Shenwan Hongyuan Strategy: Also discussing "HALO Trading"

1. Let’s also discuss “HALO Trading”: the market is beginning to focus on the evolution of industry organization forms in the AI era. Industries that may be replaced by AI + industries where barriers weaken and excess profits may be compressed in the AI era + leading tech companies that may no longer sustain their competitive advantage are being re-evaluated, with long-term expectations resetting and valuation centers facing downward pressure. Overall, Chinese assets are less affected than U.S. stocks, as China has fewer industries with monopoly barriers and excess profits.

Currently, thinking about the AI ultimate scenario is difficult to cover all key factors. The market tends to project technological industry trends to their final capabilities (AI’s ultimate ability), but it often overlooks the inevitable gradual changes in other key factors (including productivity, production relations, political systems) during AI’s progress. Therefore, the market’s concentrated pricing of the potential impact of AI’s final scenario may lead to misjudgments.

Strategic resources that are hard to replace are assets of the era: security and controllability of strategic resources + inflation in the AI era.

Globally, everyone seems to be playing the “HALO Trading” game, with markets beginning to project possible changes in industry organization forms in the AI era. Long-term expectations for three types of industries are being reset: 1. Industries potentially replaced by AI. In the short term, breakthroughs by AI model companies can cause significant adjustments in related stocks. 2. Industries where barriers weaken and excess profits may be squeezed due to AI. AI could significantly reduce supply costs and entry barriers in some sectors. Even established leading companies that survive in the AI era will need to leverage AI to offer higher-quality products amid falling prices and face fiercer competition. 3. For leading tech companies with AI layouts, the question is whether they can continue to succeed in the AI era. Among these, the second and third types of industry leaders, whose valuation centers are anchored on monopoly profits and platform value, may be downgraded to companies that participate in market competition for reasonable profits. Long-term expectations are being reset, leading to a downward shift in valuation centers.

The market’s projections of AI’s final scenario are complex and varied. We do not discuss the correctness of specific projections here. But what we can say is that, at this stage, thinking about AI’s ultimate scenario is difficult to encompass all critical factors. The market tends to project technological industry trends (AI’s final capabilities) to their end point, combined with current economic, social, and industrial organizational analyses. In reality, AI’s progress will be gradual, with incremental “partial disruptions” accumulating into “comprehensive disruptions.” During this process, productivity, production relations, political systems, industry organization forms, and social organization forms will all undergo transformation. When AI weakens human labor demand, it also means profound changes in individual capabilities and the boundaries of human society. As AI approaches its ultimate ability, the boundaries of human society will expand significantly. Analyzing AI’s final scenario within existing social organizational frameworks may be biased. On a micro level, impacted industries and companies will inevitably respond, adjust, and transform.

Concerns about 2028, even if based on principles (which we doubt), are not so urgent in timing and can be addressed through adjustments. Such concerns may themselves be opportunities to eliminate certain risks.

Regarding strategic assets that are hard to replace, we believe the logic of era assets is further reinforced: in the context of great power competition, independence and controllability of strategic resources remain core investment clues. The “inflation” logic of strategic resources and energy in the AI era may be further strengthened.

2. Short-term market characteristics: after the Spring Festival, A-shares show weak responses to long-term tech narratives but are very positive about the visible “new and old economic inflation.” This reflects the “HALO Trading” mapping in A-shares and is also related to the Fed’s easing expectations causing disturbances, leading to a relatively full interpretation of tech narratives in A-shares.

Post-Spring Festival, the short-term market features in A-shares have changed: weak response to long-term tech narratives, but strong response to visible “new and old economic inflation.” Value-oriented styles are generally favored globally. This is related to the “HALO Trading” in A-shares. Additionally, rising inflation expectations and disturbances in Fed easing expectations have limited the reflection of optimistic long-term tech outlooks. Meanwhile, the tech narrative in A-shares has played out relatively fully, with valuations reaching historical highs despite incomplete economic and earnings realization, indicating a more “pragmatic” phase for the tech sector. The pricing of China’s robotics manufacturing and motion control sectors has been established, and the ongoing progress in AI applications has also been reflected. After a spring rally driven by catalysts, these sectors have weakened. This market experiment suggests that the market needs to see “cross-stage development” in industries before initiating a new rally, aligning with our mid-term “two-stage rally” projection.

We maintain our mid-term “two-stage rally” outlook: the spring 2026 rally is an extension of the 2025 structural rally, currently still in the high zone of the first stage. At this stage, the overall PE valuation of A-shares remains at historically high levels, with internal demand for correction and consolidation. This correction phase mainly awaits further reinforcement of industry trends, validation of the economic turning point (performance digesting valuations), easing of valuation concerns, and more favorable conditions for residents’ asset allocation into equities.

There is also a “second stage” of rally in the mid-term: cyclical improvement in fundamentals + new phase of technological industry trends + more favorable conditions for residents’ asset reallocation into equities + resonance of China’s increasing influence expectations. The window for the second stage to start is likely around mid-2026. We reiterate that the two-stage rally involves consistent leading sectors and styles. In this era, “inflation assets” are primarily technology + strategic resources.

3. The most visible short-term inflation directions are the main sources of favorable structural opportunities: short-term surge in cyclical commodities (steel, coal) driven by limited supply, but demand verification in March-April raises doubts about sustained price increases. Cyclical allocation still recommends focusing on strategic asset inflation (non-ferrous metals, basic chemicals, oil, shipping). The mapping of new economy inflation to traditional sectors remains the strongest short-term direction, with investment opportunities in internal combustion engines, fiberglass, optical fibers, and storage.

The mid-term structural recommendation remains: Prosperous technology + cyclical alpha. Prosperous tech focus: overseas computing power chains, AI applications (genuine opportunities in Hong Kong internet), semiconductors, robotics, commercial aerospace, energy storage. Cyclical alpha focus: non-ferrous metals and basic chemicals. The extension of mid-term cyclical alpha investment may include export/outbound chains. Additionally, we are optimistic about revaluation opportunities in non-bank financials in the mid-term.

Risk warning: Unexpected overseas economic recession, domestic economic recovery below expectations.

(Source: Shenwan Hongyuan)