pricing ultra-low ST rates – which it does seem to be – then the yield curve at the 10s would be as low as 1% to as much as 3%, and really stretching it at 300 bps. The latter would be at the extreme upper end of a steep curve, whereas the other more consistent with an economy and monetary system experiencing “rough” enough conditions to finally expunge every last hint of waffles and put the FOMC back firmly in the more panicky rate cut category.
Right now, despite Hammack and Musalem, the swap market is mainly holding steady. Spreads aren’t decompressing any more than they had back when officials were forced to confront payroll reports rather than CPIs, likely having retraced as much of the Fed’s summertime maximum inflation resistance now that policymakers are actually back to cutting again (letting the forward rate curve steepen itself, becoming more inverted and consistent with interest rate swaps).
Neither bills nor swaps appear too concerned about either wafflers.
GIVEN EVERYTHING, IT’S NOT UNREASONABLE TO EXPECT THE 10-YEAR CMT TO GET BACK AROUND 2% OR EVEN LESS
WHEN THE CtG RATIO IS LOW AND FALLING, THE YIELD CURVE TENDS TO BE FLATTER AND FLATTENING, MEANING WE WOULD EXPECT A FLATTER NOT STEEPER YIELD CURVE IN HISTORICAL CONTEXT
To reiterate: there is no mechanical relationship between Treasury note or bond yields and swap spreads, yet, as we should expect, there is a loose one since the swap market defines the probabilities into which the Treasury market will trade, starting with the future path of ST rates which then anchor those LT maturities whenever and however we might get there.
Using other indications to define a reasonable range for expected steepness draws some rough outlines for answering those questions.
If we were to calculate a median range for expectations (which I might do at some point, working backward from swaps to figure an implied forward rate curve deviation from SOFR futures) it looks like 1.50% to 2.50% for the 10s. On the low side, it would be something like zero ST rates plus a modestly steep curve (which could end up far flatter in more worse-case scenarios). On the high side, maybe the Fed only drops to 1% plus 150 bps in calendar spread.
Given the worsening labor situation globally, the latter doesn’t seem at all unreasonable as a start.
Chinese for Beveridge
I may have mentioned the Beveridge Curve once or twice the last six months, so we’re all up to date on the possibilities stacking up in the US labor market. Beth Hammack says layoffs are low and she is probably correct on that, but it doesn’t mean what she might think it does. Transitioning to the flat path of the Beveridge Curve isn’t some instantaneous process: we have to go through a few layoffs that become sporadic then noticeable before finally getting to be overwhelming and undeniable.
Earlier in that process looks like the Establishment Survey. Being somewhat further along only gets to the QCEW.
The US labor market isn’t the only place where those concerns are high and rising. I’ve also noted over the past two months, in particular, how China’s economy really does seem to have stumbled and maybe badly.
From just the Big Three for last month, while IP did slow, retail sales only bounced back minimally from two straight months of highly unusual nominal declines while fixed asset investment appears to have crashed. Bank lending is, at best, limited, while new flows (TSF) indicate lending has also deteriorated significantly July and August, too.
The latter is almost certainly why the PBOC chose to hold benchmark credit rates (LPRs, or loan prime rates) steady for September. Banks have been resisting lower prime rates since that would only further threaten their profitability. In this environment of heavy risk aversion, they are basically only lending to borrowers at those prime rates anyway.
Since lower interest rates for them aren’t going to spur some major wave of borrowing, dropping the LPRs just makes everything worse for the banking sector, foreign them to lend at lower returns without any increase in volume to make up for them.
As it is Chinese firms appear far more likely to be laying off employees in the near term. Bloomberg reports:
To the surprise of some economists, the distress is suddenly showing up in a range of government data sets to an extent unseen in months. None of the analysts surveyed by Bloomberg News had expected urban unemployment to tick up last month, when it rose to the highest level since February.
The article also adds, “China’s labor market is deteriorating just as deflationary pressures showed initial signs of easing” as if these two outcomes aren’t related. They are the mirror image of one another.
Beijing has made it a priority to tackle “oversupply”, which simply means not enough demand as if that isn’t all over every single macro and financial statistic coming from China. The Chinese economy has been stuck in “deflation”, meaning falling producer costs and weak consumer price growth as a direct result of government directives prioritizing stable employment in a climate of falling demand.
NOT ANYMORE?
You don’t need anything other than common sense (and to have not undertaken PhD Economics training) to see the problem – if you produce more than the internal economy can demand, then the excess has to be sold somewhere else (dumping on the rest of the world) and if that doesn’t work then naturally prices are the only to way adjust. The “intense competition” everyone is talking about springs from the global economy suffering from this recession condition on the demand side.
For the CCP, they only have one choice to make: keep employment stable thereby forcing competitive price cutting; or, do something about the “oversupply” which would necessarily require pulling back on production and leading to lower levels of employment along with supply.
Xi Jinping’s government now sees persistently weak prices as the priority. Should we really be surprised that unemployment – not to mention spending and output – are about to get hit and maybe really hard?
This was always how it was going to go in China. To keep absorbing falling prices doesn’t just put jobs at risk, it puts production capacity under extreme pressure. The situation appears to be getting to that point where pulling back on supply therefore jobs is the least bad option for the Chinese central planners if only to preserve (try to) companies.
The real question is how much that is coming from the banking sector. Banks are avoiding lending in some big part because they see the losses piling up in industrial and commercial firms being pressed to hold onto workers. Are the banks so balance sheet constrained they can no longer afford to give out bad loans to finance what is really an economy-wide jobs program? If so, being at the credit limits of capacity, it would add another reason to shift in the real economy.
That is the implication, one that certainly fits only too well with where the copper-to-gold ratio is tipping. And, whether anyone realizes it or not, a Chinese economy on the cusp of some real unemployment is going to be depressing worldwide interest rates, too. US Treasuries reflect as much – if not more – global factors.
The final key to unlocking ultra-low interest rates was always Beveridge. While we’ve been tracking the US labor “curve” closely, the markets have also been keeping a close eye on China’s Beveridge curve, too. If Xi Jinping really is prioritizing prices over jobs from here on that simply means the Chinese labor market is about to get closely acquainted with its own flat part.
So, how low can rates go? Down quite a lot and it is getting less difficult to imagine.
Maybe the more pressing long run question is how low CtG is going to go. After all, just think what it might be right now if not for the supply mess holding copper relatively stable.
