The Data Weekly from Guotai Haitong's Xiangjiang Strategy Review comprises three sections: "Key Data and Market View Updates," "Highlights of Major Research Reports," and the "Daily Chart Series," designed to present weekly data trends and core viewpoints through key market indicators, selected research insights, and market chart analysis.
Key Data and Market View Updates
Review: The "Summer Cold Wind" Has Materialized as Expected, the AI 1.0 Rally Peak Has Passed
The Haitong International strategy team first issued the "Summer Cold Wind" warning on May 6 and continued publishing a series of reports under the same theme, persistently alerting that global stock markets faced significant correction risks. On August 16, in "Autumn Rally Will Shift from Deep Rebound to Choppy Divergence," the team again proactively warned of the volatility risks in the autumn rally: the 10-year US Treasury yield had already broken through its near three-year high and could reach extreme levels of 5% to around 5.3% in the coming weeks. On October 25, in "If the Late-Autumn Rebound Is Merely a Bear Market Bounce, Optimizing Holdings Is Key," a medium-to-long-term logic adjustment was proposed: the narrative for AI hardware has changed, risk premiums are rising, AI hardware valuations are contracting, and the global AI hardware bull market may have turned bearish. In the short term, US Treasuries remain favored, with opportunities to capture yield declines from high levels. On November 1, "Endure, and Endure Toward Long-Term Value" suggested, based on historical experience, that once pessimistic expectations become consensus, market pessimism may show signs of spiral self-reinforcement in the short term, potentially forming an oversold condition that warrants caution. On November 9, "The Silver Lining of Dark Clouds: Tech Narrative Reconstruction, Value Returns Prevail" specifically addressed the AI rally further: the AI wave 1.0 peak has passed, entering a transitional period of scarcity and moving toward 2.0. Tech giants continue to increase capital expenditure yet face rising financing costs, declining shareholder returns, and potential policy restrictions on data center construction. Meanwhile, AI applications remain in their infancy. Although the long-term AI industry trend has not ended, this round of AI hardware turning from bull to bear is not a short-term earnings problem but more importantly an adjustment in expectations and valuations.
Last week, US stocks experienced a choppy ride. Early in the week, falling oil prices and strong cloud computing earnings drove a rebound. However, renewed rate hike expectations suppressed valuations, and the Jensen Huang incident added uncertainty to Nvidia's ecosystem financing and supply chain delivery, causing computing hardware to give back some gains later in the week. A-shares showed structural resilience, with domestic computing power and certain value sectors providing support, while pressure on the overseas computing chain was concentrated in the latter part of the week. Hong Kong stocks were hit by both tech gaming risks and rising overseas interest rates, with weakening internet heavyweights amplifying index declines.
Outlook: Medium-Term Focus on Tech Narrative Reconstruction and Shareholder Returns, Short-Term Focus on Political Gaming and Market Stabilization Mechanisms
1. The "Scarcity" Transition from AI 1.0 to 2.0 Has Begun, and Transition Periods Often Involve Turbulence
In the just-released 2027 annual strategy report "The Silver Lining of Dark Clouds," it is emphasized that an important feature of this AI wave is that capital markets drive AI capital expenditure, which then promotes economic growth through industrial expansion and wealth effects. Now, this logic is being disrupted: investors demand higher risk compensation, political and financing environment constraints are rising, and AI application commercialization is not yet sufficient to support massive infrastructure investment. Therefore, the AI 1.0 peak has passed, and the "scarcity" transition toward AI 2.0 has begun. Looking ahead to 2027, it is important to focus not only on whether commercial returns can support capital expenditure but also on whether companies can continue to reward shareholders.
1) From AI 1.0 led by AI infrastructure to AI 2.0 led by AI application profitability, the transition period may experience significant turbulence. The future development of the AI rally depends on whether massive investments can truly translate into commercial returns. Drawing from history, during the transition from Internet 1.0 to 2.0, equipment and R&D, as well as intellectual property investment growth, experienced deep adjustments. The long-term vitality of a tech wave does not eliminate periodic large fluctuations in secondary market performance. AI applications remain in their infancy, yet capital expenditure has already reached a massive scale. The core issue during the transition is whether applications can drive sufficient demand and create sufficient returns. As computing costs decline and applications gradually materialize, AI 2.0 is expected to form a new pattern dominated by applications with hardware as a supplement, at which point hardware valuations may be restructured again as real demand is unleashed. Referring to history, during the "scarcity" AI tech transition period, the AI rally may experience a turbulent phase.
