Nvidia Hits a New High in One Move, AI Computing Power and Tech Themes Stage a Major Counteroffensive! Nonfarm Payrolls Rise Only 29K + Oil Prices Decline, Fed Rate Hike Bets Cool Sharply

After the latest nonfarm payrolls data was released, market bets on aggressive Fed rate hikes in October cooled comprehensively. Combined with international oil prices falling below $100 per barrel amid a series of positive catalysts, this jointly drove a major counteroffensive in technology stocks across U.S. and European markets. In particular, AI computing power-themed U.S. stocks, which had recently faced sustained pressure on the DCF valuation system from rising long-term U.S. Treasury yields of 10 years and above, together with sharply rising rate hike expectations, saw the most turbulent gains in early Friday trading. Nvidia, the "super overlord of AI computing power," even directly hit a record-high share price after the opening bell. U.S. job growth in September came in below expectations, with wage growth slowing, indicating that employers are showing some caution in hiring amid rising costs. Richard Moody, chief economist at Regions Financial Corp., said: "What surprises me is that this has persisted for more than a year. The low-hiring, low-firing situation remains." Overall, cooling nonfarm payrolls and falling oil prices are opening a "valuation decompression window" for AI computing power stocks and the broader global technology sector that relies on the cooling of the denominator side of the DCF valuation system. U.S. nonfarm payrolls increased by only 29,000 in September, while wage growth also unexpectedly slowed, pushing the market to further sharply reduce bets on a Fed rate hike in October, with the related probability once falling to less than 10%. At the same time, expectations for restored Middle East crude supply and coordinated release of strategic inventories in Europe pushed Brent crude below $100 per barrel. With both employment and energy prices cooling, U.S. Treasury yields fell accordingly, adding another macro catalyst to the rebound in U.S. and European technology stocks, and Nasdaq 100 futures extended gains to about 1.2%. A JPMorgan strategist team recently released a research report saying that what the institution is especially optimistic about is the reallocation opportunity formed by easing crowded positioning, falling valuations, and earnings resilience, with a focus on AI semiconductors. According to the institution's latest research report, since June, forward 12-month earnings per share forecasts for semiconductors have been raised by about 30%, while software earnings expectations have lacked corresponding improvement. The hyperscaler capital expenditure outlook it cited was approximately $950 billion for 2026, about $1.4 trillion for 2027, and about $3 trillion for 2030. Data released by the U.S. Bureau of Labor Statistics on Friday showed that nonfarm payrolls rose by 29,000 last month, while the prior two months' figures were also significantly revised down. This increase was below all baseline forecasts in a survey covering hundreds of economists before the official release of the nonfarm data. The unemployment rate rose to 4.2%, partly reflecting a significant expansion in the labor force. Strong consumer spending and robust business investment provided support for hiring, but many cost-conscious employers were reluctant to expand their workforce. Still, layoffs remained at low levels. "September's nonfarm payrolls data show that employment growth slowed after August's strong reading. Combined with the rise in the unemployment rate, this report does not support a Fed rate hike in October. We expect the Fed to keep rates unchanged for the remainder of this year," said Bloomberg Intelligence economists Anna Wong, Andrew Husby, and Eliza Winger. Because the unemployment rate remains at historically low levels, Fed officials can still focus on inflation when considering when to raise rates again. After the data was released, traders in the interest rate futures market sharply reduced bets on a Fed rate hike in October. After the nonfarm data was released, the probability of an October rate hike in the interest rate futures market was once less than 10%, while before the nonfarm release it had surged to close to the 80% threshold. "For the Fed, this data should be enough to completely eliminate the possibility of a rate hike in October," said Thomas Simons, chief U.S. economist at Wall Street financial giant Jefferies, in a note. "Now it appears that policymakers who emphasize there is still some time before another rate hike is necessary are more likely to remain patient." After the data was released, S&P 500 futures rose, while long-term U.S. Treasury yields and the dollar index both fell significantly. This was the last employment report before the November midterm elections, and economic conditions are expected to be an important factor influencing how American voters cast their ballots. The labor market is part of that, but despite consumers continuing to spend, lingering concerns about the cost of living remain a focus. It is understood that detailed nonfarm statistics show that U.S. local government employment declined, and some industries most vulnerable to artificial intelligence also saw employment decrease, including information, professional and business services, and financial activities. Investment in data centers and other AI-related infrastructure supported expanded hiring in construction and manufacturing. Employment in healthcare and in leisure and hospitality also increased. Meanwhile, average hourly earnings rose only 0.1% from August, below expectations; on a year-over-year basis they rose 3%, the slowest annual wage growth since 2021. The employment report consists of two surveys: one of businesses, used to compile nonfarm payrolls; the other of households, which provides the basis for statistics such as the unemployment rate