Expect Trouble
Doug Nolan
“Terminal Phase Excess” is fundamental to my Credit Bubble analytical framework.
Bubbles are sustained by ever increasing amounts of Credit, expanding purchasing power with myriad inflationary manifestations.
Of particular significance, rising asset prices and speculative Bubbles intensify over time.
Late in the cycle, monetary disorder takes hold — and systemic risk goes parabolic.
This destructive phase sees rapid growth of increasingly risky loans.
Expanding speculative leverage fuels unsustainable “blow off” market price inflation.
Moreover, the whirlwind of risky lending and leveraging ensures catastrophic end-of-cycle resource misallocation and structural maladjustment.
Mechanisms to intermediate skyrocketing risks are pushed to the limit.
Things don’t work well in reverse.
As I’ve discussed repeatedly over the years, Bubbles fueled by risky Credit do not generally pose major systemic risk.
A boom financed by corporate high-yield bonds, for example, won’t get too carried away – or for too long - before a market “no more junk!” revolt terminates Bubble excess.
The great global government finance Bubble continues to illustrate the precarious nature of Bubbles financed by perceived safe and liquid “money” and “money-like” debt instruments.
The uniquely insatiable demand for “money” presents a potentially catastrophic risk of uncontrollable booms and crises of confidence.
I’m again highlighting this analysis, because it’s increasingly pertinent.
The walls have begun to close in.
The historic AI-related debt boom is seeing associated debt instruments, especially at the margin, lose “moneyness.”
Meanwhile, the unending massive issuance of Treasuries and government debt internationally is facing increasing market pressure – notably waning demand and higher yields.
From Washington to Wall Street – along with Tokyo, Paris, London, Beijing, Seoul and beyond – too many are pushing the envelope to sustain increasingly fragile booms.
It’s been an illuminating six trading sessions.
Thursday August 6th: closing 10-year Treasury yield 4.68%.
The following Friday: “Weak US Payrolls, Higher Layoffs Reduce Fed Rate Hike Bets.”
Yet the unexpected loss of 23k jobs generated a modest three bps decline in yields - to 4.65%, with 10-year yields back up to 4.71% by Monday’s close.
After closing Tuesday at 4.69%, Wednesday’s “US Core Inflation Comes in Subdued, Easing Pressure on the Fed” was disregarded, with yields ending the session at 4.70%.
Thursday’s “Tame US Producer Prices to Give Fed Doves More Cover” supported a yield decline to 4.64%.
But even Friday’s “US Retail Sales Fall Most Since May 2025” couldn’t hold yields back from ending the week at 4.69%.
That yields ticked up in the face of a slew of typically constructive data corroborates the emergence of a troubling new bond market trading paradigm.
The recent backup in yields is more than justified.
WTI jumped $4.22 this week to $82.40.
With President Trump apparently digging in for economic pressure to force Iran’s hands – and Hegseth saying an Iranian blockade could be maintained “indefinitely” – energy markets adjusted to the possibility that passage through the Strait of Hormuz might be restricted for months.
Even the December 2027 crude futures price jumped $2.20 to $70.52 – versus the $60 pre-war level.
The Bloomberg Commodities Index advanced 2.8% this week. Despite last month’s somewhat encouraging data, the risk of upside inflation surprises remains elevated.
Beyond inflation issues, fragile bond markets face unrelenting massive supply.
August 12 – Reuters (David Lawder):
“The U.S. federal budget deficit for July jumped to $432 billion, a record for the month, as higher outlays and more negative tariff revenues also brought the 2026 fiscal-year-to-date budget gap to $1.799 trillion, topping the full fiscal 2025 deficit with two months left in the fiscal year…
Last month’s budget gap, which was partly inflated by calendar shifts in benefit payments, was $141 billion, or 48%, higher than in July 2025 and was the largest monthly deficit since March 2021, when it hit $660 billion due to COVID-19 relief program spending.
There have been only two other higher monthly deficits: $864 billion in June 2020 and $738 billion in April 2020…
A Treasury official said unadjusted outlays for July were also a record for that month at $766 billion, up $137 billion, or 22%, from a year ago.
But the month’s outlays were inflated by $99 billion due to the payment in July of many August 2026 benefits because the current month started on a weekend.