2) US political changes and shareholder returns jointly constrain AI capital expenditure. First, this round of AI expansion relies heavily on capital market support. When investors begin to demand clearer returns and more adequate risk compensation, capital expenditure cannot rely solely on continuously upward revisions of forward demand expectations. Unless AI applications significantly exceed expectations, caution is warranted regarding tech giants' capital expenditure falling below previously optimistic forecasts. Second, the political landscape reshaping after the US midterm elections casts a shadow over AI tech development prospects. The political gaming and supply chain shocks surrounding open-source and closed-source large model development cannot be underestimated, especially for companies in the computing power supply chain that previously received credit enhancement from Jensen Huang, making them even more vulnerable under high interest rates and tighter financing conditions. Additionally, data center construction involves local approvals, energy supply, and policy support. Political shifts in some states may affect project progress and constrain cloud vendors' investment pace. Third, the US stock market's culture of valuing shareholder returns will become a deeper constraint on the AI rally in 2027. In the first half of 2026, the five major cloud vendors' capital expenditure grew 88.4% year-over-year, buybacks decreased 74.5%, net equity returns (buybacks plus dividends minus stock issuance) turned negative, and debt financing increased. Tech giants still have financing capacity, but when free cash flow comes under pressure and buybacks shrink, shareholders are less willing to pay for large-scale capital expenditure. Stock price adjustments will in turn affect financing capacity and willingness, ultimately constraining capital expenditure. The AI bull market logic of "brute force works miracles, relying on large-scale investment" over the past few years has been questioned.
2. The Path to Victory in 2027 Depends More on Shareholder Returns
Looking ahead to 2027, finding markets, industries, and companies with high shareholder returns that can cover financing and share unlock reduction pressures is an important criterion running through US and China asset selection. How much new capital a company's growth requires, whether financing dilutes earnings per share, and whether dividends and buybacks can cover financing needs are becoming important bases for valuation divergence. First, US stocks in 2027 will demand more shareholder returns than in 2026, particularly requiring tech giants to increase shareholder returns and constraining capital expenditure through stock prices and financing conditions rather than continuing unconditional support for capital expenditure. AI will remain an important growth engine for the US economy in 2027, but both the direct boost from infrastructure investment and the wealth effect from the tech rally and its support for consumption may weaken compared to 2026. The overall shareholder return yield for US stocks declined significantly in 2026, already at a near 10-year low. The combined buyback and dividend yield for the S&P 500 is only 2.7%, noticeably below the 10-year US Treasury yield for the same period. US equity financing expanded rapidly in 2026, already changing the long-term net buyback funding pattern. Federal Reserve Z.1 data shows that the trend of negative net stock issuance by US non-financial corporations for 32 consecutive years ended in the first half of 2026, with the two-quarter seasonally adjusted annualized net issuance averaging approximately $409 billion. This full-caliber statistic includes both listed and unlisted enterprises, reflecting very strong financing demand driven by the AI wave, which may further transmit to the secondary market as large AI companies go public. Second, China's stock market should seek reliable investment returns in pessimistic areas, prospecting in fields where shareholder returns can cover financing and share unlock reduction. Hong Kong stocks in 2027, especially in the first quarter, similarly need to focus on financing and share unlock requirements for capital absorption. In the first three quarters of 2026, Hong Kong's information technology and industrials sectors together contributed about 65% of industry financing amounts; financials, telecom, energy, and utilities together contributed about 50% of dividends and buybacks, with dividends and buybacks significantly exceeding equity financing. Stock selection should consider not only earnings but also the relationship between companies' capital demands from the market and their returns to shareholders. A-share value opportunities come not only from earnings growth but also from companies converting more operating results into per-share returns. The domestic low interest rate and asset scarcity environment continues, and quality companies that can sustainably pay dividends and conduct buybacks have strong allocation value. It is also important to see whether dividends and buybacks can cover financing and shareholder reductions, and whether returns are supported by earnings and free cash flow.
