Guide
The mid-term elections triggered turmoil in the US stock market, with underlying issues such as the AI "transitioning" and high investment levels.
This rapid clearance is not a bad thing in the long run, as it helps to digest overly high expectations and positions. After all, the current valuation of US stocks is not high, and this year's gains are driven by earnings. However, short-term catalysts are also needed.
The current market needs two catalysts: either a significant breakthrough and progress in AI, or the Federal Reserve ends its tightening or even shifts to easing; these two catalysts are mutually causative.
The first quarter of next year is a key point: 1) A clear shift by the Federal Reserve is expected to occur early next year. 2) The earnings report disclosure in January may still face pressure, and more AI advancements will be needed by then; 3) The new US Congress will take office on January 3rd, making policy progression order and priorities clearer.
Before these catalysts appear, we still recommend watching cautiously in the short term, as the odds are average, controlling positions, and adjusting structures, while focusing on directions that do not rely heavily on high growth expectations within technology. In the short term, we should pay attention to whether negative feedback in liquidity will appear and whether policies will further disturb the market.
Text
Main Body
Recently, US stocks have shown significant volatility, with policy "noise" after the mid-term elections triggering market turmoil (《The Asset Implications of the Democratic Party's "Sweep"》). The key issue is that AI currently faces numerous "problems," including disconnect in demand and high investment, which have been prematurely exposed (《How to View the Plummeting US Stock Market?》).
Of course, this rapid clearance may not be a bad thing for the long-term trend, as it helps to digest excessive expectations and positions, which could be advantageous for future rebounds when new catalysts emerge, though we must wait for new catalysts to appear.
It is important to note that the gains in US stocks this year have mostly come from earnings contributions, rather than relying on valuation expansion, and after recent corrections, valuations have dropped sharply. Except for the Dow's 19.6 times valuation, which remains at the 86th percentile of the past 20 years, the S&P 500 and Nasdaq valuations have fallen to 17.8 times and 20.3 times, corresponding to the 70th percentile and 44th percentile of the past 20 years, respectively. The Philadelphia Semiconductor Index even dropped to 13.8 times, only 14% of the past 20 years' percentile.
Chart 1: Valuations of the S&P 500, Nasdaq, etc. have clearly fallen, and the Philadelphia Semiconductor Index has even dropped to the 14th percentile of historical valuations.
Data Source: FactSet, CICC Research Department
This year, the "K-shaped differentiation" of the US fundamentals has caused the market to concentrate and rely on a few directions: 1) The US Mag7 contributed 22% of the S&P 500 gain since 2026, contributing as much as 40% before the drop in mid-October; 2) The five major cloud vendors accounted for 20% of the S&P 500 index earnings in the first three quarters of 2026, contributing 25% to S&P 500 earnings growth; 3) The concentration effect of leading companies remains, with the S&P 500 rising 8% since the beginning of the year, but the proportion of stocks above the 200-day moving average has dropped from 60% at the beginning of the year to 55%, with a notable divergence in August and September. However, the current lack of models and application-level catalysts in AI, along with policy uncertainty, has led to traditional demand being suppressed by high-interest rates. In other words, the main directions face challenges, while traditional directions are yet to take over, explaining the overall turbulence.
Chart 2: The US Mag7 contributed 22% to the S&P 500's gains and losses since 2026.
Data Source: Bloomberg, CICC Research Department
Chart 3: The five major cloud vendors contributed an average of 37% to the S&P 500 earnings growth in the first half of the year.
Data Source: FactSet, CICC Research Department
Chart 4: Market breadth also showed a significant narrowing in August, with indices maintaining high levels but the proportion of constituent stocks above the 200-day moving average dropping clearly.
Data Source: FactSet, CICC Research Department
Thus, the market needs two catalysts: either AI achieves a major breakthrough, or the Federal Reserve ends its tightening or even shifts to easing. Before this happens, we recommend cautious observation: 1) Monitor whether there is short-term negative feedback on liquidity or further policy disruptions (the negotiation before the new Congress takes office early next year); 2) Control positions to preserve room for future allocations; 3) Adjust structures, even within technology, it is advisable to focus on cloud vendors and computing chips that either have low AI exposure or that have low valuations and greater pricing power.
The first quarter of next year is a key point: 1) The opportunity for a noticeable shift by the Federal Reserve is expected by early next year. The strong non-farm payroll data in October has reinforced the likelihood of a December rate hike, making it more certain that inflation will drop early next year from a high base; 2) The earnings report disclosed in January may continue to pressure ROIC for technology companies, and AI will need to demonstrate new advancements by then; 3) The new Congress in the US will be inaugurated on January 3rd. The policies and statements from Trump before the inauguration of the new Congress may continue to disrupt the market, leading to a need for digestion; however, the market may gradually desensitize after the new Congress takes office. So, by that time, either AI must show progress, or the Federal Reserve's tightening expectations must decline, potentially providing support or opportunity.
