sigeru retweeted
Oh mon dieu...
The Nasdaq during the Dotcom crash. 2000 to 2002.
Total decline: −78%.
Along the way:
+35% rally
+12% rally
+25% rally
+41% rally
+45% rally
Every single one felt like the bottom.
Every single one was a trap.
The +45% rally was the cruelest.
It came right before the final collapse.
This is the anatomy of a bear market.
Violent rallies that shake out the shorts.
Convince the bulls to re-enter.
Then resume the decline.
sigeru retweeted
BREAKING: WTI OIL is down almost 3% in the last 2 hours on reports that the US and Iran have given preliminary approval to a proposal to end the war.
sigeru retweeted
Treasury Secretary Scott Bessent has said that mortgage rates and bond yields will come down once the Iran conflict ends.
sigeru retweeted
The SINGULAR aspect of the record low dating back to the series inception in 2001 in @realtors September Pending Home Sales is every market & every major metro saw sales declines.
If Warsh intends to normalize the US housing market, he’s doing one heck of a job.
sigeru retweeted
Big reversal in crude oil today on rumours US/Iran conflict to end soon.
Stocks seem to be close to a tactical low, on watch for year-end rally.
sigeru retweeted
Hedge funds are DUMPING AI stocks, while retail investors pour billions into leveraged semiconductor ETFs:
Hedge funds sold global AI stocks for 5 straight sessions through October 14th.
In just one week, they reversed all their cumulative net buying since the start of 2025.
They also dumped the technology sector at the fastest pace in at least 10 years.
Furthermore, hedge funds cut net leverage to 43.7%, its lowest level in 5 years, but gross leverage jumped to 213.9%, as managers increased short positions rather than significantly reducing their long exposure.
Hedge funds are pulling back as concerns over AI financing intensify, with AI-related high-yield credit spreads widening +50 basis points to 371 bps, their highest in at least 12 months.
Moreover, 9 AI data-center bonds now yield above 10% and 4 exceed 14.75%.
The funding burden is set to grow, with Goldman projecting $720 billion in AI project financing and direct hyperscaler debt issuance in 2027.
Notably, AI capital expenditure already accounts for ~50% of S&P 500 earnings growth.
Meanwhile, leveraged ETFs attracted nearly $10 billion in weekly inflows, including a record $7 billion into the 3x semiconductor ETF, $SOXL, and another $1.3 billion into the 3x Nasdaq ETF, $TQQQ, driven by retail investors.
Hedge funds are dumping while retail investors are buying. Who will end up being right?
sigeru retweeted
US electricity prices are skyrocketing.
The average price of US residential electricity has risen +11% since January 2025, to 19.6 cents per kilowatt-hour in August 2026, near an all-time high.
Average prices during the first 8 months of 2026 were +40% higher than in the same period of 2019.
The largest increase in residential electricity prices over this period was in Maine, at +90%, followed by California and New York, at nearly +70%.
This comes as the AI boom is intensifying electricity demand, with data centers requiring enormous amounts of electricity to power thousands of chips and cooling systems.
A 1-gigawatt data center can consume as much electricity as 800,000 homes.
As a result, wholesale electricity prices in areas near significant data center activity surged by as much as +267% between 2020 and 2025.
The AI revolution is driving a surge in electricity demand, and households are increasingly feeling the cost.
sigeru retweeted
Copper needs $ 332bn of new mines.
Each one takes about 17 years to build.
Bloomberg new outlook sees a deficit starting in 2028 and widening to 20 million tonnes by 2050.
That’s about 44% of demand.
9 of the 10 transition metals it tracks now look covered. Copper is the odd one out.
Demand rises 61% to almost 46 million tonnes by 2050. Electric transport alone needs 11.4 million.
You can print money.... You can’t print a mine.
sigeru retweeted
The AI boom is increasingly being built on borrowed money.
AI-related bond issuance has surged to $419 billion so far in 2026, 90% higher than all of 2025.
Every equity bubble inevitably becomes a credit bubble before it bursts.
sigeru retweeted
BREAKING: The US and Iran have reportedly given preliminary approval to a new compromise proposal from Qatar and Pakistan, per Reuters
It follows a report Monday that a breakthrough deal between Washington and Tehran could be announced as soon as this week, and Iran’s president saying Friday that Tehran is ready for dialogue.
Readers added context they thought people might want to know
The report is from Al-Araby Al-Jadeed citing anonymous sources, not Reuters.
alaraby.co.uk/politics/%D8%A…
moneycontrol.com/world/us-iran-…
sigeru retweeted
SoftBank's credit risk is SURGING amid AI funding concerns:
SoftBank's 5-year credit default swap (CDS) spread is up to 432 basis points, its highest level since November 2022.
This figure has risen over +80 basis points since early September, and more than +30 bps over the last week.
Furthermore, the company's shares have cratered -50% from their June peak.
This comes as investors grow increasingly concerned about debt-fueled AI investments, rising financing costs, and SoftBank's potential acquisition of Allied Universal.
SoftBank faces at least -$6 billion in projected cash outflows in the Fiscal Year ending March 2027, with additional AI investments and a potential acquisition increasing its funding needs.
The company's loan-to-value (LTV) ratio could also approach its self-imposed 25% ceiling, potentially forcing it to raise fresh capital to finance the deal.
SoftBank's CDS spread is now the highest among major Japanese companies, exceeding Nissan's by more than 130 bps.
SoftBank's growing financial strain also highlights the cost of its aggressive AI strategy, including its $64.6 billion cumulative investment in OpenAI, representing a ~13% ownership stake, and the sale of its entire Nvidia stake to free up capital.
One of the largest investors in the global AI ecosystem is facing severe pressure.
sigeru retweeted
"AI is a fad. Nobody uses it to make money."
In the meantime, revenue per worker continues to climb.
sigeru retweeted
So far, AI has not produced an economy-wide acceleration in productivity growth. Adjusted for shifts in utilization rates, total factor productivity growth was zero in Q2.
sigeru retweeted
WTF happened to German electricity?
Most expensive in the neighborhood. Someone tell Germany that AI runs on electricity. A lot of it.
sigeru retweeted
The premise for why static support/resistance levels "work" is that traders who bought or sold at those levels will exhibit that behavior again when they see the price get back to that important level. If it was a good place to buy before, it will be a good place to buy again in the future, or so goes the thinking.
Now ask yourself, does anyone make trading decisions regarding IWM based on a price level they remember from months ago?
And furthermore, since IWM is based on the Russell 2000 Index, does it make sense that the outcome of those stocks aggregated together into an index would be driven by this supposed effect of seeing a prior ETF level get reached again, and trading based on it?
The supposed effects of static support and resistance behavior break down when stocks are pooled into an index or ETF, especially one that is not greatly watched by people. IWM arguably is watched more than many ETFs, but the breakdown effect is the same. A level seen months ago is just not that important to what happens now.
IF YOU DOUBT ME, and I am confident many readers will, try this: Pull up a chart of IWM, and start drawing horizontal lines from any prior tops or bottoms months ago which you think were important. Then see how many of those levels really mattered to future price action. But you should only do this if you are ready to have your preconceived notions get challenged by evidence.
Then try it with other small ETFs. Try it with commodity prices. Try it with currencies. See how many such lines matter across months.
sigeru retweeted
China's exports to Russia began booming right after the Ukraine invasion, but that's not the end of it. China's exports to countries in Central Asia are also up lots. You add all that up and it's 10% of Russia's GDP. China wants Russia to win in Ukraine...
robinjbrooks.substack.com/p/…