3. Short-Term Market Direction Depends on Risk Appetite Changes: For US Stocks, Watch the Jensen Huang Investigation; For Chinese Stocks, Watch Stabilization Mechanisms During APEC
US Stocks: Nvidia influences AI computing chain investment and financing; watch whether the impact of US political changes can converge. Nvidia sits at the center of US AI industry chain investment and financing, supporting customer expansion through equity investments, cloud service procurement commitments, and more. As of July 26, 2026, its ecosystem equity investments reached $99 billion, equity investment commitments were $25 billion, and cloud service commitments to certain AI cloud partners totaled $36 billion, tightly linking chip orders, customer financing, and computing procurement. Market concerns are that the investigation and the new management's higher investment return and compliance thresholds may weaken customer financing capacity, thereby delaying chip procurement, data center construction, and supporting equipment delivery. Therefore, one important opportunity for whether US stocks can shift from risk-off to risk-on in the short term lies in whether the scope of the Nvidia investigation and regulatory boundaries can be clearly defined, and whether uncertainty that is difficult for companies and financiers to price can be effectively reduced. A relevant US House committee has already linked the incident to multiple AI and chip regulatory bills. Follow-up attention should be paid to congressional progress in requiring Jensen Huang to testify on December 1. Two events next week will separately test capital support on the computing supply side and financing absorption on the application side. On November 17 after the US market close, Nvidia will release earnings, and attention should be paid to the new management's arrangements for ecosystem investment, customer financing, and existing commitments. On November 19, Anthropic is expected to go public; watch whether it creates a short-term market shock and whether it can stabilize financing expectations for AI applications and computing procurement.
China: Watch stabilization mechanisms during APEC, focusing on whether the "visible hand" can effectively improve market expectations. According to the CSRC's November 2, 2026 press conference, the People's Bank of China announced that the stock buyback and increase holding relending policy would shift from a phased arrangement to a long-term institutional arrangement, with the relending rate reduced to 1%. A CSRC official stated at the 2026 Shanghai Stock Exchange International Investors Conference held on November 12 that securities and fund companies have conducted a third batch of swap facility operations worth 100 billion yuan; since the beginning of the year, medium-to-long-term funds such as social security, annuity, and insurance have net bought over 600 billion yuan of A-shares. These developments indicate that comprehensive market stabilization support is being implemented. Looking ahead to the APEC period, if the "visible hand" stabilization mechanism can effectively and powerfully play its role, thereby breaking investors' pessimistic expectation of "meetings always bring declines," then the stabilization mechanism's pivotal effect of stabilizing expectations with minimal effort is expected to be established. This would help drive policy tool usage, listed company buybacks, and long-term capital entry to form a synergy, break the self-reinforcement of pessimistic expectations in a declining market, and reduce the impact of overseas tech volatility, geopolitical policy, and economic growth momentum uncertainties on China's stock market.
Investment Strategy: Look Long, Act Short; Short-Term Contrarian Thinking to Trade Weak Market Rebounds, Long-Term Value Discovery Based on Shareholder Returns