We have summarized the policy implications of the mid-term elections in 《The Asset Implications of the Democratic Party’s “Sweep”》 and the liquidity disturbances after market turmoil in 《How to View the Plummeting US Stock Market?》. In this article, we further address the outlook for the US stock market from a fundamental perspective in conjunction with the earnings of US stocks for the third quarter, which have basically been disclosed.
Market Drivers: Earnings are the Dominant Force, AI Contributes 60% of Growth; Growth Rate Shows an Inflection Point
First, strong earnings growth is the main driver of US stocks. From the beginning of the year to now, the gains of the major US stock indices have come from earnings contributions. The S&P 500, Dow Jones, and Nasdaq indices have increased by 8%, 7%, and 6%, respectively, with dynamic earnings growing by 29%, 17%, and 40%, offsetting valuation contraction. In the third quarter, the S&P 500 earnings grew by 25%, a decline compared to the 32% and 56% growth in the first and second quarters, but still above the average quarterly earnings growth of 12% during 2023-2025.
Chart 5: The gains in US stocks this year have primarily been from earnings contributions rather than relying on valuation expansion.
Data Source: FactSet, CICC Research Department
Secondly, AI-related sectors contributed 60% of earnings growth. By estimating AI sector profits from communications services, information technology, and Amazon, they contributed 62% to overall earnings growth in the third quarter, down from 70% in the second quarter. Among them, information technology contributed 58%, supporting the majority of growth in the AI sector, while communications services saw a significant decline in contribution due to a sharp slowdown in the media and entertainment sector. This means that overall earnings growth is reliant on AI; if the AI sector is excluded, the earnings growth in the third quarter would likely revert to 15%.
Chart 6: Information technology contributed 58% of this quarter's earnings growth, while the contribution from communications services dropped from 24% in the second quarter to 4%.
Data Source: FactSet, CICC Research Department
Chart 7: AI-related sectors contributed 60% to the overall earnings of the US S&P 500.
Data Source: FactSet, CICC Research Department
Traditional demand showed decent earnings in the third quarter, but high-interest rates still drag down profits. In terms of growth rates, sectors such as energy (150%), semiconductor equipment (110%), real estate management and development (49%), and capital goods (48%) led, while transportation (-37%), telecommunications services (-34%), financial services (-14%), and insurance (-11%) lagged. In terms of contribution, sectors including industrials, materials, utilities, and consumer staples collectively contributed 10% to earnings, while both financial and real estate contributions turned negative, dragging down overall growth by 7.5%.
Third, the growth rate shows an inflection point, marking the first slowdown since Q2 2025. The S&P 500's earnings growth of 25% in the third quarter decelerated compared to the first two quarters of the year. The market is becoming more "discriminating" regarding performance, not only demanding sustained high growth rates but also needing continuous and unexpected acceleration, which increased the difficulty of exceeding expectations for the third quarter reports. The proportion and magnitude of earnings exceeding expectations for S&P 500 constituents both declined, with the ratio slightly dropping from 86% in the second quarter to 84%, while the degree of over-expectation fell sharply from 27% in the second quarter to 3.5%, the lowest level since Q4 2022. Among them, the communications services sector, including Google and Meta, saw the largest decline, with earnings changing from 99% above expectations in Q2 to 19% below expectations in Q3.
Chart 8: The proportion of S&P 500 companies exceeding expectations dropped slightly from 86% in Q2 to 84%, with the magnitude of exceeding expectations plummeting to 3.5%.
Data Source: FactSet, CICC Research Department
AI Direction: Income Trails Investment; Returns Decline, but Still Exceed Costs; Can Withstand Interest Rates Dropping to 6~6.5%
1) Income lags behind investment. Capital expenditures are accelerating, free cash flow has turned negative, and external financing pressures are rising. The five major cloud vendors' total capital expenditure in the third quarter increased from $181.5 billion to $230 billion, with the year-on-year growth rate accelerating from 87% in the second quarter to 118%. However, the growth rate of operating cash flow significantly slowed, dropping from 38% in the second quarter to 19% in the third quarter, totaling $198 billion, which is insufficient to cover the capital expenditures during the same period. This has caused the five major cloud vendors' free cash flow to turn negative, from $4.8 billion in the second quarter to -$31.8 billion in Q3. Except for Microsoft, Amazon, Meta, Google, and Oracle all had negative free cash flows in the third quarter, further increasing the demand for external financing. In the context where AI demand has not yet broken through the existing application scenario's “ceiling” and investment return outlook remains unclear, market scrutiny of financing costs and sources of funds has significantly increased, with credit risk concerns being amplified, as the CDS rates of the five major cloud vendors have all risen to historical highs.