Accounting for these and other calendar shifts brought the adjusted July deficit to $333 billion, up $50 billion, or 18%, from the prior year...”
Interest expense on Treasury Debt rose to $118 billion last month, second only to the $251 billion of expenditures at the Department of Health and Human Services (i.e., Medicare and Medicaid).
Department of Defense expenditures came in at $86 billion.
It’s worth noting that Interest expense was $40 billion in July 2019.
Through 10 months of the 2026 fiscal year, debt service of $1.170 TN ran 14.5% ahead of comparable 2025.
The full-year deficit is poised to exceed $2.1 TN (6.5% of GDP), which would be 17% ahead of last year.
August 13 – Financial Times (Myles McCormick and Kate Duguid):
“The US has paid the highest borrowing costs to sell 30-year bonds since 2001…
A $25bn Treasury auction of 30-year bonds on Thursday drew yields as high as 5.22%...
It marked the highest yield since the 5.52% paid in August 2001, after which 30-year auctions were suspended for almost five years.
The yield on Thursday’s auction compares with 5.06% at the previous 30-year sale in July, and 4.91% just before Trump’s second term began in January 2025.
The jolt higher in borrowing costs comes as the US debt pile has swollen to almost $40tn, pushing the debt-to-GDP ratio towards an all-time high.”
August 13 – Bloomberg (Jonnelle Marte):
“Federal Reserve Bank of Cleveland President Beth Hammack said she’s monitoring three things when it comes to US financial stability, including leverage in the Treasuries market.
‘We have a high amount of debt outstanding, and right now there’s a lot of leverage that’s being used to buy that debt,’ Hammack says…
‘The purchasers of that debt are actually borrowing funds, and so that can create some instabilities in the system when you think about that.’
‘If you look at Congress, I don’t get a sense of a lot of fiscal discipline coming back in.’ Growth in private credit is another area to monitor, she said.”
The VIX (equities volatility) Index ended the week at a carefree summer yacht rock 14.25 – the lowest close since December.
The MOVE (bond volatility) Index declined more than 2 points this week to a lowly 69.6 (5-yr avg. 99.2).
The High Yield Volatility Index sank 16 to below 92 (lowest close since January) – and half the five-year average (187).
JPMorgan CDS dropped to a near one-year low 37.9, with most bank CDS ending the week not far from one-year lows.
Emboldened by Bessent’s yen intervention gambit, equities and risk markets signal “the fix is in” through at least the midterms.
Treasuries are sensibly uncomfortable.
And this week, in particular, global bond market trading suggested incipient recognition of mounting deleveraging and liquidity risks.
The problem children were viewed with wary eyes.
With its own budget issues, France’s 10-year yields surged 13 bps this week to 4.05% - the high all the way back to June 2009.
The spread versus German yields widened six to 85 bps, the widest level since early October.
Japan’s 10-year yield jumped seven bps to 2.87% - the high back to 1996.
UK yields surged 12 bps to 5.04%, within 13 bps of highs back to pre-crisis 2008.
Governments have company when it comes to recklessly vociferous borrowing appetites.
August 13 – Financial Times (Robin Wigglesworth):
“On Monday, Alphaville explored just how big the lease commitments of the AI hyperscalers have become as their financing approach shifts from plain vanilla bonds to more creative avenues.
Goldman Sachs analysts had scoured through the footnotes of the hyperscalers’ regulatory filings and counted $1.5tn of lease commitments, of which $1tn hadn’t started yet and therefore didn’t appear in their financial accounts as conventional liabilities.
As those analysts obliquely noted: ‘From a credit perspective, this treatment can understate leverage and future liquidity needs as these obligations are eventually recognised and contractual payments come due.’
We also threw in an interesting titbit from an earlier Morgan Stanley report from July, which also toted up the purchase commitments of Alphabet, Microsoft, Amazon, Nvidia and Oracle.
These are typically contractual obligations to buy chips, compute, electricity to power data centres, and other equipment, and came to another $982bn at the end of the first quarter.”
Especially with global yields marching higher, ever-inflating AI borrowing requirements (stoked by briskly inflating costs) will prove increasingly challenging.
Central to my analysis throughout the mortgage finance Bubble period was the concept of “Wall Street alchemy” – the transformation of increasingly risky mortgage Credit into perceived (mostly “AAA”) “money-like” instruments.
