Not Idiosyncratic but Systemic
Doug Nolan
Twenty-five days until midterms.
No reason to expect normal.
The AI Bubble/arms race survived another week.
“Big Tech Rally Extends Into Fourth Week at AI Bubble Fears Fade.”
Meanwhile… “AI Borrowing Spree Hammers Tech Debt in Rush to Reprice Risk.”
Another Nasdaq100 record close.
Lingering equity enthusiasm only underscores the importance of closely monitoring waning market confidence in AI-related debt.
Axios co-founder Mike Allen (October 3, 2026):
“The Financial Times, probably your favorite paper, said ‘Bessent fails to break fever in U.S. bond market.’ What did you learn from that?”
Treasury Secretary Scott Bessent:
“Well, I learned that the Financial Times is anti-American, anti-business.
And they have a fever – they’re constantly trying to create a problem for the U.S.
Let’s leave that disgraced publication in London aside…”
Allen: “So it’s fair to say it’s no longer your favorite paper?”
Bessent:
“Never was.
But they do have some excellent journalists.
But what I would say – what I’ve learned – everyone in the market knows you don’t win every hand.
But you win over time.
Everyone knows I’m a big basketball fan.
Big fan of John Wooden.
I’m a big fan of Dean Smith.
And how did they win?
You trust the process – you put a good process in place.
What we can see with bond yields, I would be concerned if we were having some kind of idiosyncratic rise.
It’s been a global rise.
US bond performance since the President came in has been the best in the world.
So, we are not seeing people sell Treasuries to buy German bunds or Japanese bonds.
Like I said, I can’t control the bond market.
What I can do is try to get people to slow down and think…”
“Never was,” a fan perhaps.
But I have a sneaking suspicion that Mr. Bessent never was without his Financial Times paper during peak mortgage financial Bubble excess.
The FT’s Gillian Tett has a place in history for her phenomenal reporting on U.S. Credit derivatives, structured finance, and Wall Street excess.
While the U.S. media was generally asleep at the wheel, the talented team at the Financial Times burnt the midnight oil reporting on developments leading up to the financial crisis.
“Excellent journalists,” indeed.
From the previous Bubble period, the names John Authers, Henny Sender, Paul Davies, Michael Mackenzie, Aline van Duyn, Saskia Scholtes, Krishna Guha, Francesco Guerrera, Brooke Masters and others remain worthy of recognition.
The FT was certainly not an enemy of the American people then - and it is not today.
Our nation faces a terrible debt problem, while our Treasury Secretary keeps saying dumb things.
The FT tradition of deep investigative journalism continues, as their journalists distinguish themselves in reporting on U.S. markets and finance, including recent focus on hedge fund leverage, the repo market, AI debt issues, debt sustainability, the Federal Reserve, US politics, the Iran War, and increasingly fraught geopolitics.
Just this week:
“Hedge Funds as Systemic Risk,”
“How a Trillion-Dollar Hedge Fund Borrowing Spree Became Wall Street’s Cash Cow,”
“Investors Look to Shelter Portfolios From Rising AI Concentration Risks,”
“OpenAI Annualised Revenues $20bn Less Than Previously Signalled,”
“Insurance Claims to Test Altman and Amodei Liability for ‘Rogue’ AI,”
“Wall Street Banks Launch Record $60bn Chip Deal for Broadcom and Anthropic,”
“Crowding Out Pits Sovereigns Against AI,”
“Surge in Borrowing Costs Hits Corporate America,”
“How US Mortgage Bonds Can Trigger a ‘Vicious Loop’ for Treasury Yields,”
“Debt in the Spotlight as Paramount Closes $111bn Deal for Warner Bros,” and
“Russia’s New Drive to Crush Ukraine” – just to name a few.
The finest financial journalism, bar none.
But I digress.
A little fact checking doesn’t hurt.
Since inauguration day (1/20/25), U.S. 10-year yields have increased 61 bps.
Canada yields are up 60 bps, Mexico 39 bps, Peru 42 bps, and Chile 36 bps.
Yields over this period were down in Colombia (24bps), Brazil (39 bps), and Panama (113bps).
Yields rose 19 bps in Switzerland, 40 bps in New Zealand, and a single basis point in China.
Sure, Treasuries have outperformed Europe, Japan and many Asian markets since the inauguration.
But keep in mind that U.S. yields surged over 100 bps in the five months preceding inauguration day, as the Fed commenced a misguided loosening cycle.
Back to September 30th, 2024, of the major sovereign markets, only Japan (215 bps) and France (194 bps) have seen larger rises in 10-year yields than the U.S. (146 bps).”
Ten-year Treasury yields traded to 5.36% intraday Wednesday.
Yields have not posted a higher close since April 2, 2002.
Sugar-coating and subterfuge will not cut it anymore.
Bessent: “I would be concerned if we were having some kind of idiosyncratic rise.”
“Idiosyncratic” has been bandied about occasionally over recent decades as various global debt crises erupted.
