WTF Just Happened To Global Markets

Global Finance News
11 followers
#finance
#financement
Yes, the stock market got ugly, but the real story is the dramatic shift in bonds.
The Nasdaq just plunged more than **5%**, falling into correction territory. The S&P 500 dropped **4%**, heading for its worst day since April 2025. The Dow lost more than **1,650 points**, erasing almost its entire post-election rally in a single session.
And the epicenter was artificial intelligence.
But this is no longer only a technology selloff.
AI Generated·May contain errors
Everybody thinks this is about AI stocks, but now we've got the bond
market that's actually in rally mode on top of everything else.
But the stock market, that's got today's attention.
The NASDAQ just plunged more than 5%,
falling into correction territory.
The S&P 500 dropped 4%, heading for its worst day since April 2025.
The Dow lost more than 1,650 points as of this recording,
erasing almost its entire post-election rally in a single session.
Again, the epicenter seems to be artificial intelligence.
Broadcom and NVIDIA fell approximately 9%.
Oracle dropped 8%.
AMD and Micron lost around...
15% each.
These are just one-day totals.
Semiconductor ETF fell roughly 11%,
putting it on track for the worst day
since the COVID outbreak in March of 2020.
But this is no longer just a technology sell-off,
at least today.
Caterpillar fell 5%,
JP Morgan dropped 4%,
Boeing lost 4%,
Nike declined 3%.
But at the same time,
maybe more importantly,
treasury bonds are rallying.
Not just a fluctuation,
we're seeing a bond market rally develop.
The two-year yield has fallen to a recent low near 4.48%,
down almost half a percentage point
in five weeks. The 10-year is around 483,
heading toward 480, roughly 50 basis points below
its recent high. September's technical effects are gone,
just as I told you they would be.
And the market is increasingly free to price
the fundamentals. And the fundamentals
are becoming more concerning.
So the immediate explanation for today's stock crash
is that the midterm election results may
make data centers harder to build.
Investors are worried about more regulation, environmental
restrictions, and even proposals for a national moratorium.
But that is only the proximate cause.
It is one more concern, perhaps one too many concerns,
piled on top of anxieties that have
been building for months. The spending was already enormous.
The financial commitments
We're becoming increasingly difficult to justify credit markets.
They've been resisting the weakest AI borrowers.
Too much had to go perfectly
to support current stock prices and valuations.
Now the path to perfection looks increasingly less likely.
And bonds are not merely pricing an AI bust.
They're pricing energy-driven demand destruction,
weak economic fundamentals,
and the possibility that if the AI boom does turn into a bust,
which we know it will at some point,
That will remove one of the few major sources of investment,
keeping the economy afloat.
We all know this boom will eventually break.
They all do.
What we don't know is when and how violent
that will be when it happens.
But that is always in the back of everyone's minds,
especially those who are investing
in stock markets as well as the bond market.
And so because it's always there,
it's always something you have to keep in mind and think
about.
The reactions can be violent in the short run
because everybody knows how this ends.
We just don't know when,
which means you really got to pay attention to all of these signals.
Now, again, the immediate catalyst
for today's sell-off is the possibility that America's
data center expansion will become politically more difficult.
Following a Democratic sweep of both chambers of Congress
and victories in several important
gubernatorial races, investors are now reassessing
the regulatory outlook for AI infrastructure.
Senator Bernie Sanders, the crazy guy from up there in Vermont,
called for a national data
center moratorium, arguing that Americans don't
want technology companies damaging the environment,
eliminating jobs, and threatening the public's future.
The national moratorium is not currently law
It may never become law.
That isn't the point.
Financial markets have spent
the last several years assuming data centers could be constructed
as quickly as financing, as quickly as equipment,
and as quickly as power became available.
Political resistance was treated as a manageable local inconvenience.
That assumption is now being increasingly challenged.
There is always a correction.
Data centers require enormous amounts of electricity and water.
They need transmission lines, backup generation, cooling systems,
land connections that are already straining the power grids.
These requirements are becoming increasingly visible to voters.
Communities are being asked to tolerate higher electricity demand,
new infrastructure, and potentially higher utility bills.
The politics were always going to become more difficult.
Investors simply weren't pricing that difficulty
because the AI narrative assumed infrastructure
would be built almost without limitation.
Today, along with some of the stuff in the credit markets,
the markets are beginning to price limitations.