1. How to Apply Contrarian Thinking in Short-Term Weak Market Rebounds?
When market sentiment indicators are at extreme levels, follow the rhythm of the "visible hand" in an orderly manner. Recently, China's stock market sentiment indicators remain sluggish, but stabilization mechanisms are continuously being improved and reinforced by policy. Considering the APEC leaders' summit in Shenzhen in mid-November, it is recommended to closely follow the rhythm and intensity of the "visible hand" for orderly positioning. Key time points for this round of China's stock market rebound since late October include the APEC summit, the December G20 leaders' summit, the Federal Reserve rate meeting, and China's December Central Economic Work Conference. Whether the long-standing psychological suggestion of "meetings always bring declines" in the A-share market can be broken by stabilization mechanisms in a weak market environment will determine whether this rebound can form sustainable risk appetite improvement and actionable profit realization. If A-share risk appetite can be boosted by stabilization mechanisms, it will help drive Hong Kong stocks, at least bringing potential momentum for short covering. Currently, Hong Kong's overall short-selling turnover ratio remains at 20%, still at a historically relatively high level, while southbound funds and foreign capital have begun buying. Subsequent short covering is expected to provide additional momentum for the rebound. After a sustained decline, rebounds with more obvious elasticity may come from directions with healthier chip structures. Recommendations: 1) Prioritize sectors with relatively optimistic industry prospects such as biopharmaceuticals, defense and military, embodied intelligence, and high-end equipment. 2) The previously hard-hit AI hardware rally has entered a differentiation and rebound phase. It is recommended to carefully select within the self-controllable domestic chain rather than blindly bottom-fishing, and especially to guard against cross-market negative feedback: financing positions in Japan and Taiwan remain high, and once the Philadelphia Semiconductor Index continues to weaken, passive deleveraging in Asia's hardware chain may in turn intensify global AI hardware selling.
2. How to Discover Value Based on Shareholder Returns in the Medium-to-Long Term and Optimize Holdings During Short-Term Rebounds?
First, from a medium-to-long-term perspective, leading central state-owned enterprises with stable dividends remain the underlying assets for dividend allocation. Varieties with high crowding after previous gains should wait for more suitable buying prices. For mature industry leaders, focus on earnings stability, declining capital expenditure, and willingness to pay dividends. Whether shareholder returns can be sustained determines whether the safety cushion for value assets is reliable. Retain "friends of time" that can withstand long-term scrutiny, and patiently wait for medium-term buying opportunities after adjustments. Banks and dividend stocks are currently relatively crowded, with crowding exceeding February 2024 levels, and attention should be paid to the risk of relative return giveback. In a weak market environment, short-term adjustments brought by market style shifts instead provide better medium-term allocation opportunities. From a medium-to-long-term perspective, quality core assets with good cash flow and dividend foundations are the "base positions" in a low interest rate and asset scarcity environment. Second, "quasi-convertible bond" high-quality strong players—find "last man standing" winners in "traditional industries" related to domestic demand and overseas expansion chains. Prospect in overlooked mature industries and select "winners" with improving free cash flow. Industry demand may not accelerate significantly, but after survival-of-the-fittest, leaders that have established brand, channel, technology, or scale advantages—as long as they can maintain relatively stable earnings while reducing new capacity construction and other capital investment—may actually see improved free cash flow, creating conditions for increased dividends and buybacks. What is valued is these "last man standing" companies' ability to accumulate cash and enhance shareholder returns in a low-growth environment. Third, look long, act short; during the transition from AI 1.0 to 2.0, optimizing holdings and positions during short-term AI rebounds may have a higher success rate. The global AI 1.0 peak has passed, long-term optimistic fundamental expectations have been shaken, micro-level capital crowding in the AI hardware rally has not been thoroughly cleared, and volatility in AI hardware tech stocks is notably higher than the broader market. Therefore, bottom-fishing AI hardware in the short term requires too high trading capability, and excessive optimism is not advisable. It is recommended to position on the left side and take profits promptly, and not to use leverage to chase highs; instead, based on medium-term trends, take advantage of each short-term rebound to optimize holdings and positions.
A-shares: From November 9 to November 12, A-share margin financing saw a net inflow of 19.5 billion yuan; on the ETF side, last week semiconductors, gold, banks, communications, and innovative drugs ETFs saw net inflows, while media, chemicals, artificial intelligence, robotics, and non-ferrous metals ETFs saw net outflows. Hong Kong's overall short-selling turnover ratio remained at 20%, Hang Seng Tech short-selling ratio rose to 21%, and internet leaders' short-selling ratio rose to 18%. Last week, southbound funds turned to a net inflow of 11.9 billion HKD, with the turnover ratio rising to 21.02%. At the individual stock level, SMIC and Zhipu continued net inflows of 1.5 billion and 1.6 billion HKD, Alibaba turned to a net inflow of 1.4 billion HKD, while Tencent and YOFC continued net outflows of 800 million and 300 million HKD. At the industry level, during November 5-11, e-commerce, industrials, and metals sectors led net inflows, with e-commerce turning from net outflow to net inflow; communications services, pharmaceuticals, and financials saw net outflows; real estate ended nine consecutive weeks of net inflows, turning to net outflow. From November 9-11, Hong Kong stocks saw net inflows of 191.4 billion HKD from foreign intermediaries, 5.5 billion HKD from local Hong Kong intermediaries, and 2.1 billion HKD from Chinese intermediaries. Domestic Hong Kong Stock Connect ETFs turned to a net inflow of 300 million yuan. Last week, Hong Kong's market saw 5.1 billion HKD in share unlocks, and next week is expected to see 1 billion HKD in unlocks.