It’s code for “someone else’s problem – nothing to see here in the land of the world’s safe haven Treasury market and preeminent central bank liquidity backstop.”
Perhaps Bessent has an argument with “idiosyncratic.”
But this completely misses key analysis: idiosyncratic is hardly relevant in an era of deepening monumental systemic risk.
As we’ve witnessed repeatedly, idiosyncratic is relatively easy to manage – simply flush a troubled market with Credit and liquidity and watch it reflate.
Manifesting at the “periphery,” idiosyncratic risk is inherently neutralized by a robust “core.”
But when the point of an unsound and fragile “core” is reached, you confront today’s predicament.
Sovereign debt problems are these days systemic rather than idiosyncratic.
And, importantly, the U.S. is the epicenter of global government debt.
Moreover, the U.S. is at the epicenter of a historic international corporate debt boom.
Perhaps most pressing, the U.S. is today the epicenter of the global AI Bubble and arms race.
The world will surely yearn for the days of “idiosyncratic.”
October 8 – Bloomberg (Caleb Mutua, Preeti Singh, and Laura Benitez):
“The rush to finance AI is rattling investors in the more than $10 trillion US corporate market, sparking a repricing of risks around some of the biggest technology companies…
Heavyweights including Oracle Corp., Broadcom Inc. and SpaceX are among the borrowers that priced nearly half a trillion dollars of new debt this year to pay for the infrastructure powering artificial intelligence…
The tally is widely expected to rise by orders of magnitude in the months and years ahead.
Broadcom alone may raise about $600 billion to finance computing power in the coming years…
The pace and sheer scale of the borrowings, combined with unknowns around the technology and rising interest rates, is putting growing pressure on the sector…
In recent days, news of a wave of potential fresh financings from SpaceX and Broadcom, possibly topping $100 billion, sent credit derivatives linked to hyperscalers and chipmakers climbing to prices that suggest traders see a growing risk of a default over the next five years.
For Oracle, it’s now above 20%, for SpaceX it’s about 16% — and even Nvidia Corp… is given a more than 7% risk of default in that time.”
Oracle (’36) yields surged to a record 7.62% intraday Wednesday, before closing the week at 7.57%.
Oracle yields traded below 6% in early June.
CoreWeave’s six-year yields traded to a record 12.84% intraday Friday, before closing the week up 44 bps at 12.73%.
CoreWeave yields were 147 bps higher over the past month, and up a staggering 423 bps since its June 18th issuance.
In price terms, the bond has sunk from par (100) to 83 cents (down 17%).
Meta (’36) yields rose to 6.34% Wednesday, with a Friday close of 6.20%.
This yield ended the week 40 bps higher on the month and up 85 bps over four months.
Meta bond yields traded at 5.27% on June 29th.
Wednesday had that unraveling feel.
Treasury yields rose to highs back to 2002 (UK yields 28-year high), on the back of higher crude prices (Houthis fire missiles on Riyadh); a stronger-than-expected NY Fed inflation expectations reading (3-yr high 3.90%); and from Fed minutes, “officials see another hike coming.”
Fortunately, the 10-year Treasury action was well-received (any poor auctions before the midterms?).
Ten-year yields closed the session eight bps below intraday highs.
Notably, fragile technology debt lagged, reeling from “SpaceX’s $40 Billion Financing Shock.”
October 6 – Financial Times (Michelle Chan, Sujeet Indap, Stephen Morris and Michael Acton):
“Elon Musk’s SpaceX is seeking to raise $40bn to purchase Nvidia chips as the AI and rocket group amplifies its bet on the chipmaker’s advanced technology.
The company is seeking to raise about $10bn in bank loans and $30bn in investment-grade debt to fund its blockbuster chip order, according to people familiar...
Private capital group Apollo is expected to lead the financing effort and help sell the debt to a broad base of investors.
Bond group Pimco was among a small group of lenders in talks to fund the deal…
SpaceX’s financing plans underscore the vast sums that are being raised to pay for investments in data centres, chips and other infrastructure underpinning AI.”
The FT article included a couple sentences that underscore late-cycle craziness:
“Investors whom SpaceX has previously approached about financing its multibillion-dollar chip purchase said they only received a short two-page deal memo with pictures of outer space and an arrow pointing out that the company was going to build data centres ‘somewhere in the universe'.
‘How are we supposed to take that to the IC?’ one of the people said, referring to the in-house investment committee that approves transactions.”
SpaceX’s 10-year (’36) yield jumped to 7.35% intraday Wednesday, before ending the week at 7.18%.
This bond was issued (at par) on June 26th at a yield of 5.90%.
The price closed the week at $91.
SpaceX CDS closed Thursday up 20 bps w-t-d at 200 bps, before ending the week with a one-month gain of 36 bps to 199.
Curious to see the stock up 10% over the past month, as the company’s bonds were clocked and CDS spiked.
High-yield CDS traded above 345 bps in intraday Thursday trading – which would have been the highest close since early April (ended the week at 337).
Investment-grade tech bonds were under pressure.
