Shenwan Hongyuan Group: Signals of the late stage of short-term adjustment have emerged

Shenwan Hongyuan Group released a research report stating that the market is currently digesting pessimistic expectations. The adjustment phase objectively exists, and the adjustment process is also a process of digesting pessimistic expectations. In the tail region of the short-term adjustment, policies to stabilize capital market expectations are timely, with the direction of policy efforts leaning more toward growth, which is more conducive to rapid short-term stabilization. If the direction of policy efforts continues to lean toward value and the clearing of trading chips remains constrained, attention should still be paid to repeated fluctuations in the tail of the adjustment. The deduction of long-term, medium-term, and short-term wave positioning is reiterated: In the long term, A-shares are in a consolidation period between two major upward waves. In the medium term, the consolidation wave continues and still leans toward a weak oscillation phase. In the short term, the market has already oversold, and a small-scale rebound can be expected. The main views of Shenwan Hongyuan Group are as follows: I. Pessimistic expectations currently being digested by the market: 1. The long-term outlook for the AI industry does not generally include computing power inflation, and the industry is currently accelerating its correction. Financing for overseas CSP leaders is under pressure, which reflects both credit risk concerns and the impact of liquidity shortages. The pressure naturally transmits to the computing power inflation chain. Concerns about a decline in AI capital expenditure growth in 2027 are more persistent. Breaking through supply bottlenecks in the computing power inflation chain is also an inevitable trend. 2. The overseas asset allocation environment is unfavorable, and macro constraints on industry are increasing. 3. The capital supply and demand environment is unfavorable, with all types of funds reducing positions to some extent in October. The adjustment phase objectively exists, and the adjustment process is also a process of digesting pessimistic expectations. Sorting out the medium-term issues currently facing the market: 1. Disturbance in the computing power inflation industry trend. The long-term outlook for the AI industry does not include general computing power inflation. A series of current industry changes can be viewed as the industry accelerating its self-correction. There are opportunities in the phase dominated by computing power inflation, and there are also opportunities in the phase dominated by computing power deflation (explosive growth of AI applications), while the transitional phase in between may be a headwind period. The problems observed include: (1) CDS for overseas CSP leaders have generally risen, and financing is under pressure, which reflects both fermenting credit risk concerns and the fact that under the Fed's tightening trend, new financing demand easily triggers interest rate shocks. With CSP leaders under pressure, the pressure naturally transmits to the computing power inflation chain. (2) The current expectation for capital expenditure growth is that 2027 will be lower than 2026, and 2028 will continue to decline. Similar expectations also appeared at the end of 2025 when looking ahead to 2026. This time, however, concerns about financing constraints and power constraints are also superimposed, making concerns about declining capital expenditure growth more persistent and potential breakthrough points fewer. (3) There are no strong constraints on capital expenditure expansion in the computing power inflation chain, and the gradual breakthrough of supply bottlenecks is an inevitable trend. Globally, large AI models and the application side still respond positively to favorable news, but it is relatively difficult for the computing power inflation chain to build new consensus. Recently, U.S. stocks have outperformed Asia-Pacific stocks, and within U.S. stocks, large models, CSPs, and software have outperformed hardware. 2. The overseas asset allocation environment is unfavorable. Resource nationalism pushes up inflation, and in a global environment constrained by inflation, geopolitical disturbances are more likely to intensify. This makes high oil prices and high U.S. Treasury yields resilient. At present, divergences in global economic governance approaches are increasing. Raising interest rates to fight inflation, tightening fiscal discipline, addressing structural economic contradictions, and maintaining overall stable economic operation need to be balanced, yet are difficult to achieve simultaneously. The contradiction between the traditional monetary policy framework and asset prices has become very prominent. The damage to asset prices from elevated U.S. Treasury yields is still unfolding in the short term. 3. The capital supply and demand environment is unfavorable. Short-term macro and industry signals are somewhat headwind, and in the adjustment phase, negative capital feedback loops are playing out again. In October, all types of funds reduced positions to some extent (small and medium-sized insurers, annuities, and fixed income plus products defending key break-even levels, active public funds adjusting and reducing positions, and private funds adjusting positions again after the rebound). At this stage, offensive directions are still decreasing, and the pattern of insufficient offensive capital continues. II. The stage of confirming the adjustment phase has now passed, and more discussion should focus on positioning within the adjustment phase. After the adjustment to date, there is already a timetable for long-term cost-effectiveness clearing, and short-term cost-effectiveness improvement has already shown results. Indicators indicating extremely high short-term cost-effectiveness continue to increase. It is not far from the state in which various cost-effectiveness indicators uniformly point to short-term rebound demand. The short-term adjustment is already in its late stage. The stage of confirming the adjustment phase has now passed. Discussing subsequent adjustment space by confirming a major wave adjustment is more akin to circular reasoning and is not an effective method. At this stage, more discussion should focus on positioning within the adjustment phase. The adjustment process must be a process in which subsequent downside space narrows. Shenwan Hongyuan Group notes: 1. There is already a timetable for long-term cost-effectiveness clearing. At present, the 1/PE of electronics and communications has rebounded to near the historical median. With the disclosure of the 2026 third-quarter report, the static valuation of communications will fall below the historical median; at that time, if 2027 growth improves relative to 2026 and there is no clear long-term expectation of decline, directions may diverge and strengthen. With the disclosure of the 2026 annual report, the static valuation of communications will approach historical lows; at that time, if no inevitable negative growth can be seen in the foreseeable future, asset prices should show strong resilience. This may be a clue to a mid-cycle or even long-cycle low region. 2. Short-term cost-effectiveness indicators have shown results. Floating profits on communications and electronics holdings have fallen back to the bottom region of historical major wave adjustment phases, the downward momentum effect has risen to a historical high, and for representative computing power inflation industries such as electronics and communications, the profit-making effect has also contracted to an extremely low historical region. In addition, all types of funds have reduced positions to some extent, which is also a signal of effective short-term cost-effectiveness improvement. The short-term cost-effectiveness indicators that have not yet been effectively cleared are mainly trading crowding measures and relative profit-making effect measures. It is not far from the state in which various cost-effectiveness indicators uniformly point to short-term rebound demand. This indicates the positioning that the short-term adjustment is already in its late stage. In the tail region of the short-term adjustment, policies to stabilize capital market expectations are timely, with the direction of policy efforts leaning more toward growth, which is more conducive to rapid short-term stabilization. If the direction of policy efforts continues to lean toward value and the clearing of trading chips remains constrained, attention should still be paid to repeated fluctuations in the tail of the adjustment. III. The short-term AI chain still faces headwinds, but it is entering a stage of increasing medium-term opportunities, and the effective aesthetic for the AI chain remains unchanged: 1. A small number of directions whose fundamental expectations can exceed those at the end of June. 2. Directions where 2027 prosperity continues to improve relative to 2026 may be those that diverge and strengthen after earnings plus pullback digest valuation. 3. Attach importance to investment opportunities in technology themes and non-technology themes. Non-technology outperformance is extending in time, and absolute returns from high dividends are extending in time. The focus in non-technology directions remains unchanged: CXO, innovative drugs, real estate, coal, shipping, precious metals, industrial metals, and basic chemicals. High dividends and technology tracks exhibit a "seesaw effect," and accordingly attention should be paid to banks and utilities. Before signals of the market returning to strength emerge, high dividends can still be held based on a defensive approach. Risk warning: Overseas economic recession exceeds expectations, and domestic economic recovery falls short of expectations.