Highlights of Major Research Reports
Overseas Macro Strategy: The Silver Lining of Dark Clouds: Tech Narrative Reconstruction, Value Returns Prevail — Yidong Zhang
Looking ahead to 2027, US economic growth may slow, and AI will remain an important growth engine for the US, but its boost to investment and consumption will weaken. After the midterm elections, the US political landscape will undergo major changes, and data center construction and AI capital expenditure growth will both face slowdowns. The AI infrastructure-led AI tech wave 1.0 may have passed its peak. In 2027, the AI tech wave will begin transitioning toward AI 2.0 led by AI applications. AI applications are expected to accelerate, but the market will pay more attention to whether commercialization returns can support large-scale capital expenditure and will focus more on secondary market shareholder returns. In 2026, US equity financing expanded rapidly, ending the 32-year trend of negative net stock issuance by US full-caliber non-financial corporations; US stock shareholder return yields declined significantly, at a near 10-year low, with tech giants shifting from continuously rewarding shareholders to net financing. In 2027, US political changes may lead to cooling Middle East tensions and falling oil prices. Combined with debt ceiling gaming, long-end US Treasury yields may begin a moderate decline after high-level turbulence in the first half. Additionally, comparing US macro factors with those of the Eurozone and Japan, the dollar is expected to oscillate downward in 2027. According to historical patterns, under the triple-weak scenario of a weakening US economy, weakening US Treasury yields, and a weakening dollar, globally interest-rate and exchange-rate sensitive elastic assets all benefit relatively, with low-valuation, fundamentally stable Asian emerging markets often outperforming developed markets.
Overseas Pharmaceuticals: Spring Blossoms, Autumn Harvest, Moving Toward the New — Hong Kong Pharmaceuticals 2026 Winter Strategy — Kehan Meng
In 2027, multiple positive catalysts for Hong Kong pharmaceutical companies—including profitability, BD overseas expansion, and AI technology upgrades—are expected to further materialize. It is recommended to focus on four main lines: innovative drug overseas expansion, CXO, AI drug discovery, and leading consumer healthcare or internet healthcare companies with stable performance. ① Innovative drugs: overseas expansion entering the realization phase; focus on leading Biotechs with key overseas data readouts and Pharma leaders with continuously rising innovative drug revenue share. ② CXO: global demand continues to prosper, orders continuously translate into performance; domestic demand sentiment is gradually recovering. ③ AI drug discovery: short-term wet lab validation demand materializes first; long-term model progress is expected to improve new drug R&D efficiency, reducing costs and increasing efficiency in early research and clinical stages. ④ Consumer healthcare and internet healthcare: domestic demand is gradually stabilizing, and valuations offer good value.
Overseas Energy Technology: Energy Industry Strategy Report: How Energy Technology Continues to Support AI Development Under New Circumstances — Baiqiao Xu
AI infrastructure: modularization is the guarantee under new circumstances. North American AI power consumption may accelerate its contraction toward "clean + stable" electricity. Energy storage is the "reservoir" for new energy and AI. Oil, gas, and chemicals: high cash returns resonate with spread repair.