Chart 9: The free cash flow of the five major cloud vendors turned negative from $4.8 billion in Q2 to -$31.8 billion in Q3.
Data Source: FactSet, CICC Research Department
Chart 10: The CDS of the five major cloud vendors have also risen to historical highs.
Data Source: FactSet, CICC Research Department
2) Returns exceed costs. Marginal ROIC has further decreased, but overall returns can still cover financing costs. With continuous capital investment expansion, the new operating returns have not kept pace, leading the five major cloud vendors' marginal ROIC to decline from a high of 44% in Q2 2024 to 6.1%, causing the overall ROIC to further drop to 15%. However, the overall ROIC is still above the current weighted average cost of capital (WACC) of about 10% for the five major cloud vendors, indicating that cloud vendors can still achieve excess returns.
Chart 11: The overall ROIC of the five major cloud vendors declined from 16% in Q2 to 15%, and the marginal return rate also dropped to 6.1%.
Data Source: FactSet, CICC Research Department
The decline in marginal returns has led to a decrease in the interest rate level that AI can bear to between 6% and 6.5%. Previously, we calculated that if the five major cloud vendors maintained stable ROIC, assuming the difference between ROIC and WACC narrowed to the historical average of 8 percentage points before the rise of AI trends in 2015-2023, the bearable risk-free interest rate would be 7-7.5% (《AI: From "Friend" to "Enemy" of Traditional Demand?》). However, with further weakening of marginal investment returns in Q3, the gap between ROIC and WACC has converged to historical averages, thus relegating the bearable risk-free interest rate level for AI to between 6% and 6.5%.
Chart 12: With the return expectations for 2027, the bearable risk-free interest rate level for AI has reduced to between 6% and 6.5%.
Data Source: FactSet, CICC Research Department
3) Cloud vendors' "heavy assetization" is putting pressure on capital turnover rates. Further decomposing ROIC into after-tax operating profit margin (NOPAT/Revenue) and capital turnover rate (Revenue/IC), the after-tax operating profit margin of the five major cloud vendors has been continuously increasing since 2023, rising from 14.8% to 22.1% in Q3; however, the capital turnover rate has been declining, dropping from a high of 109% in Q1 2023 to 66% in Q3, marking a new low since 2015. At the same time, since Q2 2025, the slowdown in margin improvement has been unable to offset the drag from capital turnover rate, leading to a continuous decline in the ROIC of the five major cloud vendors.
Chart 13: Since Q2 2025, improvements in margins have not been able to offset the negative drag from capital turnover rates.
Data Source: FactSet, CICC Research Department
Further differentiation within technology, parts of the hardware segment still need to digest growth expectations, and cloud vendor pricing is relatively reasonable. Some hardware infrastructure sectors with high growth expectations, such as network interconnection/optical modules, semiconductor manufacturing, liquid cooling/temperature control, have recently shown a significant valuation correction; for example, the valuation of the semiconductor manufacturing sector has quickly dropped from the 90th percentile in early October to around 20%. In contrast, cloud vendors have not provided noticeable valuation premiums for new profits, and the valuation adjustment space is relatively limited.
Traditional Sectors: Interest-sensitive areas under pressure from high-interest rates, and the squeezing from AI is also gradually showing.
Traditional demand has not formed comprehensive improvement. In terms of consumer spending, service consumption is stronger than goods consumption. Earnings from consumer services, food and beverage, and automotive parts grew by 12%, 8.7%, and 7.3%, respectively, while durables and apparel, household and personal care fell by 5.4% and 10.3%. The industrial sector shows even greater divergence, with capital goods earnings growing by 48.3%, commercial and professional services growing by 16.2%, yet transportation, affected by high oil price costs, decreased by 37.1%. In the financial sector, bank earnings saw a slight increase of 4.8% supported by net interest income, while insurance and financial services fell by 10.7% and 13.6% due to investment profit and loss volatility.