It was risk intermediation to behold – that is until “terminal phase” crazy saw 2006’s fateful $1 TN of subprime derivatives.
August 11 – Wall Street Journal (Jack Pitcher, Anissa Gardizy and Peter Rudegeair):
“Nvidia CEO Jensen Huang is running into a problem: Many of his customers can’t afford to buy his company’s coveted AI-powering chips.
That explains why Huang teamed up with an array of Wall Street firms on a $500 billion plan that will theoretically standardize chip financing, creating asset-backed pools of capital for AI companies—while leaving Nvidia partly on the hook if things go wrong.
Executives involved in the strategic partnership Huang unveiled this week with Apollo Global Management, BlackRock, Blackstone, Brookfield Asset Management, Goldman Sachs and KKR tout the effort as the launch of a new asset class akin to the securitization of everything from airplanes to credit cards to mortgages.
To critics, it is a system that will cover up weaknesses in some corners of the AI marketplace.
The concern isn’t about tech heavyweights—Meta, Microsoft, Google—with fortress balance sheets.
Instead, it is about the many smaller AI labs, cloud companies and enterprises that have ravenous demand for Nvidia chips but face high interest rates if they want to finance purchases of them.”
I suppose a $500 billion scheme specifically to finance Nvidia chip purchases qualifies for “Expect the Unbelievable.”
But the craziest part of this so far is that it’s not dismissed as crazy late-cycle phenomena.
Using overpriced chips (at peak mania) as loan collateral could indeed work miraculously to sustain the boom.
But gosh.
For a system already acutely vulnerable to any sudden tightening of financial conditions, Wall Street is just hankering for trouble.
August 12 – Financial Times (Antoine Gara, Michelle Chan and George Hammond):
“Wall Street is betting that AI chips can defy one of finance’s basic rules: that fast-moving technology quickly loses its value.
Nvidia unveiled this week a $500bn deal under which tech groups will be able to lease semiconductors with financing from groups including Apollo Global, KKR, Brookfield, BlackRock and Goldman Sachs.
The blockbuster partnership was underpinned by expectations that chip prices would remain higher for longer than many analysts had anticipated as a result of the clamour to secure the components powering the AI boom, finance executives involved in the deal told the FT.
Nvidia chief Jensen Huang has said the pact will create a new asset class underpinned by chips, which would be ripe for investment from the $22tn private capital industry.”
Using homes, whose prices only go up, as collateral for risky loans intermediated through sophisticated Wall Street instruments and vehicles (i.e., ABS, CDOs, “CDO-squared”, synthetic CDOs, SIVs, etc.) worked miraculously - until subprime mortgage finance suddenly tightened and home prices tanked.
A likely scenario: markets are slammed by a bout of de-risking/deleveraging, with risk aversion and illiquidity forcing an abrupt reassessment of prospects for the multi-trillion AI arms race buildout.
The chip price collapse will make the 2007/08 Phoenix home price deflation look like a walk in the park.
I have a hard time believing investors will today be willing to pony up a half Trillion to such a high-risk proposition.
It’s worth noting that Nvidia CDS gained another three this week to 75.5 bps, after trading below 40 in early June.
This $500 billion Nvidia plan is the latest example of a wildly emboldened Wall Street exuberantly playing with fire.
August 10 – Financial Times (Robin Wigglesworth):
“You may have noticed that the hyperscalers have become increasingly inventive in how they raise the money required for a titanic series of data centres being built around the world.
The first example of what this looks like was the record-breaking bond sale for Meta’s ‘Hyperion’ data centre in Louisiana, which Alphaville explored in depth here.
In short, instead of just issuing bonds off its own balance sheet, Meta formed a joint venture with Blue Owl called Beignet that would develop and own Hyperion.
Meta owns just 20% of Beignet, but made a rock-hard commitment to lease Hyperion for at least 20 years.
That guarantee allowed Beignet to issue an amortising $27bn bond, but this debt doesn’t actually appear as debt on Meta’s balance sheet, even if it is on the hook for the payments.
Anyway, a lot of the other hyperscalers seemed to think that this was a tremendous idea, and have since explored their own increasingly creative ways of raising a lot of money while limiting the optical impact on their balance sheets.
And various lease structures vaguely along the lines of Beignet are the favoured way of doing so.
