Microsoft leads tech stock rebound in July

U.S. technology stocks staged a powerful late-July rebound led by cloud computing, artificial intelligence infrastructure and memory-chip companies, but the advance remained narrow and heavily dependent on the same high-volatility names that had fallen hardest in the preceding selloff. Microsoft’s earnings-driven surge, Amazon’s higher capital-spending target and an 8% jump in the Philadelphia Semiconductor Index shifted attention back toward whether the industry can build enough AI capacity rather than whether it is spending too much.

Microsoft shares rose more than 15% after the company said its cloud division had surpassed a $100 billion annual revenue run rate and was growing by more than 40%. The move marked Microsoft’s strongest daily performance since 2008 and added roughly $450 billion to its market value in one session. The company’s results gave markets a concrete revenue figure to place alongside the enormous infrastructure budgets now being committed by large technology groups.

Amazon added another 9% following its earnings report and said it had raised its full-year capital expenditure plan to $120 billion. That figure reinforced the view that demand for data centers, advanced chips, networking equipment and electricity remains strong. It also puts pressure on suppliers to expand capacity without creating the kind of excess inventory that has historically hurt semiconductor cycles.

Chip shares recover after steep three-day decline

The Philadelphia Semiconductor Index rose 8%, nearly recovering a decline of almost 10% over the prior three sessions. Lam Research gained 18%, its largest one-day rise since 1999, as traders returned to equipment makers expected to benefit from new fabrication plants and memory expansion.

Samsung Electronics added to the optimism after reporting record quarterly operating profit and warning that tight memory supply could persist until 2028. Memory chips are central to AI servers, which require far more high-bandwidth memory than conventional computers. A prolonged shortage would support prices and supplier margins, though it would also raise costs for cloud providers racing to deploy larger AI systems.

The recovery was especially dramatic in South Korea, where memory-linked equities had been under severe pressure. South Korea’s main stock index was up as much as 18% intraday despite remaining lower for the week. A two-times leveraged long ETF focused on Korean memory stocks rose 70% in one day, demonstrating how rapidly leveraged products can amplify reversals when underlying shares move sharply.

That scale of movement complicates efforts to judge whether selling pressure has fully cleared. One assessment by a major U.S. bank estimated that about 90% of hedge-fund deleveraging in Korea had already taken place and that foreign selling had slowed. The bank estimated net outflows of $110 billion over a year, with about 90% concentrated in two memory-heavy stocks.

It also described the KOSPI as having fallen more than 40% from its June 22 peak, leaving the index trading at a forward price-to-earnings ratio of 5 and a price-to-free-cash-flow multiple of 5. Those valuations can attract bargain-focused traders, but low multiples alone do not resolve concerns about earnings concentration in a handful of chipmakers.

Rebound favored the most heavily sold shares

The day’s cross-sectional trading suggested that the rally was driven partly by position-covering rather than a broad reassessment of corporate fundamentals. Lower-margin companies outperformed stronger-margin peers, while the shares that had fallen most during the earlier decline generally produced the largest rebounds.

That pattern matters because it differs from a rally led by companies delivering accelerating earnings or expanding profit margins. It can emerge when traders buy back short positions, when leveraged funds rebalance after sharp gains, or when risk controls force portfolios to add exposure as volatility falls.

The data on leveraged Korea-focused products offered a similar caution. Share counts were broadly unchanged because redemptions were close to zero, indicating that existing exposure was repriced upward rather than removed from the market. Cumulative inflows into leveraged products remained near recent highs and had stalled rather than reversed into meaningful outflows. If prices continue to rise, the leverage embedded in those products could rebuild mechanically.

Meta provided a contrasting example of the market’s increasingly selective response to AI spending. Its shares fell nearly 8% after results and guidance disappointed expectations, even as Microsoft was recording one of the largest single-day gains in its history. The divergence suggests markets are rewarding capital expenditure when companies can link it to clear cloud revenue, capacity utilization or demand, while punishing spending plans that do not yet show a comparable payoff.

Higher yields challenge the risk-on narrative

The technology rally also occurred alongside higher Treasury yields and a stronger gold price, an unusual combination that pointed to unresolved tension over growth, inflation and fiscal conditions. The 10-year U.S. Treasury yield rose to 4.66%, its highest level since 2007, while U.S. equities advanced.

Only technology and consumer discretionary stocks finished higher among major U.S. equity sectors, leaving the market advance concentrated rather than broad-based. Rising long-term yields increase the discount rate applied to future profits and can be especially difficult for growth stocks whose valuations depend on earnings expected years ahead. The fact that major AI names rallied anyway reflected the strength of the earnings and spending narratives, but it also left the move vulnerable if bond yields continue rising.

Currency markets added another source of volatility after Japan’s Ministry of Finance intervened during New York trading. The yen moved from around 164 per dollar to about 157, a one-day gain of nearly 3% and its largest daily move since late 2022. The intervention came between scheduled policy meetings of the Federal Reserve and the Bank of Japan, catching traders positioned for a later response.

The move did not produce a broad unwind of yen-funded carry trades, which involve borrowing cheaply in yen to buy higher-yielding assets elsewhere. A sustained carry-trade unwind would typically require a sharper compression in the yield gap between U.S. and Japanese government bonds. Selling U.S. Treasuries to repatriate funds into yen could also push long-term U.S. yields higher and tighten financial conditions.

Durable recovery would require broader participation

The rebound has revived the AI trade, but its durability depends on evidence beyond sharp recoveries in semiconductors and leveraged Korea products. Broader participation across equity sectors, declining long-term yields and fewer forced liquidations would provide a more stable foundation than a rally concentrated in the market’s most heavily sold technology shares.

Historical comparisons offer a reminder that infrastructure leadership can remain volatile even when the underlying technology is transformative. During the dot-com era, hardware and network suppliers were often treated as safer “picks-and-shovels” beneficiaries, yet Cisco and Intel later suffered declines of roughly 80% to 90% from their peaks. The present memory cycle may ultimately prove more resilient, especially if supply remains constrained, but rising capital budgets have also increased the market’s sensitivity to any sign that AI revenue growth is slowing.

Disclaimer: The content on this page is provided for general informational purposes only and does not represent the views or financial advice of Toobit. We make no guarantees regarding the accuracy or completeness of this information and shall not be held liable for any errors, omissions, or outcomes resulting from its use. Investing in digital assets involves risk; users should independently evaluate their financial situation and the risks involved. For further details, please consult our Terms of Service and Risk Disclosure.