and labor force participation rate. The household survey also has an independent employment measure, which rose sharply during the month. The September labor force participation ratethe share of the population working or actively looking for workrose to 61.8%, a four-month high. However, stability in the labor market also has its drawbacks. In some cases, insufficient labor mobility makes workers feel trapped in jobs they dislike. It also makes it harder for the unemployed to find another job. This may help explain why many Americans are pessimistic about the job market: fewer people think jobs are plentiful, while more think jobs are hard to find. 29K Nonfarm Payrolls Repel Rate Hike Bets, Tech Stocks Welcome a "Valuation Decompression Window" The semiconductor and AI computing power sectors, previously suppressed by both rising long-end rates and rising rate hike expectations, are becoming an important main line of this counteroffensive. European technology stocks had already risen about 2.1% before the nonfarm release, led by semiconductor equipment leaders. In early U.S. trading, AI computing power leader Nvidia posted the most dazzling gain, rising more than 2% directly at the 21:30 Beijing time U.S. market open to a record high of $236.420, with a total market value of about an astonishing $5.7 trillion. For AI infrastructure companies whose earnings expectations are still being revised upward, easing rate pressure means the valuation offset to profit growth is diminishing, and the market is thus welcoming a positive window shifting from "valuation killing" to "earnings taking over." The simultaneous cooling of nonfarm payrolls and wages, combined with restored crude supply and falling international oil prices, is easing the two forces suppressing technology stocksthe policy pressure of further rate hikes and the valuation squeeze on long-term earnings from high rates. U.S. nonfarm payrolls rose by only 29,000 in September, clearly below market expectations of about 90,000; July and August payrolls were revised down by a combined 60,000. At the same time, average hourly earnings rose only 0.1% month over month and 3.0% year over year, the unemployment rate rose to 4.2%, and the labor force participation rate increased to 61.8%. This set of data significantly weakened the basis for the view that "employment and wages remain overheated, forcing the Fed to raise rates again quickly"; the rise in participation also shows that part of the increase in unemployment came from more people entering the labor force and cannot be entirely explained as worsening layoffs. Global interest rate futures markets immediately lowered tightening bets. The latest market conditions after the U.S. stock market opened at 21:30 Beijing time on October 2 showed that the probability of an October rate hike once fell to less than 10%; U.S. two-year, ten-year, and thirty-year Treasury yields fell by about 7, 6, and 4 basis points, respectively, to 4.716%, 5.176%, and 5.569%, while the dollar index fell by about 0.2%. Short-end yields are most sensitive to policy expectations and fell more; the simultaneous decline in long-end yields provided a direct buffer for equity valuations and new corporate financing costs. Falling oil prices added another layer of support to this "great wave of valuation decompression," and this change occurred before the nonfarm release. As of 20:00 Beijing time, Brent crude was at $99.78 per barrel, down 2.47%; WTI was at $89.55, down 3.57%. However, using a simple comparison with the pre-war closing prices on February 27, the two were still up about 37.7% and 33.6%, respectively: the energy shock is easing, but its cumulative gains have not disappeared. The improvement on the supply side has substantive content: according to statistics cited by people familiar with the matter, throughput on the east-west pipeline has approached 6 million barrels per day, about 86% of its 7 million barrel design capacity; after deducting West Coast refinery use, crude available for export is about 4.5 million barrels per day. At the same time, France is pushing for coordinated release of strategic inventories, and the latest proposals under discussion include Europe releasing about 50 million barrels of diesel and International Energy Agency members releasing about 50 million barrels of crude. It is worth noting that this corresponds to pipeline transportation recovery and reserve release plans, and cannot be directly equated with the G7 having already increased production or completed releases. However, it should be noted that U.S.-Iran military risks remain: the United States continues to send additional military forces to the Middle East, while Iran is preparing to expand retaliation if attacked. The current oil price pullback reflects the suppression of risk premium by improved supply, and is not yet enough to show that geopolitical conflict has de-escalated. From "Valuation Killing" to "Earnings Taking Over": JPMorgan Sees a New Round of Momentum for the Technology Sector! Can AI Computing Power Themes and Broader Tech Stocks Enter the "Earnings Takeover Zone"? These changes do indeed constitute important positive catalysts for global technology stocks, especially semiconductors and AI computing power themes. The core point is that the valuation headwind to earnings growth is easing. But the reaction of global long-term bonds needs to be distinguished by trading session: U.S. Treasuries have clearly fallen after the nonfarm data; the UK 10-year gilt yield had already fallen to about 5.33% in early European trading along with oil prices. The main trading session for Japanese cash bonds ended earlier than the release of U.S. nonfarm payrolls, and the local 10-year yield remained near a high of about 3.1%, affected by Tokyo inflation and Bank of Japan policy expectations. Therefore, it can now be confirmed that pressure in U.S. and European bond markets has eased, and the probability of