Since 2026 (as of March 13, the same below), the A-share market has presented a sharply differentiated picture against the backdrop of the &#34;14th Five-Year Plan&#34; kickoff and geopolitical conflicts.
At the index level, the small and mid-cap style and dividend strategy have performed particularly well, with the Wind Micro-Cap Index leading with a 14.47% increase, the Dividend Index rising by 10.58%, the CSI 500 up by 10.37%, and the CSI 1000 up by 8.15%, significantly outperforming large-cap blue chips. During the same period, the Wind All A Index rose by 5.31%, the Shenzhen Component Index increased by 5.59%, the SSE Index went up by 3.19%, while the CSI 300 only slightly increased by 0.85%, and the Hang Seng Tech Index fell by 9.75%.
The core logic driving the market can be clearly seen from the industry gainers and losers. The leaders can be roughly categorized into three major camps:
First, the resource cycle sector represented by coal (25.96%), oil and petrochemicals (23.17%), and non-ferrous metals (14.00%), with non-ferrous metals seeing a monthly increase of over 22% in January, becoming the leading force at the start of the year, while the coal sector has recently accelerated upward, topping the gainers list;
Second, the infrastructure industry chain represented by building materials (15.74%), construction decoration (13.83%), and steel (11.72%), benefiting from the expectations of &#34;14th Five-Year Plan&#34; infrastructure investment and the promotion of new urbanization;
Third, the high-end manufacturing sector represented by electrical equipment (15.18%) and machinery (8.13%), driven by both commercial aerospace policy support and industrial upgrades.
In contrast, the financial sector and consumer sector have collectively come under pressure, with commercial retail leading the decline at 8.40%, non-bank financials down 8.10%, banks down 4.33%, and food and beverage down 1.99%, forming a stark contrast to the strength of cyclical stocks.
Since 2026, the gap between the highest-yielding coal industry and the lowest-yielding commercial retail industry has widened to 34.36 percentage points, indicating a clear degree of market differentiation.
<img src="https://imageproxy.pbkrs.com/https://inews.gtimg.com/om_bt/OKQxuCVCQYRS6zQZff3AQCc8kMkNQ-ySH1cVqlUjyv2XIAA/641?x-oss-process=image/auto-orient,1/interlace,1/resize,w_1440,h_1440/quality,q_95/format,jpg" alt="Image" original-src="https://imageproxy.pbkrs.com/https://inews.gtimg.com/om_bt/OKQxuCVCQYRS6zQZff3AQCc8kMkNQ-ySH1cVqlUjyv2XIAA/641"/>
Behind this extreme differentiation is a concentrated reflection of the mainstream trading logic in the market.
Firstly, the kickoff effect of the &#34;14th Five-Year Plan&#34; is fermenting, with a significant characteristic of policy-driven dynamics. The National Two Sessions clarified the focus direction of new productive forces, including humanoid robots, 6G, and commercial aerospace, with fiscal efforts supporting equipment upgrades and safety capability construction, providing continuous catalysis for related sectors.
Secondly, the long-term nature of geopolitical factors is profoundly reconstructing the value logic of resource products. The escalation of geopolitical conflicts in the Middle East has directly pushed up prices of crude oil, gold, and others, with risk aversion sentiment and supply contraction expectations resonating, providing solid support for the resource sector
More importantly, the ongoing escalation of conflicts is accelerating the global energy substitution process. The rise of the power equipment sector (15.18%) is not merely a reflection of the traditional energy cycle but is deeply benefiting from the energy security and substitution logic catalyzed by geopolitical conflicts—demand for new energy infrastructure such as wind power, photovoltaics, and energy storage is being systematically boosted, becoming a key track that combines safety barriers and growth attributes.
Third, the explosion of AI agents is reshaping the demand pattern for computing power and industry narratives. The open-source agent framework represented by OpenClaw has achieved a closed loop of &#34;cognition, execution, and memory,&#34; capable of autonomously completing complex tasks such as cross-application coordination, marking the transition of AI from the &#34;dialogue&#34; era to the &#34;execution&#34; era. The most direct impact of this transformation is the exponential increase in demand for computing power—Token consumption has upgraded from &#34;human-machine dialogue&#34; to &#34;machine self-circulation,&#34; with a single complex task potentially consuming hundreds to thousands of times more Tokens than traditional models.
Industry estimates show that the daily Token consumption of mainstream large models in China has surged from the 100 billion level at the beginning of 2024 to 180 trillion by February 2026. In the face of the sudden increase in reasoning computing power demand, mainstream manufacturers have successively experienced tightness in computing power, with unprecedented supply-demand mismatches.
This demand for computing power generated by the explosion of agents is resonating deeply with the quietly emerging &#34;HALO trading&#34; on Wall Street.
HALO stands for &#34;heavy assets, low elimination rate,&#34; and its core logic is: go long on &#34;heavy assets that are difficult for AI to replace and rely on,&#34; while going short on &#34;light assets that are easily disrupted by AI.&#34; Data shows that over the past year, the U.S. infrastructure stock index has risen by more than 70%, while the SaaS index has significantly declined.
The rise of HALO trading is essentially a market re-examination of the &#34;moat&#34; of the light asset model. When AI agents can automatically generate enterprise-level processes, the traditional SaaS profit model based on seat subscriptions faces fundamental questioning. Investors are turning to pursue tangible certainty—entities such as power grids, pipelines, and minerals that cannot be replicated by code are becoming a safe haven in an era of uncertainty.
It is worth emphasizing that the HALO strategy is not opposed to AI; rather, it benefits from the development of AI. Training trillion-parameter large models requires a large number of transformers, liquid cooling equipment, and thick copper cables. The extremely high physical reset costs form the cornerstone of the revaluation of heavy asset valuations. As millions of &#34;digital employees&#34; continuously operate in the cloud, their consumption of electricity and reliance on computing power clusters are continuously directing demand towards those fields of &#34;heavy assets, low elimination rate.&#34;