Daily Chart Series
- Hang Seng Tech Expansion: Broader Industry Coverage, About 10% New Weight
Key Data Update
Last week, market expectations for easing US-Iran tensions initially pushed oil prices lower. On Thursday, Trump stated the US need not reach a deal with Iran, and oil prices rebounded, but still fell 2.06% for the week to $96.29 per barrel. US October CPI rose 0.4% month-over-month, above market expectations, with core CPI up 0.3% month-over-month. Combined with Fed Governor Waller's Tuesday comment that even if inflation improves, a 50 basis point hike should be considered in December. Rate hike expectations intensified, and the dollar index rose 1.10% for the week to 103.41. The 2-year US Treasury yield rose 33.66bp for the week to 4.844%, and the 10-year US Treasury yield rose 27.78bp to 5.135%. In precious metals, spot gold touched a more than one-year low during Thursday trading, with London gold falling 4.56% for the week to $3,883.98 per ounce and London silver falling 6.10% to $48.74 per ounce. Japan's 10-year and 30-year government bond yields rose 22.75bp and 18.35bp respectively, to 2.865% and 4.000%. The dollar was essentially flat against the offshore yuan at 6.6749. Last week from November 9-12, A-share margin financing saw a net inflow of 19.5 billion yuan; on the ETF side, last week semiconductors, gold, banks, communications, and innovative drugs ETFs saw net inflows of 6.5 billion, 1.8 billion, 1.2 billion, 1 billion, and 500 million yuan respectively, while media, chemicals, artificial intelligence, robotics, and non-ferrous metals ETFs saw net outflows of 400 million, 400 million, 700 million, 1.1 billion, and 1.6 billion yuan respectively. Hong Kong's overall short-selling turnover ratio remained at 20%, Hang Seng Tech short-selling ratio rose to 21%, and internet leaders' short-selling ratio rose to 18%. At the individual stock level, Meituan and Alibaba's short-selling turnover ratios fell to 16.3% and 12.3% respectively from the previous week, corresponding to rolling 1-year percentiles of 19.0% and 22.6%; Tencent, JD.com, NetEase, and Baidu's short-selling turnover ratios rose to 20.3%, 36.1%, 33.9%, and 23.2% respectively from the previous week, corresponding to rolling 1-year percentiles of 96.8%, 81.3%, 84.1%, and 38.9%. Last week, southbound funds turned to a net inflow of 11.9 billion HKD, with the turnover ratio rising to 21.02%. At the individual stock level, SMIC and Zhipu continued net inflows of 1.5 billion and 1.6 billion HKD, Alibaba turned to a net inflow of 1.4 billion HKD, while Tencent and YOFC continued net outflows of 800 million and 300 million HKD. At the industry level, during November 5-11, e-commerce, industrials, and metals sectors led net inflows, with e-commerce turning from net outflow to net inflow; communications services, pharmaceuticals, and financials saw net outflows; real estate ended nine consecutive weeks of net inflows, turning to net outflow. From the perspective of Hong Kong stock funding sources, last week (November 9-11), Hong Kong stocks saw net inflows of 191.4 billion HKD from international intermediaries, with cumulative net inflows of 499.1 billion HKD year-to-date; 5.5 billion HKD from local Hong Kong intermediaries, with cumulative net inflows of 71 billion HKD year-to-date; and 2.1 billion HKD from Chinese intermediaries this week, with cumulative net inflows of 98.6 billion HKD year-to-date. Since mid-October, international intermediary capital inflows have continued to accelerate. From October 15 to November 11, international intermediaries saw cumulative net inflows of 292.1 billion HKD, with incremental flows mainly concentrated in internet and biopharmaceutical leaders. Alibaba and BeiGene saw cumulative net inflows of 155.8 billion HKD and 33.5 billion HKD respectively, together accounting for 64.8% of international intermediary net inflows during the period; additionally, Xiaomi Group and Zijin Gold International saw net inflows of 12.44 billion HKD and 11.44 billion HKD respectively, while ICBC, China Feihe, Tencent Holdings, and Horizon Robotics saw net inflows of 5.02 billion HKD, 5.00 billion HKD, 4.91 billion HKD, and 4.89 billion HKD respectively. Domestic Hong Kong Stock Connect ETFs turned to a net inflow of 300 million yuan, with technology inflows of 700 million yuan; innovative drugs, dividend, consumer, and financials saw outflows of 300 million, 100 million, 40 million, and 30 million yuan respectively; since the beginning of 2026, the entire market has seen cumulative net outflows of 79.8 billion yuan, with broad-based, technology, dividend, consumer, financials, innovative drugs, and other pharmaceutical sectors seeing cumulative net outflows of 14.7 billion, 15.7 billion, 22.9 billion, 2.4 billion, 22.2 billion, 1.3 billion, and 600 million yuan respectively. Last week, Hong Kong's market saw 5.1 billion HKD in share unlocks, and next week is expected to see 1 billion HKD in unlocks. Since the beginning of 2026, Hong Kong's cumulative share unlock amount is 1,053.3 billion HKD.