Neo cloud companies were hit particularly hard.
Coreweave, that's one that's been in the news a lot lately,
fell approximately 13%.
Nebius lost 16%.
Iron dropped 17%.
Not just today, it's been coming for some time.
And these companies are far more dependent
on continuous AI construction
than say Microsoft, Amazon, and Google.
They don't have gigantic advertising,
retail, and software businesses
that are capable of subsidizing years of uncertain returns.
So if data center projects are delayed,
their revenue is delayed.
If projects are canceled,
their expected revenue path disappears
while financing obligations they've already taken on,
those remain.
And if lenders conclude political risk has
made future construction less predictable,
the cost of financing rises before
a single project is officially stopped.
That is why one proposed moratorium can have
a much larger financial effect
than its immediate probability would suggest.
It exposes just how many valuations depended
on almost everything going exactly right
and how that was never a realistic possibility.
There's a big mistake that a lot of gold investors
are making right now. It has nothing to do with
the price of gold. You might have bought gold five,
ten years ago, watched the price go up and
thought, I've done pretty well.
But here's the question almost nobody asked. How many more ounces
of gold could you own today if your gold had
been earning you more gold this entire time?
Think about it.
Most people buy gold, pay to store it and insure it,
and then just sit there waiting for the price to go higher.
Your gold isn't producing anything.
That's what makes monetary metals interesting.
They've created a way for eligible investors
to earn up to 4% annually on their gold
through their gold yield marketplace.
But here's the important distinction.
That yield is paid in additional physical gold, not dollars.
So instead of only watching the dollar price of your gold change,
you have the opportunity to increase the amount of gold that you own.
And those additional ounces can potentially appreciate
alongside your original holdings if gold prices rise.
Now, Monetary Metals has been doing this for nearly a decade,
and they've paid out thousands of ounces
of precious metals to their clients.
So here's the question.
If you're already holding gold for the long term,
Why wouldn't you at least investigate
whether it could be earning you more gold?
The real opportunity cost may
not be the price that you paid for your gold.
It could instead be all the ounces that you're not accumulating.
Go to monetary-metals.com and learn how
their gold yield marketplace works.
If this was only a one-day re-reaction
after yesterday's non-reaction reaction,
we wouldn't be seeing what we're seeing in the bond market.
The bond market, the treasury market in particular, is now rallying.
It's not just a fluctuation, not just a short-run blip.
This is an honest-to-goodness treasury market rally,
which has much broader implications,
particularly since it began long before
the election results on Tuesday.
As of last check, the two-year yield is now at 4.48%,
down almost half a point from its
recent high, at 10 years around 480 or so,
approximately 50 basis points below its peak.
This matters a great deal
because the earlier rise in yield had two main components to it.
The first was, as I pointed out to you several times, September.
Treasury markets often encounter seasonal pressure
at the end of the third quarter
as dealer balance sheets become constrained
and intermediaries become less willing to warehouse
duration.
This year, those pressures amplified an already difficult sell-off.
The second component, the primary component, was the Fed.
Following Kevin Warsh's Jackson Hole remarks,
markets began pricing the risk
The Fed would respond aggressively
to an energy-driven increase in headline inflation,
even though expensive energy was simultaneously weakening
the real economy underneath.
That pushed expected short-term rates higher
and farther into the future.
September is now over, which removes the seasonal pressure.
Therefore, the September effects are now fully into,
well, I guess it's not, thank God it's
October.
Thank God it's October, now November.
More importantly, though.
The economy is beginning to limit how much room
the Fed has to overreact and tighten itself.
The worse the economic outlook becomes,
the less credible an extended series of rate increases also becomes.
The upward portion of the forward rate curve,
that's termed so for futures,
it has been compressing hard because traders are pricing less room
for the Fed to continue hiking before something really breaks.
That process accelerated during today's stock market sell-off.
The probability in options markets
of a quarter-point increase in December from the Fed
fell to approximately 70%, down from nearly 83% just yesterday.
The probability the Fed does nothing
in December increased to almost 30%.
Markets are still pricing a meaningful risk of another truce,
another rate hike out of policymakers,
but the Fed's margin for doing so is shrinking.
And the reason it is shrinking is because the market
is becoming less confident in the economy's ability
to absorb energy expenses, a worsening credit cycle,
and now a potential reversal in AI investment all at the same time.
