The impact of AI is reshaping the valuation logic of global stock markets. In light of the structural peculiarities of the Asia-Pacific market, Bank of America Securities believes that the crowded &#34;HALO&#34; (heavy assets, low elimination rate) trades have obvious limitations, and identifying with the REAL (regulatory barriers, durable cycles, heavy assets, local services) framework may be more appropriate—especially in markets with abundant capacity and a lack of scarcity moats in heavy industries, where the protective effect of the HALO strategy is at risk of being overestimated.
According to news from the Wind Trading Desk, Bank of America analyst Winnie Wu's team stated in a report on the 16th that the U.S. software sector has seen its market value evaporate by over $2 trillion in the past five months, the Indian IT sector has cumulatively dropped over 40% from its peak in December 2024, and stocks in the Asia-Pacific media, e-commerce, and fintech sectors have also faced massive sell-offs. Meanwhile, funds have rapidly rotated into heavy asset sectors such as semiconductors, capital goods, energy, and utilities, leading to a surge in popularity for &#34;HALO&#34; trades.
However, Bank of America warns that the HALO strategy has limitations in markets with abundant industrial capacity. In heavy industries lacking regulation or scarcity moats, heavy assets may turn into liabilities, as AI can shorten R&amp;D cycles and give rise to alternative technological paths, exacerbating overcapacity and competition.
Therefore, the REAL framework is introduced to reassess the survival risks of enterprises. This framework filters from four dimensions: regulatory barriers, durable cycles, heavy asset characteristics, and local service intensity. Bank of America research shows that even in the software and consumer internet sectors most severely impacted by AI, leading companies with REAL characteristics still demonstrate exceptional resilience against declines and long-term investment value.
<img src="https://imageproxy.pbkrs.com/https://wpimg-wscn.awtmt.com/c8a4459a-abb6-47d9-87b7-5030096026b1.png?x-oss-process=image/auto-orient,1/interlace,1/resize,w_1440,h_1440/quality,q_95/format,jpg" alt="" original-src="https://imageproxy.pbkrs.com/https://wpimg-wscn.awtmt.com/c8a4459a-abb6-47d9-87b7-5030096026b1.png"/>
<h2>Limitations of the HALO Strategy: Heavy Assets Are Not a Universal Moat</h2>
The core logic behind the market's pursuit of HALO sectors is that the construction cycle of tangible assets is long and difficult to be quickly replaced by AI. This logic is reasonable in areas with guaranteed asset scarcity, but its limitations are evident in markets where engineering talent and industrial capacity are extremely abundant.
Bank of America points out that in heavy industries (such as automobiles, solar energy, steel, and cement) lacking regulatory constraints or scarcity moats, supply can easily exceed demand, leading to fierce price competition. Heavy assets in these areas not only fail to provide protection but may also become a burden—especially when AI further shortens R&amp;D cycles and opens up alternative technological paths. In contrast, some light asset industries that heavily rely on human services (such as healthcare and dining) may exhibit more enduring resilience against AI replacement.
<img src="https://imageproxy.pbkrs.com/https://wpimg-wscn.awtmt.com/81aa2ce1-5b03-4312-b681-488e752a9455.png?x-oss-process=image/auto-orient,1/interlace,1/resize,w_1440,h_1440/quality,q_95/format,jpg" alt="" original-src="https://imageproxy.pbkrs.com/https://wpimg-wscn.awtmt.com/81aa2ce1-5b03-4312-b681-488e752a9455.png"/>
Bank of America also notes that the REAL framework does not imply that companies in related sectors are immune to AI. AI can expand these companies' addressable markets, reduce operating costs, and may also compress innovation cycles and lower industry barriers. The core judgment is that leading companies in high REAL moat industries face significantly lower survival threats compared to their counterparts in low REAL industries, especially in a rapidly disruptive environment driven by AI
The valuation compression of the low REAL sector may be more persistent and profound.
<h2>REAL's Four Moats: Redefining Defensiveness</h2>
Bank of America defines the four dimensions of the REAL framework as follows:
Regulatory Critical: Systemically important banks, telecommunications operators, electricity and energy suppliers, and defense-related industries. These sectors involve social stability and national security, with regulatory agencies imposing strict regulations on access licenses, foreign ownership ratios, and the operation of critical infrastructure. The introduction of AI often brings higher monitoring requirements and burdens of human oversight, increasing compliance costs rather than lowering barriers.
