Original author: Zhao Ying

Original source: Wall Street Watch

Chip trading driven by artificial intelligence is experiencing a rapid cooling. The Philadelphia Semiconductor Index has fallen about 19% from its peak in June, just one step away from confirming a bear market, as funds begin to withdraw from overvalued chip and memory stocks and shift towards sectors like finance, retail, and transportation that more directly benefit from economic resilience.

In Thursday's U.S. stock trading, the Philadelphia Semiconductor Index fell 4.3%, with all 30 component stocks lower than the record high of June 22. According to Dow Jones market data, if the index falls further into the 20% range from its peak, it will confirm a technical bear market. On Friday, the sell-off further spread to Asian and European markets, with Japan's Nikkei 225 index closing down 4%, and stocks such as TSMC, Kioxia, and European chip equipment firm ASML under pressure.

The reversal of AI trading is dragging down global risk appetite. The Philadelphia Semiconductor Index has fallen a cumulative 8.5% this week, on track to record its worst single-week performance since last year's "Tariff Day" shock. Nasdaq 100 futures fell 1.6%, and S&P 500 futures fell 0.9%, as concerns about the returns on investments in AI infrastructure, inflation risks, and the outlook for monetary policy intensify.

Goldman's trading chief described the current AI market as an "elastic band" that is constantly being stretched. As hyper-scale cloud computing companies continue to ramp up capital expenditures, the key question in the market has shifted from "How large are the investments?" to "When and how will these investments convert into returns?" The upcoming earnings reports from tech giants may become a crucial first checkpoint to validate this logic.

Chip Index Nearing Bear Market, Profit Taking Evolving into Widespread Cooling

Chip stocks were previously among the most sought-after trades this spring. As investors were once concerned about the "Seven Giants" bearing most of the costs for AI data center construction, funds shifted to chase chip manufacturing, storage, and semiconductor equipment companies, betting that they would become direct beneficiaries of the capital expenditure cycle.

However, this trade is rapidly reversing. As of Thursday, the Philadelphia Semiconductor Index had fallen 19% from its historical high on June 22. The index dipped 4.3% on that day, just one step away from confirming a bear market with a 20% pullback.

Individual stocks are experiencing even more severe volatility. Marvell Technology has dropped nearly 40% since the Philadelphia Semiconductor Index peaked, yet is still up 121% year-to-date. This indicates that the current adjustment is more concentrated on previously high-flying AI beneficiaries, as investors reassess whether high growth expectations have been fully or even excessively priced in.

In Thursday's U.S. stock market, Sandisk, Western Digital, and Seagate all fell over 9%, while Intel and Micron dropped about 6%. On Friday, Asian markets continued to decline, with Japanese memory chip maker Kioxia at one point dropping more than 16%, having retreated over half from its June peak; TSMC's stock price also significantly declined.

Earnings Still Strong, but Market Begins to Question Sustainability of Growth

The chip sector is not facing a short-term profit collapse. FactSet data shows that the market expects S&P 500 constituent companies to report a 23.6% year-on-year increase in earnings for the second quarter, while the semiconductor and related equipment industry is expected to see earnings growth as high as 131%.

The question is whether the strong current performance is enough to support the valuations that have reflected years of growth expectations. David Russell, Global Market Strategy Director at TradeStation, stated that tech companies may deliver impressive performance, but the market is questioning whether this growth can be sustained in the next one to three quarters.

TSMC's performance highlights this contradiction. Despite announcing record quarterly profits, the stock price has weakened significantly this week. The Financial Times reported that TSMC fell more than 7% on Friday. The better-than-expected performance failed to stop the decline in stock prices, indicating that the market's focus has shifted to the sustainability of orders, returns on capital expenditures, and the growth slope of AI demand, rather than the profits of a single quarter.

Kevin Gordon, Macro Research and Strategy Director at the Schwab Center for Financial Research, believes that the sharp adjustment in chip stocks does not necessarily constitute a severe warning signal. Over the past decade, the Philadelphia Semiconductor Index has seen six pullbacks of more than 20% and 31 adjustments of at least 10%, exhibiting volatility significantly higher than that of the S&P 500 Index. However, frequent adjustments also mean that this sector is highly sensitive to changes in valuations, inventory cycles, and expectations of capital expenditure.

Funds Shifting to Economically Sensitive Sectors, Market Breadth Expanding

While chip stocks are under pressure, a clear rotation of funds is occurring within U.S. stocks. The financial sector set new closing highs for the second consecutive day on Thursday, driven by strong bank performance. The Dow Jones Transportation Average has risen over 30% year-to-date, approaching historical highs, and retail ETFs have also risen to near the highs seen since early 2022.

David Royal, Chief Financial and Investment Officer at Thrivent, stated that the expanding market breadth is a healthy signal, and recent employment and retail sales data also show the economy remains resilient. Funds shifting from overvalued tech sectors to finance, consumer, and transportation fields indicate that investors are not fully retreating from risk assets but are reallocating to assets that are more sensitive to economic growth.

This rotation has also weakened the previous relative advantage of chip stocks. When the economic outlook remains stable, investors have more options and do not need to continue placing concentrated bets on the highest valued and most crowded companies in the AI infrastructure chain.

Goldman Warns: Tension Exists Between AI Capital Expenditure and Realization of Returns

Mark Wilson, Head of EMEA Equity Hedge Fund Business, and Rich Privorotsky, Head of EMEA Equity Flow Intermediary Business at Goldman Sachs believe that the mismatch between AI infrastructure investments and commercialization returns is becoming the most core risk variable in the market.

The two pointed out that hyper-scale cloud computing companies like Microsoft, Amazon, Alphabet, and Meta are investing in AI infrastructure at a scale surpassing the growth rate of their own operating cash flow. However, in the short term, how much revenue, profit, and cash returns these investments can generate remains highly uncertain.

Privorotsky described the AI market as an "elastic band," noting that the key is not whether the market still views AI's long-term direction positively, but how long this stretching of valuations and capital expenditures can continue. Goldman also noted that the rapid dissemination of frontier models and decreasing inference costs may change the AI value chain: the premium for hardware and computing power scarcity may decrease, while platform companies that master distribution channels and workflows may capture more value.

If any one of the hyper-scale cloud computing companies leads in cutting capital expenditures, the market may quickly reassess demand expectations for the entire AI hardware chain, triggering a broader chain reaction.

Earnings Season Will Test Whether AI Trading Can Regain Support

The next market focus will shift to the earnings reports and capital expenditure guidance of large-tech companies. Alphabet and Tesla will announce their earnings on July 22, and their statements regarding AI investment, data center construction, and commercialization progress may affect whether the chip sector can stop its decline.

In the short term, the high earnings growth rate of chip stocks still provides some valuation support, but the market is no longer satisfied with "high growth" alone. Investors need to see improvements in income, profit margins, and cash flow resulting from AI investments, as well as confirm that the capital expenditures of hyper-scale cloud computing companies will not slow down due to financing pressures, rising inflation, or returns falling short of expectations.

Whether the Philadelphia Semiconductor Index officially falls into a bear market may only be a technical demarcation. For the market, a more important boundary lies in whether AI trading can shift from prepaying for growth over the coming years to validating the path of real returns.

免责声明：本文章仅代表作者个人观点，不代表本平台的立场和观点。本文章仅供信息分享，不构成对任何人的任何投资建议。用户与作者之间的任何争议，与本平台无关。如网页中刊载的文章或图片涉及侵权，请提供相关的权利证明和身份证明发送邮件到support@aicoin.com，本平台相关工作人员将会进行核查。