easing pressure in the Japanese bond market is also rising sharply. For popular stocks related to the AI computing power theme, a major decline in rates/U.S. Treasury yields first improves the present value of future cash flows. AI infrastructure construction requires upfront investment in chips, servers, networking, and power facilities, with cloud service revenue realized gradually afterward; the later the returns are realized, the more sensitive they are to the discount rate. Previously, even if orders and earnings expectations continued to be revised upward, they could be offset by P/E contraction. Now, if long-end rates stabilize and fall while earnings forecasts continue to be raised, share prices have a chance to enter the "earnings takeover zone"gains can rely more on profit growth while also receiving support from reduced valuation pressure. A recently released bullish research report from JPMorgan happens to provide industry and valuation support for this judgment: JPMorgan calculations show that the forward 12-month P/E ratio of the Magnificent Seven relative to the broader market is at a ten-year low, while forward 12-month EPS forecasts for semiconductors have been raised by about 30% since June. The former means the relative valuation premium has clearly contracted, while the latter means earnings expectations are still expanding. Combined with falling rates, this is conducive to the market refocusing on "whether earnings upgrades can continue," which is also why semiconductors may benefit more than tech sectors lacking earnings improvement. From the perspective of actual AI inference engineering, the agent applications represented by Muse and Astra push computing demand from "generating one answer" to "continuously completing a task." A task often includes planning, retrieval, tool invocation, code execution, and result verification, and may involve retries and multi-agent collaboration. This simultaneously increases demand for accelerator computing, CPU task orchestration, memory and context caching, persistent storage, and network data exchange, extending the beneficiaries of AI infrastructure investment from GPUs to server CPUs, HBM and DRAM, enterprise SSDs, and high-speed interconnects. Multi-step execution and tool loops are important features that distinguish agent workloads from simple question-and-answer interactions. Lower rates may also improve the investment economics of data center projects: with credit spreads remaining stable, a decline in the risk-free rate helps lower new financing costs and raise project net present values, making some expansion plans near the return threshold easier to advance. This forms a positive chain worth watching: expanded agent use brings computing demand, eased financing pressure improves construction feasibility, and improved delivery and utilization then revenue and profit. The research report released by JPMorgan shows that the overall valuation adjustment of the major U.S. technology giants that carry heavy weight in U.S. equities (the Magnificent Seven, or Mag 7) may be largely complete, and earnings growth is expected to once again become the main force supporting share prices. JPMorgan said the forward 12-month P/E ratio of the Magnificent Seven relative to the broader market has fallen to about one standard deviation below the historical median, at a ten-year low. Apple, Microsoft, Google parent Alphabet, Amazon, Meta, Nvidia, and Tesla are not only important components of the market-cap-weighted U.S. equity index, but also influence global AI investment expectations through AI chips/AI semiconductors, cloud computing, AI applications, and broad end-market positioning. Previously, through the end of 2025, the roughly $30 trillion market value expansion during the three-year super bull market in the S&P 500 was importantly driven by the Magnificent Seven and the broader AI computing power infrastructure supply chain. Therefore, changes in the earnings and valuations of the Magnificent Seven can simultaneously affect benchmark index performance, global equity market risk appetite, and the growth prospects of the AI computing power industry chain. The JPMorgan strategist team believes that as bullish positioning in the U.S. technology sector becomes less crowded, earnings performance remains strong, and valuations become more realistic, technology stocks will regain some of the momentum lost since the end of the first half, and therefore advises investors to re-enter the sector during market pullbacks. Technology stocks are still leading the S&P 500 by a wide margin this year, but the rally has cooled in recent months amid concerns that massive AI spending may not deliver the returns optimists assume. Within the technology sector, the valuations of the U.S. "Magnificent Seven" are at their lowest level in 10 years, while semiconductor stocks are emerging from a difficult stretchAnthropic's Dario Amodei and OpenAI's Sam Altman had both called for coordinated slowing of advanced AI development, worsening the sector's troubles. "We doubt there will ultimately be a clear slowdown, because this race remains an existential, winner-take-all contest," wrote the JPMorgan equity strategy team led by Mislav Matejka. JPMorgan said that although the kind of gains seen in the first half are unlikely to recur, major opportunities still exist. The Magnificent Seven's valuations being at a ten-year low is not an isolated judgment by JPMorgan. Data from Morgan Stanley Wealth Management's Global Investment Committee show that the valuation premium of the "Magnificent Seven" relative to the other 493 stocks in the S&P 500 is currently only 10%, the lowest in more than a decade, while the seven giants as a group still have an annual earnings growth advantage of about 45%. "By comparison, we think these hyperscalers now look downright cheap," Lisa Shalett, chief investment officer at Morgan Stanley Wealth Management, wrote in a report.