Enduring Cycles: Semiconductors, capital goods, aerospace and shipbuilding, pharmaceuticals and biotechnology, and gaming IP. The barriers in these industries stem from the irreducible time accumulation in the real world—advanced chip processes rely on multi-generational process knowledge and EUV equipment accumulation, with new wafer fab construction cycles lasting several years; aircraft must undergo complex airworthiness certification; drugs must pass through multiple stages of clinical and regulatory review; gaming copyrights typically last 50 to 70 years, supporting continuous IP monetization. AI can optimize some processes but cannot bypass the multi-year certification and validation requirements mentioned above.
Asset Heavy: Natural resources and commodities, power grids and utilities, railways and ports, and livestock farming. The scarcity in these industries arises from the hard constraints of physical resources—mineral reserves are limited, licensing approval processes are lengthy, infrastructure construction costs are high, and new entrants find it extremely difficult to economically justify replicating existing assets.
Local Services: Hotels, catering and property management, childcare, elderly care, and pet care, as well as on-site deployment and operation and maintenance in the IT services sector. These jobs involve non-standardized on-site environments, requiring a high degree of human adaptability, and have a very low tolerance for errors; current AI and robotic systems are not yet economically viable substitutes.
<h2>REAL Distribution Across Markets: Short-term Advantages and Long-term Concerns in ASEAN</h2>
From the sector composition of the MSCI Asia-Pacific Index, there are significant differences in AI risk exposure across markets. Southeast Asian markets are heavily tilted towards high REAL sectors: Approximately 79% of Singapore's market capitalization, about 87% of Malaysia's, and around 94% of Indonesia's are concentrated in these sectors, with banks being the core weight. This structure provides relative resilience in a market environment dominated by AI narratives, with Thailand's year-to-date increase of 14.6% and Malaysia's rise of 5.1%, both outperforming India (-10.6%).
However, Bank of America also warns that the short-term resilience of Southeast Asian markets carries the risk of transforming into long-term fragility. If AI-driven automation reduces manufacturing costs in developed economies and makes nearshoring more economically attractive, the offshore demand for labor-intensive production will decline. Vietnam, Malaysia, and Thailand have high ratios of foreign direct investment to GDP and export dependence, and without sufficient digital infrastructure and AI talent reserves, these economies face structural challenges such as widening technological gaps and obstacles to global supply chain integration
<h2>The Impact of AI Combined with Population Aging: A Dual-Axis Analysis of Sector Patterns</h2>
Bank of America analyzes the disruptive impact of AI alongside Asia's low birth rates and accelerating population aging, forming a dual-axis matrix to assess the medium- to long-term structural positioning of various sectors.
Healthcare, semiconductors, capital goods, and insurance are in the best position: They all have a relatively high near-term moat against AI disruption, while benefiting from increased automation demand driven by aging and the spending growth of a wealthier elderly consumer group, making them &#34;structural opportunity&#34; sectors with defensive attributes and long-term growth potential.
Real estate, utilities, and banks, while having a high REAL moat that can buffer the short-term impact of AI valuation shocks, all face long-term pressure from aging on housing formation and discretionary consumption. Bank of America judges that these three sectors &#34;can provide near-term downside protection against AI volatility, but have limited upside potential for long-term valuation reassessment.&#34; Consumer durables, media and entertainment, retail distribution, and automotive sectors are under the most significant pressure—low AI moats combined with the drag of aging on discretionary consumption create a dual disadvantage, making them the most concentrated risk sector combination in the current landscape

ASEA

AI impact is disrupting the Asia-Pacific stock market, and Bank of America warns that crowded "HALO" heavy asset trades pose risks in an oversupply market. It promotes the REAL framework—screening opportunities through four dimensions: regulatory barriers, durable cycles, heavy assets, and local services. The Southeast Asian market has strong short-term resilience due to a high proportion of high REAL sectors like banking; meanwhile, industries such as healthcare and semiconductors, which combine AI defensive capabilities, are seen as more attractive

- The impact of AI is reshaping global stock market valuation logic, with Bank of America emphasizing the limitations of the "HALO" strategy in oversaturated markets.  
- Their REAL framework, focusing on regulatory barriers, enduring cycles, asset heaviness, and local services, suggests that industries meeting these criteria show resilience against AI disruptions.  
- However, the analysis warns that regions like Southeast Asia may face long-term vulnerabilities as AI-driven automation affects labor-intensive sectors and economic dependencies without sufficient digital infrastructure or talent.

Is REAL trading more suitable than HALO for the Asia-Pacific stock market?