Flight to safety, lower growth and inflation expectations,
less interference from the Fed.
Lower bond yields.
The treasury market is now in rally mode.
The data center backlash that we're now seeing emerge
didn't create the AI industry's financial problems
that we've been talking about for a lot longer
than just this past week.
In fact, it became harder and harder to ignore
by the time we got to the election on Tuesday.
So again, maybe one straw too many
for the camel to keep going.
Amazon recently spent more than $77 billion on property
and equipment in a single quarter,
contributing to a free cash flow outflow
of approximately $42.4 billion.
The company then established supplier financing arrangements
that could allow it to borrow as much
as $80 billion against the hardware it purchases.
Alphabet faces more than $8 trillion in reported future payments
connected to data centers, computing capacity, and energy.
Even as investors continue asking how generative AI will be monetized
without damaging Google's existing search advertising businesses.
Microsoft offered voluntary buyouts to approximately 60,000 employees
as it seeks additional flexibility
while continuing its own enormous AI infrastructure program.
These companies, these specific ones I just went over,
are not speculative startups.
They are among the most profitable corporations in history.
Yet even they are being forced to conserve cash.
Find new financing structures
and prioritize AI investments over all other kinds of expenses.
And below them, a much weaker layer of neoclouds,
data center operators, and AI laboratories
without profitable independent businesses.
They need financing before revenue even exists.
Lenders are being asked to fund GPUs,
the networking equipment, the land purchases,
and all the power associated with everything based
on projections of what customers may
eventually pay after the infrastructure is completed.
That was becoming a difficult proposition
long before we got to the election.
Banks nearly backed away
from a reported $63 billion financing package for Broadcom chips
and networking equipment that was intended for Anthropic.
Lenders demanded higher interest rates
and stronger protection against those losses.
Broadcom reportedly had to provide residual value support
because the banks weren't comfortable
relying entirely on the finance chips as collateral.
It had told us something very important
before we got to the election this week.
There's always a collection.
The same chips described as the most valuable
and strategically important assets in the world
apparently aren't sufficient collateral
on their own without additional guarantees from Broadcom,
the supplier. Why?
Because uncertainty is enormous and getting bigger. Add political
and construction risks to that equation
and financing becomes even more difficult. The
election didn't create the wall.
It may have pushed the industry into a wall that was already
in front of it.
You've got to keep in mind, too, that this AI possible bust,
we aren't there yet, but the possibility
of an AI bust adds to an already difficult situation.
That's what the bond market has been pricing,
even though everybody was hyper-focused on the September sell-off
and Kevin Warsh and the Federal Reserve's role in that sell-off.
The Fed has been hiking rates for CPI pressures that it sees
and therefore imagines down the road
while the bond market continues to price all
the negatives in the real economy that exist
and have evidence for, and AI would just be one more of those.
It is much broader than just strictly technology.
The energy shock has been described primarily as an inflation problem.
That interpretation ignores what expensive energy does
when incomes and credit aren't
growing nearly fast enough to absorb it.
Households can't create money because gasoline,
electricity, and now their utility bills are
going to cost a lot more.
If consumers have to spend an initial $100 on fuel,
that's $100 they can't spend at
restaurants, retailers, or other discretionary expenses.
One category's higher price becomes another category's lost revenue.
That's what the bond market, especially tips,
have been explicitly pricing this entire time.
Businesses face the same constraint, and so that's in bond yields too.
It really bonds spreads.
Higher transportation, electricity, production costs,
they reduce margins unless they can
pass those costs on to their customers.
But customers already face higher household expenses,
therefore they can't accept another
round of price increases.
That's what we keep hearing all across the economy.
Without rapid wage growth, which, let's face it,
is not happening, expanding credit or
Or some other broad increase in income, the energy shock.
Can't generate an endless inflationary spiral.
It destroys volume.
That's why gasoline demand
and other measures of discretionary activity are such a big part
of it.
If fuel consumption weakens, for example,
while energy remains expensive, the economy
is revealing that demand is being rationed.
Treasuries are really recognizing the higher likelihood
and maybe even a sooner arrival
of that destruction.
Falling yields don't mean the energy shock has disappeared,
they mean the market increasingly
believes the economy can't tolerate the shock without,
at the very least, slowing down more
than it has. That also explains why today's broad stock decline
is a big part of it too.