Growth remains resilient, but growth areas sensitive to interest rates continue to suffer. Recent macro data reflects that the current US growth still has resilience, with a Q3 annualized growth rate of 3.4%, exceeding expectations of 3.2% and Q2's 2.2%. Aside from AI-related investments, personal consumption remains strong, particularly in medical, dining, and accommodation services. The October ISM manufacturing and non-farm payrolls also greatly exceeded expectations, with the unemployment rate declining from 4.2% to 4.1%. However, high interest rates continue to suppress housing purchases and credit demand, with the 30-year mortgage rate consistently increasing since the beginning of the year, reaching as high as 7.4% in October. Real estate sales remain at low levels, and although existing home sales saw a slight month-on-month increase in September, the annualized sales were only 3.99 million units, and new home sales dropped to 676,000 units. On the other hand, the weakening returns from AI might also begin to "crowd out" the physical and financial resource factors of traditional demand. As emphasized in 《AI: From "Friend" to "Enemy" of Traditional Demand?》， the positive externalities brought by AI demand previously supported US economic growth, but as AI requires more external capital to support investment, the crowding-out effect on traditional demand is also unavoidable, with AI issuing large amounts of debt and the Federal Reserve still needing to raise interest rates being a typical example.
Chart 14: Interest-sensitive areas like housing sales have continued to suffer since 2023.
Data Source: FactSet, CICC Research Department
A larger magnitude of repair still requires interest rates to drop to cooperate. We previously estimated using real estate and non-AI companies that traditional demand could withstand a 10-year Treasury yield of about 4%. Although there is resilience, high interest rates are continuing to squeeze the release of residential housing and small business credit demand, reflected as insufficient credit expansion momentum in the household sector. In the short term, the resilience of growth and employment data means the Federal Reserve lacks immediate motivation to shift, but the weakness of traditional demand does not support significant rate hikes by the Federal Reserve. If AI expectations continue to weaken in the first quarter of next year, the market's tightening expectations for the Federal Reserve are likely to decline, creating conditions for interest rates to fall and traditional demand to recover.
Prospective Judgment: The market needs a catalyst for the end of tightening or advancements in AI, with early next year being a key node.
Currently, the market's expectations for AI are in a "transitioning" phase, with the critical issue being the absence of new narratives and catalysts without demand and models, which also amplifies the risks of credit and regulatory pressure. High levels of capital expenditures continuously squeeze the free cash flow of cloud vendors, leading to rising concerns regarding the speed of realizing AI revenue and investment returns - a situation reminiscent of the market's fears over private equity and credit risk late last year when AI developments were insufficient.
In this context, the market needs two catalysts: either AI must achieve significant breakthroughs, or the Federal Reserve must end tightening or even shift to easing. If neither occurs, it may affect investors' willingness to re-enter the market and increase volatility. In the short term, we need to focus on whether negative feedback on liquidity arises or any further policy disruptions (if Trump will utilize this window for policy negotiations before the new Congress takes office early next year). The first quarter of next year is a key verification point, with the Congress being inaugurated, earnings periods featuring AI realizations, and inflation changes expected.
Until then, we still recommend cautious observation in the short term, and before the aforementioned catalysts appear, closely monitor to see if negative liquidity feedback arises and whether policy disturbances will further influence the market. Currently, the S&P 500 index has returned near the annual line, with previous adjustments having released some emotions, but the market still lacks new upward catalysts, and liquidity factors, along with the Federal Reserve's rate hikes in December, may bring further disturbances. Therefore, we suggest waiting for the emergence of the aforementioned catalysts and clearer prospects for investment returns before proceeding with allocations. In extreme cases, if the additional earnings growth brought by AI expectations is completely "erased," the dynamic earnings growth rate may fall to 10%, and if valuations simultaneously decrease to 16-17 times, it would correspond to about a 10% decline in the S&P 500 index.
Aside from managing positions well, structural adjustments to holdings can also be made, even prioritizing focusing on cloud vendors with low AI exposure or that have low valuations and greater pricing capabilities within technology. Recent market performance has reflected this difference, with segments that previously incorporated higher growth expectations showing more apparent pullbacks, while cloud vendors have remained relatively resilient. Based on the average valuation over the past 10 years, sectors such as optical modules, semiconductor manufacturing, and grid and distribution will still require 40-70% profit upgrades to match current pricing, while profits of cloud vendors retain nearly 20% of downward adjustment space.
Chart 15: Overall valuations of US stocks are not currently at extreme highs, but there is significant differentiation within the AI industry chain.
Data Source: FactSet, CICC Research Department
Chart 16: Sectors like optical modules, semiconductor manufacturing, and grid and distribution that have accounted for many high growth expectations need profit upgrades of 40-70%, while cloud vendors have retained nearly 20% room for subsequent profit growth.
Data Source: FactSet, CICC Research Department
Source
Source Acknowledgment
This article is extracted from: "Has the US Stock Market Hit Its Bottom?" published on November 9, 2026.
Liugang, CFA Analyst SAC Certificate Number: S0080512030003 SFC CE Ref: AVH867
Yang Xuanting Analyst SAC Certificate Number: S0080524070028 SFC CE Ref: BXS665
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。
