BlackRock: The Link Between Crude Oil and US Treasury Yields Has Broken Down; AI Risk Has Replaced Geopolitical Risk

Cailianshe Financial News Agency

Open APP

Cailianshe Xiaoxiang, Editor-in-Charge Lin Qi 2026-11-02 08:47

① When the market broke the trading logic of "oil prices moving in tandem with US Treasury yields" in October, it sent us an important signal; ② BlackRock senior portfolio manager Rosenberg admitted frankly, "In the market's view, geopolitical risk has given way to AI risk."

Cailianshe, November 2 (Editor: Xiaoxiang) — If there was one "can't-lose" iron rule in the 2026 market, it was that US Treasury yields would rise in sync with climbing crude oil futures prices. Yet over the past month, that rule has come to an end. Jeffrey Rosenberg, senior portfolio manager of the Systematic Alternative Investment ETF (IALT) under BlackRock, the world's largest asset management company, noted that the collapse of this trading logic has far-reaching implications for the market.

"Crude oil and US Treasury yields had maintained an astonishing correlation, but in October, that linkage came to an abrupt halt," Rosenberg said.

The relationship between crude oil and US Treasury yields was also highlighted at last Wednesday's press conference following the Federal Reserve Chair Kevin Warsh's policy meeting. Warsh pointed out that the continued rise in crude oil futures prices is a factor the central bank must consider in fighting inflation.

But Rosenberg believes that what Warsh avoided discussing is precisely the "most core takeaway."

"The correlation between US Treasuries and crude oil reached as high as 80% this year," Rosenberg noted, a level far exceeding historical norms. "People have basically been trading around oil prices, and for very good reason."

So what exactly happened to cause the correlation to break down last month? Rosenberg believes it all comes down to the cooling AI trade. When the market broke the trading logic of "oil prices moving in tandem with US Treasury yields" in October, it sent us an important signal — he admitted frankly, "In the market's view, geopolitical risk has given way to AI risk."

Over the past month, the AI trade has suffered a notable pullback — last Friday's broad selloff on the Nasdaq market, and Google's largest single-day decline in company history earlier that week (with $500 billion in market value wiped out in a single day), laid that pain bare.

Rosenberg analyzed that US Treasury yields turned downward because the market viewed the weakness in the AI trade as a signal of overall US economic weakening. And last week's US GDP report further confirmed this logic. "AI is everywhere, dominating everything at all times," he said, whether it is economic growth driven by the investment side or the wealth effect generated on the consumption side, all are inseparable from artificial intelligence.

"The third-quarter GDP data was extremely impressive, but it was all thanks to AI," Rosenberg added. He pointed out that the growth in equipment investment and consumption far exceeding expectations were both driven by the wealth effect brought by AI; especially against the backdrop of continued declines in real incomes for the American public, the better-than-expected performance on the consumption side appears even more unusual.

Data shows that US corporate investment grew 12.7% in the third quarter, the largest increase since 2011 excluding the pandemic period, led by a 37% increase in information processing equipment investment. Industrial equipment investment also grew by nearly 25% — although this is not purely AI investment, it partly reflects the massive spending on data center construction.

Admittedly, the "market breadth" indicator measuring the upward momentum of S&P 500 constituent stocks improved significantly in October after hitting a 2026 low in late September, and defensive sectors such as healthcare and consumer staples also rebounded. But Rosenberg pointed out that the core conclusion drawn by the overall market in October is: any hint of "trouble" in the AI trade will be seen as a wavering of the US economic growth outlook and will directly trigger expectations of slowing inflation (disinflation).

And in a disinflation environment, "only one asset can serve as a hedge," Rosenberg said, and that is "the variety everyone previously avoided like the plague — long-end US Treasuries."

"This is the truly important takeaway from last month's market performance," he concluded.

How to view the outlook?

However, despite holding the above view, Rosenberg still cautiously stated that one cannot therefore assert that this dynamic has returned bonds to the ranks of "safe-haven assets." "Simply put, the market has not returned to the past norm of the 'stock-bond seesaw,'" he said bluntly.

Rosenberg believes the reason is that the bond market is being pulled by two opposing forces: rising oil prices and inflation remaining above target for an extended period have not disappeared, which still poses challenges to fixed-income assets and weakens what he calls the "effectiveness" of fixed-income assets as a hedging tool.

"Inflation is still too high, which ties the Fed's hands in implementing easing policy, and we have not yet emerged from this predicament," he explained. "But last month's market also made us realize that if the shock is severe enough — once the AI trend reverses and backlashes on the macroeconomy and inflation trajectory, it could trigger a full reversal of market sentiment, bringing a significant disinflation effect, and at that point bonds would repeat October's rally."