Fourth, the long-term mainline of industrial upgrading and security autonomy is still deepening. The power equipment sector benefits from energy transformation, while sectors such as basic chemicals are steadily rising under the dual drivers of price increase expectations and domestic substitution. Breakthroughs in AI computing hardware and commercial aerospace further enhance the attractiveness of the technology growth sector
In contrast, the decline of the financial and consumer sectors exposes the market's cautious attitude towards the endogenous growth momentum of the economy. The drop in banks and non-bank financial institutions reflects multiple concerns about narrowing interest margins, the transmission of real estate risks, and the weak recovery of brokerage valuations. The weakness in consumer sectors such as food and beverages and retail points to the slow recovery process of consumption and the ongoing impact of online diversion.
Market funds are choosing to embrace policies and resource logic that are more directly influenced by macro factors, while temporarily avoiding areas deeply tied to residents' consumption willingness and the real estate cycle. This itself is a cautious expression of the current economic recovery structure.
Looking ahead, given that the market logic has undergone significant changes since the beginning of the year, investment strategies should also be adjusted accordingly.
In the short term, market style may still lean towards small and mid-cap value, and the performance of resource cycles and infrastructure industry chains may still have room for development during the policy implementation and data verification period. However, it is particularly important to be cautious as the conflict between the U.S. and Iran is evolving from short-term pulse disturbances to a long-term pattern. The capacity blockage in the Strait of Hormuz, rather than the reduction of exports from Iran alone, is the core contradiction of this round of supply shocks—this global maritime oil lifeline carries about 14 million barrels of crude oil daily, and the continuous decline in its efficiency means that the supply premium for crude oil will shift from event-driven to systemic cost increases.
This shift will strengthen the price resilience of energy, coal chemical products, pesticides, and polyurethane, while also accelerating the replacement process of new energy for traditional energy. Wind power, photovoltaics, and energy storage will welcome new development windows, putting continuous pressure on cost-sensitive industries such as aviation and shipping. The long-term nature of the conflict will further validate the importance of unmanned equipment, bringing continuous catalysts to the AI and drone sectors in the military industry. Therefore, while the short-term structural market continues, it is necessary to examine the risks and opportunities of conflict diffusion from a layered perspective.
In the medium term, as the global semiconductor cycle recovery trend gradually confirms and the demand for AI hardware continues to grow, the allocation value of the technology growth sector (electronics 5.78%, computers 1.92%) will gradually become apparent. Consider moderately shifting from cyclical resource sectors towards technology and high-end manufacturing.
Looking upstream along the main line of explosive demand for computing power, computing power infrastructure is the first to be impacted—chip and server manufacturers with large-scale reasoning capabilities are undoubtedly the &#34;shovel sellers&#34; of the intelligent economy. Further advancing, industry leaders that can first implement intelligent capabilities, restructure business processes, and generate real commercial returns will gain significant first-mover advantages.
Long-term layouts should closely follow two main lines: One is the technological innovation (AI applications, semiconductors, commercial aerospace), strategic emerging industries, and green transformation (new energy, energy storage, environmental protection) determined by the &#34;14th Five-Year Plan,&#34; and the recovery of domestic demand (consumption upgrades, high-end manufacturing going abroad); the second is the HALO core track—those heavy asset fields that are not easily disrupted by AI but rather benefit from AI development, including power equipment, non-ferrous metals, oil and gas pipelines, and transportation infrastructure. They are both the physical foundation of AI and a &#34;safe haven&#34; for funds to avoid technological risks. Against the backdrop of global supply chain restructuring, physical assets such as mines, power grids, and chemical production capacity are being systematically reassessed for their scarcity premium due to their high capital barriers and long cycle attributes
In the current market environment, market volatility has increased, objectively making it more difficult to obtain returns. What investors need to do is to maintain a clear understanding of the alignment between valuation and fundamentals amidst the noise of leading sectors, and to identify opportunities that have been mispriced in neglected sectors. Balanced allocation, avoiding chasing highs, and seeking entry points that align with long-term themes amidst volatility are essential for navigating this round of structural market conditions steadily and successfully.
[Note: The market has risks, and investment requires caution. Under no circumstances does the information or opinions expressed in this subscription account constitute investment advice for anyone. Unless specifically noted, the research data in this article is supported by Tonghuashun iFinD.]
This article is original from the public account &#34;Xingtu Financial Research Institute,&#34; authored by Xue Hongyan, Executive Vice President of Xingtu Financial Research Institute.
Editor: Hu Wei

515080

515890

563300

Since 2026, the A-share market has shown significant differentiation against the backdrop of geopolitical conflicts and the start of the "14th Five-Year Plan." The small and mid-cap style and dividend strategy have stood out, with the Wind Micro-Cap Stock Index rising by 14.47% and the Dividend Index increasing by 10.58%. The resource cycle, infrastructure industry chain, and high-end manufacturing sectors have led the gains, while the financial and consumer sectors have faced pressure, with the commercial retail sector declining by 8.40%. The mainstream trading logic in the market is concentrated on policy-driven and emerging industry support

Attention! The investment logic of A shares has undergone significant changes