Highlights of Major Research Reports
Overseas Macro Strategy: The Silver Lining of Dark Clouds: Tech Narrative Reconstruction, Value Returns Prevail — Yidong Zhang
Looking ahead to 2027, US economic growth may slow, and AI will remain an important growth engine for the US, but its boost to investment and consumption will weaken. After the midterm elections, the US political landscape will undergo major changes, and data center construction and AI capital expenditure growth will both face slowdowns. The AI infrastructure-led AI tech wave 1.0 may have passed its peak. In 2027, the AI tech wave will begin transitioning toward AI 2.0 led by AI applications. AI applications are expected to accelerate, but the market will pay more attention to whether commercialization returns can support large-scale capital expenditure and will focus more on secondary market shareholder returns. In 2026, US equity financing expanded rapidly, ending the 32-year trend of negative net stock issuance by US full-caliber non-financial corporations; US stock shareholder return yields declined significantly, at a near 10-year low, with tech giants shifting from continuously rewarding shareholders to net financing. In 2027, US political changes may lead to cooling Middle East tensions and falling oil prices. Combined with debt ceiling gaming, long-end US Treasury yields may begin a moderate decline after high-level turbulence in the first half. Additionally, comparing US macro factors with those of the Eurozone and Japan, the dollar is expected to oscillate downward in 2027. According to historical patterns, under the triple-weak scenario of a weakening US economy, weakening US Treasury yields, and a weakening dollar, globally interest-rate and exchange-rate sensitive elastic assets all benefit relatively, with low-valuation, fundamentally stable Asian emerging markets often outperforming developed markets.
Overseas Pharmaceuticals: Spring Blossoms, Autumn Harvest, Moving Toward the New — Hong Kong Pharmaceuticals 2026 Winter Strategy — Kehan Meng
In 2027, multiple positive catalysts for Hong Kong pharmaceutical companies—including profitability, BD overseas expansion, and AI technology upgrades—are expected to further materialize. It is recommended to focus on four main lines: innovative drug overseas expansion, CXO, AI drug discovery, and leading consumer healthcare or internet healthcare companies with stable performance. ① Innovative drugs: overseas expansion entering the realization phase; focus on leading Biotechs with key overseas data readouts and Pharma leaders with continuously rising innovative drug revenue share. ② CXO: global demand continues to prosper, orders continuously translate into performance; domestic demand sentiment is gradually recovering. ③ AI drug discovery: short-term wet lab validation demand materializes first; long-term model progress is expected to improve new drug R&D efficiency, reducing costs and increasing efficiency in early research and clinical stages. ④ Consumer healthcare and internet healthcare: domestic demand is gradually stabilizing, and valuations offer good value.
Overseas Energy Technology: Energy Industry Strategy Report: How Energy Technology Continues to Support AI Development Under New Circumstances — Baiqiao Xu
AI infrastructure: modularization is the guarantee under new circumstances. North American AI power consumption may accelerate its contraction toward "clean + stable" electricity. Energy storage is the "reservoir" for new energy and AI. Oil, gas, and chemicals: high cash returns resonate with spread repair.
Daily Chart Series
Risk warning: Intensifying Middle East geopolitical risks, US Treasury yields rising beyond expectations.
Analysts of this report: Zhang Yidong, SFC HK License No.: BIS749; Zhou Linhong, SFC HK License No.: BEO418; Huang Yuxin, SFC HK License No.: BRD659
Comments