Last week, major U.S. stock indices closed higher, but AI chip stocks did not follow suit. This week, pressure on tech stocks continues, as bank earnings, TSMC’s financial results, and U.S. inflation data are set to be tested by the market.
This is where the current market cycle becomes particularly worth discussing. While indices remain near their highs, the market has begun reassessing the quality of growth across different companies. Capital has not withdrawn en masse, but expectations around AI investment, profit realization, and financing costs have become significantly more specific.
Last Week in Review: Index rose, with increased divergence among sectors
According to BiyaPay data, as of the close on October 9, the S&P 500 stood at 7,811.54, rising approximately 1.2% for the week; the Dow Jones Industrial Average closed at 51,654.95, up about 0.9%; and the Nasdaq Composite reached 27,366.17, advancing roughly 0.6%. The Russell 2000, however, declined by approximately 0.9%. The divergent performance between large-cap and small-cap stocks indicates that the index gains do not necessarily reflect a synchronized market-wide recovery.
Hong Kong stocks also showed signs of recovery. On October 9, the Hang Seng Index rose approximately 1.8% to close near 24,211 points, gaining about 1% for the week; the Hang Seng Tech Index rose about 3.1% on the day. The rebound in internet stocks supported Friday’s market performance, but whether this single-day recovery can translate into a sustained rally will depend on future industry performance and earnings expectations.
While indices rise, sectors show clear divergence—monitoring the market requires comparing different assets side by side. For example, BiyaPay, a global one-stop asset allocation platform, uses USDT as the core entry point for funds, connecting multiple scenarios—such as U.S. and Hong Kong stocks, cryptocurrencies, commodity futures, and cross-border remittances—through a unified account system. It supports exchanges between over 30 fiat currencies and more than 200 digital assets, enabling users to trade real stocks like those on U.S. and Hong Kong exchanges with zero commissions via USDT. For those frequently monitoring multiple markets, this centralized entry point offers seamless continuity in tracking market trends, converting assets, executing trades, and managing subsequent fund usage.
AI chips are under pressure as the market begins to question how revenue will be realized.
NVIDIA closed last week at $229.28, declining approximately 2.00% for the week based on the closing prices from the previous and current Friday. AMD closed at $608.10, falling about 4.07%. On October 8, both companies declined by approximately 2.94% and 3.90%, respectively, indicating concentrated pressure on AI-related stocks during the week.
An important backdrop to this market correction is the reassessment of AI companies’ revenue scales and their reporting methodologies. According to reports, OpenAI disclosed its annualized revenue for September to be close to $50 billion, differing from the previously widely cited figure of approximately $70 billion. However, the report also noted that the discrepancy between these two figures relates to differences in reporting standards when comparing with other AI companies. Therefore, it would be inaccurate to state that “OpenAI’s revenue suddenly dropped by $20 billion”; annualized revenue does not represent actual full-year revenue already realized.
The impact of this on the market lies primarily in the need to reassess growth expectations. Chip orders, cloud service contracts, and data center construction ultimately depend on end-user demand. If AI application revenues continue to grow, infrastructure investments will have a clearer return pathway; if revenue realization lags behind expectations, the market will further question customers’ funding sources, compute utilization rates, and future purchasing capacity.
Therefore, the correction in chip stocks does not indicate that AI demand has reversed, but rather shows that capital markets are raising their expectations for growth metrics. Order volumes, contract values, and annualized revenue must align with actual deliveries, profits, and cash flow—especially for high-valuation companies, whose stock prices often already reflect highly optimistic growth expectations.
Energy and utilities strengthened, as the impact of high interest rates has not disappeared.
Last week’s relative strength was not limited to the technology sector. Using sector ETFs as a sample, the energy ETF XLE rose approximately 3.60% based on closing prices, and the utilities ETF XLU increased by about 3.97%. These figures reflect the performance of these two ETFs and cannot be directly used to represent the movement of all related companies.
The business models of energy and utility companies also differ. Rising oil prices may boost revenues for some upstream oil and gas companies, but increase costs for transportation, manufacturing, and consumer-facing businesses. The performance of utility companies must be analyzed in the context of demand, financing costs, and capital expenditures. The simultaneous strength of both asset classes indicates that the market is seeking diverse sources of profitability, and this does not imply that capital has completed a unified "defensive shift."
As of October 9, the U.S. 10-year Treasury yield was approximately 5.24%, and the WTI crude oil futures settlement price was around $91.85 per barrel. Both variables remain significant: the former affects valuation and financing costs, while the latter influences corporate expenses and inflation expectations. Even if stock indices rise, the constraints imposed by high funding costs on operations and valuation will not disappear.
Whether the recovery in Hong Kong stocks can continue depends on the industry, not just the Hang Seng Index.