But just how meaningful are these non-debt financial obligations?
Fortunately, Goldman Sachs’ analysts have gone through all the fine print for us, and totted up a massive $1.5tn of lease commitments, of which about $1tn doesn’t appear in the financial statements of the hyperscalers.”
It’s tempting to group Friday afternoon’s Jane Street news into the “Expect the Unbelievable.”
But it’s only too believable.
August 13 – Financial Times (Jill R Shah and Joshua Franklin):
“Jane Street posted a roughly $15bn loss in July after turmoil at AI-focused hedge fund Situational Awareness wrongfooted the US trading firm.
The… firm disclosed the figure to lenders as part of a deal to shift its roughly $11bn public debt pile to private investors including Pimco, according to people familiar…
The loss represents a rare setback for a secretive trading shop that has in recent years become a dominant player in global markets.
Jane Street has generated more than $40bn in net trading revenues in the year to Friday, even accounting for the July loss, which exceeds its entire haul for 2025…
The drawdown in July came during a tumultuous month for US AI stocks, which fell sharply in a powerful reversal of a rally that carried them sharply higher throughout most of 2026.
The sell-off also hit several AI-focused hedge funds including Leopold Aschenbrenner’s Situational Awareness, which counted Jane Street among its investors.”
August 14 – Financial Times (Toby Nangle):
“Jane Street came to market this week with a $14.6bn multi-tranche monster bond issue.
The lion’s share was used to refinance existing debt.
Nothing unusual about that: most new bonds are issued to repay existing debt.
However, the proprietary trading firm famed for both its financial acumen and fastidious secrecy paid through the nose to retire existing debt that had no business being retired.
Why?
MainFT… wrote (our emphasis) that: ‘[a] shift towards private markets would… allow the proprietary trading firm to limit disclosures on its financials, which it currently reports quarterly to a large group of debt holders.’”
A two-year-old hedge fund rides highly levered AI bets to inflate assets to $45 billion, only to get hit with a 67% loss in July – forced deleveraging that required liquidating positions to Ken Griffin’s Citadel hedge fund.
In all the chaos, highly levered Jane Street loses a cool $15 billion, only to then refinance a huge chunk of debt in the private debt market.
The S&P500 closes Thursday at a record high, with the VIX ending the week below 15.
Now that chain of events qualifies as “Expect the Unbelievable.”
Throw in rising global yields in highly leveraged bond markets, and it seems we’re witnessing enough to Expect Trouble.
China’s latest Credit data deserves a brief mention.
Following typically booming Junes (final month of the quarter), Julys are usually slower months for Chinese lending growth. Last month was notably weak.
New Loans contracted a record $51 billion (vs. July ‘25’s $7bn contraction), with one-year growth of $2.045 TN down 21% from comparable 2025.
Corporate loans declined a record $19.2 billion, the first contraction since July 2016’s less than $1 billion decline.
This reduced y-t-d Corporate loan growth to $356 billion, 19% below comparable 2025.
One-year growth slowed to 8.2%, matching the weakest reading since April 2017 (itself the weakest growth in at least a decade).
Consumer (chiefly mortgages) loans contracted another $68 billion, causing a one-year decline of $18 billion (1.3%).
It’s worth recalling that years of double-digit consumer loan growth ended with the bursting of the apartment Bubble back in 2022.
Meanwhile, Total Aggregate Financing mustered a respectable $179 billion gain in July (to $68.7 TN), outpacing July ‘25’s $150 billion.
While down y-o-y, one-year growth of $4.75 TN indicates massive ongoing Credit excess.
Leading the charge, Government Bonds expanded $196 billion in July, up about $10 billion from July ’25 to $15.23 TN.
In the year’s first seven months, Government Bonds surged $1.15 TN, with one-year growth of $1.882 TN.
Outstanding Government Bonds have surged 39% in two years and 61% in three.
In addition, with Q1 data available, Chinese bank assets surged $2.18 TN during the first quarter to a record $73.39 TN – with one-year growth of a blistering $5.40 TN (7.9%).
Bank assets inflated 15% over two years, 25% over three, 50% over five, and 137% over ten years, in one of history’s most spectacular Credit booms.
China bank assets ended 2008 at $9.3 TN.
A crisis of confidence in Chinese finance is overdue.
0 comments:
Publicar un comentario