Even so, he reminded investors not to bet that this trade will play out one-sidedly or without cost. Investors still need to weigh: whether to place more emphasis on the risk-diversification value of bonds demonstrated by October's market action, or to guard against the alternative possibility that AI does not drag down the economy.

In any case, AI trading itself has become a pivotal part of the bond market. Over the past week, both the stock price plunge triggered by Google's earnings report and Microsoft's return to the public bond market have profoundly confirmed this. "This surprised the market quite a bit, after all, Microsoft had not appeared in the public bond market for nearly a decade," Rosenberg said.

As the capital expenditure needs of tech giants increasingly frequently exceed their free cash flow, the public bond market has to prepare for an impending flood of debt issuance. "When the bond market — especially the public bond market — looks at where these bonds are coming from... this will be a huge new issuer, and its issuance scale is increasingly frequently and continuously growing," he said.

At present, the financing scale of hyperscalers is gradually approaching the capacity limits of bank loans and the private credit market. Although Rosenberg emphasized that the global bond market, worth as much as $40 trillion to $60 trillion, has the capacity to absorb the projected $4 trillion in AI-related debt over the next four years, "the market's absorption capacity is ultimately not unlimited."

Such enormous financing demand will inevitably push down bond prices and push up credit spreads.

However, Rosenberg also said that pushing more hyperscaler debt into the public bond market may not be a bad thing, because it means bond prices and credit default swap (CDS) spreads will bring greater transparency. Although Microsoft's CDS spread is currently hovering near a five-year high (about 55 basis points), it is still far from Oracle's CDS spread (close to 300 basis points). This provides another trading opportunity for investors who are unwilling to buy bonds directly due to the U.S. inflation environment and fiscal challenges, but still want to hedge risks related to the AI boom.

Rosenberg believes that the message the CDS market is currently conveying to investors is not that large hyperscale tech companies have a high risk of default, but that as these companies seek more debt financing, the ability of banks and financial intermediaries to find counterparties for these risks is becoming increasingly limited.

"Think back to the massive investment booms in financial markets throughout history; they never charge forward in a straight line, but rather leap in steps, then go through a shakeout and reset," Rosenberg concluded. The previous AI boom had only shown "the soaring side"; "this is still a grand narrative with enormous potential, only with the emergence of AI pessimism, the market has finally regained a bit of balance."

Global Market Intelligence

U.S. Treasury Dynamics

Cailian Press Statement: The content of this article is for reference only and does not constitute investment advice. Investors who act on it do so at their own risk.

Business Cooperation

Popular Unlocks

[Research Select • Data] Chinese equity funds record net inflows for 3 consecutive weeks, with scale hitting a new high since 2025

Chinese equity funds have recorded net inflows for 3 consecutive weeks, with the scale hitting a new high since 2025.

[ETF Playbook] Leading companies post impressive results, top firms are entering a critical period of "volume expansion + overseas expansion," institutions say they are bullish on this sector's performance amid a volatile market; equity ETFs continued to see net inflows of over 100 billion yuan in subscriptions and redemptions last week

① Leading companies post impressive results, top firms are entering a critical period of "volume expansion + overseas expansion," institutions say they are bullish on this sector's performance amid a volatile market; ② equity ETFs continued to see net inflows of over 100 billion yuan in subscriptions and redemptions last week; ③ Institutions: Looking ahead to November, against the backdrop of strengthened external disturbances and persistently high oil prices, this style is expected to continue to outperform.

[Nine O'Clock Special] China plans to achieve a moon landing before 2030! Institutions say China's space launches will accelerate significantly in 2026, and the two domestic internet satellite constellations, Guowang and Qianfan, are expected to achieve early application deployment in the coming years; price raised by 3,000-5,000 yuan/ton, dye giant issues latest price adjustment letter

① China plans to achieve a moon landing before 2030! Institutions say China's space launches will accelerate significantly in 2026, and the two domestic internet satellite constellations, Guowang and Qianfan, are expected to achieve early application deployment in the coming years; this company has developed key materials for aerospace engines; ② Price raised by 3,000-5,000 yuan/ton! A dye giant issues its latest price adjustment letter, analysts believe the industry's long-term landscape is set for certain optimization and are bullish on the continued upward shift of the industry chain's price center; this company has 32 series and more than 400 varieties of reactive dyes; ③ Nvidia recently publicly demonstrated its next-generation co-packaged optics technology, called OOI, at SEMICON West 2026.

BlackRock: The Link Between Crude Oil and US Treasury Yields Has Broken Down; AI Risk Has Replaced Geopolitical Risk

Topics

Tags

Body · AI translation