CPI ignites sell-off, AI market begins to face the test of financing costs

After the Inflation Shock, the Market Begins to Worry About Who Can Withstand Higher Interest Rates

Wall Street has entered a trading phase where multiple risks are being released simultaneously. Hot inflation data has raised interest rate hike expectations, the energy market remains under pressure, and the decline of AI-related bonds is starting to affect stock investors' judgments.

Data released by the U.S. Bureau of Labor Statistics shows that the CPI rose 0.6% month-on-month in September and 3.7% year-on-year; the core CPI, excluding food and energy, rose 0.3% month-on-month and 2.6% year-on-year. Gasoline contributed to more than half of the overall monthly increase, but housing, healthcare, and airfare also rose, indicating that price pressures are not limited to energy.

The market's initial reaction aligned with a typical inflation shock: U.S. Treasury yields rose, interest rate hike expectations heated up, the dollar strengthened, and gold, cryptocurrencies, and interest rate-sensitive stocks came under pressure.

However, U.S. Treasuries regained buying interest as stocks weakened further. Innes believes this change is more noteworthy than the initial sell-off: investors are beginning to consider how much pressure current financing costs will place on the economy and businesses.

High inflation suggests that the Federal Reserve may need to maintain tighter policies; falling stock prices and widening credit spreads could further tighten financial conditions beyond policy measures. Corporate financing becomes more expensive, investor risk tolerance declines, and economic activity may also be suppressed.

This complicates policy judgments. Even if the Federal Reserve has not yet taken new actions, the market itself may have already brought about additional tightening effects.

Therefore, the rebound in U.S. Treasuries does not mean that the inflation problem has been resolved. In the author's view, it resembles the market beginning to trade two risks simultaneously: prices not coming down and high financing costs gradually eroding growth.

AI Valuations and Profit Expectations Are Under Pressure Simultaneously

For AI-related stocks, rising interest rates are only part of the pressure.

Higher interest rates reduce the present value of future cash flows, making it harder to maintain valuations that rely on future growth. At the same time, changes in the competitive landscape may also affect how much profit companies can ultimately achieve.

In the past, investors were willing to pay high valuations for certain AI companies, with an important premise: demand is sufficiently large, and leading companies have the ability to maintain high profit margins over the long term.

Innes questions this premise. Taking memory chips as an example, if Chinese manufacturers narrow the technology gap, industry capacity continues to increase, and the supply tightness of server DRAM eases, the current high profit margins may contain strong cyclical factors that are difficult to extrapolate directly to the long term.

This competitive scenario still requires subsequent data verification, but it explains why the market is adjusting both valuations and profit expectations simultaneously.

When the cost of capital rises and expected profits are threatened by competition, the return threshold that companies need to cross will further increase.

This is especially true for infrastructure companies. Data centers, servers, and supporting energy facilities require significant upfront investment, with revenues gradually realized in the future. The more expensive financing becomes, the more projects need to rely on high utilization rates, stable customers, and sustainable pricing to support returns.

Bond Market Questions: Can Companies Hold Out Until Revenues Materialize?

Compared to stocks, the concerns conveyed by the credit market are more direct.

CoreWeave's previously issued bonds, with a coupon rate of 9.75% maturing in 2031, fell to a market value of about 81% on October 14, corresponding to a yield of over 15%. The cost of debt default protection has also risen significantly, indicating that investors are demanding higher risk compensation.

It is important to distinguish between coupon rates and market yields. A decline in bond prices and an increase in yields do not mean that a company's interest expenses on existing fixed-rate debt immediately increase; it reflects a change in the returns that investors require for holding that bond, which may also affect the conditions for the company's future borrowing and refinancing.

Innes believes that AI trading is entering a new phase as a result.

Whether demand for computing power is growing remains important; however, for creditors, a more pressing question is whether the companies providing funding for these demands can maintain their ability to repay debts and access financing before revenues fully materialize.

A company may have a large number of orders but could face financial pressure due to excessive upfront investments and slow cash recovery. There is no automatic equation between business growth and financial stability.

Once such concerns persist, risks may reinforce each other: widening credit spreads increase the cost of new financing; expensive financing depresses project returns; declining return expectations impact stock valuations; weakening stock prices may further undermine a company's ability to raise funds through equity financing.

The capital markets that previously supported expansion may also become more cautious as a result.

What can break this cycle is stronger profits, clearer cash flow guidance, and operational data sufficient to prove project returns. The market needs to see that companies can not only build facilities but also convert investments into cash available for debt repayment and ongoing operations.

Leading Stocks Weaken, Market Buffering is Also Diminishing

The reason AI financing pressures may affect the broader market is that related companies have become important pillars of the market.

When gains are concentrated in a few companies, their profit expectations, capital expenditures, and financing capabilities have a greater impact on overall risk appetite. Once leading stocks come under pressure simultaneously, other sectors may not be able to step in promptly.

Innes also incorporates the hedging behavior of options traders into the analysis. Under certain holding structures, traders' hedging buys and sells can help buffer market volatility; however, as prices and positions change, this effect may weaken and, in some cases, amplify volatility.

This does not mean that breaking below a certain level will necessarily trigger a chain sell-off. A more accurate interpretation is that when leading stocks weaken and buying interest declines, new selling pressure may lead to larger price changes.

The energy market adds uncertainty to this situation. In addition to crude oil prices, the author is also focused on diesel processing profits and tanker freight rates. If pressures in the transportation and refined oil segments persist and further transmit to end prices, it may become more difficult for inflation to recede.

For AI companies that rely on financing for expansion, this means that interest rate pressures may not quickly ease.

Innes also discusses the possibility of government support for strategic technology companies in the future, but this is speculative and should not be assumed to mean that AI companies have already secured financing support.

At present, investors still need to make judgments based on the financial capabilities of the companies themselves. The core question raised by this round of adjustments is: under higher capital costs, which companies can convert AI demand into sufficient cash flow to support expansion and fulfill debt commitments?