Last Friday’s rebound in Hong Kong stocks continued to show divergence this week. On October 13, the Hang Seng Index rose approximately 0.3%, while the Hang Seng Tech Index fell about 0.79%. Stocks in sectors such as banking, real estate, and optical communications advanced, while semiconductor and storage-related stocks came under pressure, indicating that the index performance does not fully align with that of the technology sector.
For Hong Kong-listed tech companies, changes in the U.S. AI supply chain may influence sentiment, but revenue structure remains the key determinant of mid-term operational performance. Internet companies should monitor returns on advertising, e-commerce, cloud services, and AI investments; consumer electronics firms should focus on end-demand and order volumes; semiconductor companies must further differentiate by product type, capacity utilization, and price movements. The impact on different companies cannot be uniformly explained by the broad term "tech stocks."
This week’s focus: Inflation and earnings reports jointly test market expectations
Last Monday’s U.S. stock market performance reflected this divergence. On October 12, the Dow Jones Industrial Average rose approximately 0.4%, while the Nasdaq declined about 0.4%, and the S&P 500 closed slightly lower. Meta increased by around 3.4%, whereas NVIDIA, AMD, and Micron fell approximately 1.3%, 2.3%, and 3.9%, respectively. Even within the technology sector, there was no unified direction—chip stocks came under pressure while some platform companies strengthened.
CPI and retail sales should be viewed together.
According to the official schedule, the September CPI will be released on October 14, the September PPI on October 15, and September retail sales will also be published on October 15. All of these data releases are scheduled for 8:30 AM Eastern Time, which corresponds to 8:30 PM Beijing Time on the same day.
CPI influences market assessments of consumer price pressures, PPI provides insights into production-side costs, and retail sales help gauge demand resilience. These three data sets must be analyzed together. If prices rise while demand remains strong, the market may continue to expect higher interest rates to persist longer; if consumer spending weakens but price pressures do not ease simultaneously, companies may face the dual constraint of slowing revenue growth and persistently high costs.
Retail sales require particular attention to methodology. They measure nominal sales, and the increase in value may stem from higher sales volumes or rising prices. To assess whether consumption is improving, further analysis of product categories and price changes is necessary. Stronger-than-expected data does not automatically translate to positive outcomes, nor does weaker data necessarily cause immediate stock price gains—the key lies in how it alters interest rate and earnings expectations.
Bank financial reports look at profitability as well as credit quality.
JPMorgan Chase and Goldman Sachs are both scheduled to release their third-quarter earnings on October 13. As of the time this article was written, the results have not yet been announced and cannot be described as having exceeded or fallen short of expectations.
Key indicators to watch in bank earnings reports, beyond earnings per share, include net interest income, deposit costs, loan demand, and credit loss provisions. Higher interest rates may boost asset yields, but competition for deposits and rising funding costs can erode some of those gains. Trading, underwriting, and M&A activities provide another set of signals regarding the vibrancy of capital markets. Only by examining these metrics together can you determine whether profit growth stems from operational improvements or from a few selective businesses and one-time factors.
Banks also serve as a window into economic stress. If consumer credit, delinquency, and non-performing loan data remain stable, the market has stronger grounds to assess demand resilience; if credit quality begins to deteriorate, even strong quarterly profits should prompt consideration of the potential impact on future provisions and loan growth.
TSMC and ASML verify how AI investments are transmitted
ASML plans to announce its third-quarter results on October 14, and TSMC’s third-quarter earnings call is scheduled for October 15 at 2:00 PM Beijing Time. The two companies, operating in semiconductor equipment and wafer manufacturing respectively, can provide insights at different levels of the supply chain.
TSMC has announced its September revenue of approximately NT$511.857 billion, representing a 54.6% year-over-year increase and a 0.6% month-over-month decrease; cumulative revenue for the first nine months of the year increased by 41.1% year-over-year. This indicates that revenue growth remains robust, but monthly revenue announcements do not include complete profit and cash flow data, so revenue growth alone cannot be used to assess operational quality.
This week’s earnings report is more noteworthy for gross margin, demand for advanced processes, capital expenditures, and guidance for the next quarter. For ASML, assessment should be based on new orders, order composition, and customer expansion plans. There is a time lag between equipment orders and current deliveries, and a transmission delay between wafer manufacturing revenue and end-application revenue. Both companies provide evidence of industry trends, not a unified answer for the entire AI market.
Last week’s index rise indicated that the market is still willing to pay for growth in certain companies; the pullback in chip stocks suggests this willingness is increasingly dependent on concrete evidence. This week’s inflation data will shape expectations for capital costs, bank earnings will test the health of the economy and credit conditions, and the results from TSMC and ASML will assess demand across the supply chain. The key question moving forward is whether companies’ reported profits and cash flows can continue to justify the growth expectations already priced in.