Caterpillar, JP Morgan, Boeing, Nike,
they aren't direct AI infrastructure trades at all. Their
declines demonstrate that investors are moving beyond
a narrow reassessment of semiconductor valuation
Caterpillar reflects investment in construction.
JP Morgan, credit in the wider financial system, the credit cycle.
Boeing, industrial demand and financing conditions.
Nike, consumers, but a lot more trouble there at Nike.
When all of them fall alongside technology,
the market is no longer only asking whether NVIDIA's valuation,
for example, is too high.
It's asking whether the economic
foundation beneath the entire marketplace
is weaker than previously believed,
or really weaker than previously told,
Or really, really, really.
And believe it or not, the United States
is not the only market to suffer these types of questions.
In fact, the Chinese marketplace, the Chinese tech marketplace,
the Chinese position in the AI bubble has already
been far more advanced than what we're seeing here in the U.S.
In fact, China's technology bubble may have already burst.
The Star 50 index of major Chinese technology
and semiconductor companies
has fallen roughly 40% from its mid-year peak,
destroying approximately half a trillion
U.S. dollar equivalent in market value.
China's broader CSI 300 dropped around 8.5% in October,
its worst monthly decline in more than a decade.
Technology and AI shares have become
an enormous portion of the market capitalization
and trading activity, and once those stocks reversed,
Very few investors escaped the damage.
Beijing has, of course, attempted to stabilize the market.
Authorities announced another round of central bank swap operations,
with securities firms and fund managers applying
for more than 100 billion yuan
for approximately 15 billion US dollar equivalent of liquidity support.
Financing costs for share buybacks were reduced,
and major technology companies
were encouraged to repurchase their stocks.
So far, the measures haven't created
a durable recovery in the marketplace.
And that's...
It's a warning American investors
and global investors overall should heed.
Now, liquidity from any source can slow a decline,
share buybacks can create temporary demand,
and government support can prevent disorderly selling,
but none of those measures are going to be able to manufacture
a profitable end-demand state for AI services.
China's sell-off was also intensified by its weak domestic
economic situation, its prolonged property downturn, but also,
maybe this is a point we should all pay most attention to,
disappointing technological progress
and concerns about restrictions on Chinese equipment.
The exact catalyst was less important than the accumulated imbalance.
Valuations required sustained technological progress.
Rapid demand growth forever into the future extrapolating
and continued support from all parts of the Chinese system.
When investors became less certain about all three,
prices fell much faster than the underlying story, which had changed.
And this is where bonds and technology stocks finally converge.
The AI capital expenditure boom has been one of the few,
one of the few legitimate economic tailwinds for the United States
and really around the rest of the world.
Just look at global trade
and how much that has been this year in particular,
really boosted by the technology sector.
So today's stock crash is being blamed
on the political fallout from the election and the
possibility that building data centers
would become politically more difficult.
And that is technically correct,
but is also an incomplete explanation for everything that
we're seeing here. Political resistance is only the latest concern.
The industry was already
consuming extraordinary amounts of cash.
Financing structures were becoming increasingly circular.
Lenders were demanding stronger guarantees.
Credit protection costs were already rising.
Chinese AI stocks had already collapsed.
Too many assumptions had to remain true simultaneously.
Data centers had to be approved.
Power had to become available.
Chips had to retain their value.
Financing had to remain cheap.
AI demand had to expand exponentially and continuously.
Customers had to accept profitable prices.
And the broader economy had to remain resilient enough
in truth to support all of it.
Markets are now seeing a narrower path to that perfect outcome.
Stocks are pricing. Stocks are pricing.
The direct threat to earnings and valuations,
at least the narrative about the future.
Bonds are pricing the broader economic consequences.
Energy-driven demand destruction.
The energy shock hasn't gone anywhere. Weak real economy
potential. Tighter credit from the credit cycle downturn
that downturns even more. And the
possibility that an AI reversal removes one
of the economy's largest remaining investment engines.
The bubble hasn't burst yet. Not fully.
Not in the United States, at least.
But the warning signs, the warning signals in more places,
they are beginning to pile up.
And when both stocks and bonds begin sending the same message,
it's much harder to dismiss
this as a post-election hangover for a single daily crash.
The market is beginning to ask,
what are these AI stocks actually worth?
The wider marketplace is beginning to wonder what happens
to the broader economy if that worth never gets realized.