Notably, Amazon yields spiked intraday Wednesday high to a record 6.13% - up 48 bps over the past month.
Microsoft yields saw a midweek surge to a record 5.63% - up a notable 82 bps over the last month.
Broadcom yields traded Wednesday to a record 6.53% - up 57 bps over the previous months and 127 higher over four months.
Market debt issues proliferated from space to the sky and beyond.
FT:
“Skydance’s Debt Pile Takes the Spotlight.”
Other headlines included “Skydance’s $80 Billion Debt Bomb: How Will Warner Bros. and Paramount’s New Owner Make It Work?,” “Skydance’s High-Wire Act Begins,” “Skydance Junk Bonds Hit Fresh Lows as Broader Market Soft,” “Paramount’s Monster Debt Offers Few Safeguards to Investors.”
In the “we’ll see” category, “David Ellison: Skydance Can Pay Down Debt While Still Growing.”
Jumping another 14 bps Friday, Skydance’s high-yield unsecured (9 1/8 ’36) bond yield surged another 35 bps this week to 10.16% - with the yield rising an ominous 103 bps since the September 30th issuance.
In price terms, this bond sank from par (100) to 93.60 in only seven sessions.
I often discuss how “terminal phase excess” fuels a parabolic rise in systemic risk.
Skydance’s $100 billion of debt and literally Trillions of AI-related borrowings illuminate this dynamic exceptionally well.
I refer to “terminal” for good reason.
First, there is accelerated expansion of increasingly risky debt.
Second, late-cycle euphoria and the perception of endless liquidity ensure distorted markets and epic mispricings.
Major losses become inevitable – and the longer excesses run rampant, the greater the parabolic rise of problem Credit.
A new paradigm is upon us.
It hasn’t for a while, but risk matters today – and it will only matter more over time.
It’s worth noting that leveraged loan (LSTA Index) prices declined another 25 cents this week (65 cent 3-week drop) to 94.97 – now only two cents from the low back to April.
It has quickly become an inhospitable environment for high-risk borrowers seeking new funds or needing to refinance.
Surging borrowing costs are shattering scores of business models.
The good news: high yield spreads, closing Friday at 308 bps, remain 140 bps below “liberation day” spike levels from April 2025.
Typically, in acutely unstable markets, safe haven buying and sinking Treasury yields are a significant force behind widening corporate spreads.
The bad news: because of the systemic nature of the unfolding Credit crisis, Treasury and sovereign yields have risen substantially.
Spreads don’t tell the story.
At 8.24%, high yield spreads are today only 34 bps below April 2025 spike highs.
On the one hand, levered corporate “carry trade” positions (i.e. short Treasuries to fund higher-yielding corporate debt) have so far been supported by sinking Treasury prices (rising yields).
Only marginal deleveraging has materialized thus far.
On the other hand, significant losses have been endemic throughout the corporate Credit investor universe.
When deleveraging does begin in earnest, it will rock an already impaired marketplace.
And the upheaval will likely be systemic rather than idiosyncratic – Treasuries, MBS, corporates, munis, and structured finance.
Moreover, systemic U.S. deleveraging surely means global de-risking/deleveraging.
It was curious to see money market fund assets surge $72 billion last week to a record $7.964 TN – along with a chunky $79 jump in outstanding commercial paper.
That this monetary inflation corresponded with a rally to new highs in tech stocks (i.e., Nasdaq100 and MAG7) is not surprising.
Money fund assets at record highs support the thesis of only modest deleveraging.
It also suggests that big tech evolved into a powerful bastion of levered speculation and liquidity creation (i.e. options, ETFs, margin debt, and derivatives).
But even marginal deleveraging coupled with massive AI-related borrowings has triggered tighter financial conditions.
And this dynamic has spurred a powerful rush to the markets to issue debt before illiquidity becomes a serious issue.
SpaceX… “SoftBank Seeks $100 Billion From Gulf States for AI.”
With investors already nursing substantial losses, it’s unclear who will have the appetite to accommodate what will surely be an even more intense push to raise funds for the global AI arms race.
It has become a deeply systemic issue: Deleveraging and the AI Bubble are irreconcilable.
This is a serious dilemma for the S&P500 and equity index universe, for bond indices and ETFs, for Wall Street structured finance, for bank loan portfolios, and for the U.S. and global economies.
Bond markets are sending a signal that deserves to be taken seriously.
October 9 – Bloomberg (Julia Fioretti, Sangmi Cha and Manuel Baigorri):
“In 48 hours, Firmus Grid Ltd. went from a rising star of the artificial intelligence data center boom to one of the biggest deal flops in recent memory.
After months of touting itself as an essential artificial intelligence play, the Nvidia Corp.-backed company was on the cusp of pulling off one of Australia’s biggest-ever initial public offerings of up to $5.5 billion, and at a valuation fit for a top 10 player in the industry.
But that story wilted under the scrutiny of US fund managers that bankers were counting on to buy a chunk of the deal.”
0 comments:
Publicar un comentario