Wall Street Soars, KOSPI Stays Trapped... U.S. Has Big Tech and Chip Design, Korea Only Memory in English - The Asia Business Daily

The U.S. stock market has been on an upward trajectory, even amidst high interest rates, while South Korea's KOSPI index has remained stagnant. Analysts attribute this disparity to the significant differences in the structure of the artificial intelligence (AI) value chain between the two countries, particularly as the focus shifts from infrastructure investments to monetization strategies.

On October 12, the KOSPI and KOSDAQ indexes were visible on an electronic display at Hana Bank's headquarters in Jung-gu, Seoul. According to a report from iM Securities, the United States boasts leading publicly traded firms across five key segments of the AI value chain: applications (such as Palantir and ServiceNow), cybersecurity (including CrowdStrike and Palo Alto Networks), cloud platforms (like M7), semiconductor design (featuring NVIDIA, Broadcom, and Marvell), and electronic design automation (EDA) for semiconductors (with major players like Synopsys and Cadence). In contrast, South Korea's industry is predominantly focused on manufacturing, particularly in foundries, memory production, and related equipment.

U.S. companies operating in these advanced segments have less exposure to inventory and capital equipment costs, benefiting from high customer retention rates and subscription-based revenue models. Consequently, they are less impacted by fluctuations in memory prices. For instance, Marvell, a fabless semiconductor design firm, relies on data centers for 79% of its revenue by outsourcing its production. In the EDA market, Synopsys, Cadence, and Siemens collectively dominate, holding between 70% and 80% of the global market share, with subscription-based sales contributing to 80% to 85% of their revenues and having nearly 100% customer retention rates.

In contrast, South Korea’s semiconductor sector is heavily reliant on memory-focused integrated device manufacturers (IDMs), which directly bear the costs associated with capital investments and inventory. This dependency means that market oversupply can lead to reduced average selling prices and inventory write-downs, adversely affecting earnings. Even during an AI-driven market boom, companies situated lower in the value chain struggle to maintain margins, while those positioned higher can capitalize on growth opportunities.

The KOSPI's significant reliance on semiconductors presents another challenge. Samsung Electronics and SK hynix, two major players in the semiconductor industry, account for over half of the total market capitalization of the KOSPI, whereas semiconductors constitute only around 18% of the S&P 500 index. This disparity means that South Korea's stock market often lacks sectors that can benefit from capital rotation witnessed in the U.S. market. While U.S. capital can shift from beneficiaries of infrastructure investments to AI applications and cloud services, South Korea lacks comparable upper-tier sectors, leading to capital outflows from its stock market instead of rotation within it.

Foreign investment trends suggest that this capital reallocation may already be taking place, as the ongoing absence of foreign investors indicates a withdrawal from South Korean stocks in favor of other sectors. As foreign investors reduce their exposure to AI infrastructure, they are simultaneously decreasing their investment in the South Korean market.

Another concern is the slowdown in investment growth. The share prices of infrastructure companies are sensitive to the growth rate of capital expenditures rather than the absolute dollar amounts spent. Even if major tech firms continue to invest, a deceleration in growth rates could signal a peak for companies lower down the value chain. Predictions suggest that capital expenditure growth among hyperscalers may peak in the third quarter of 2026.

Ultimately, South Korea's role in the AI cycle is that of an intermediate-goods supplier, making it increasingly susceptible as the AI market transitions to monetization and application phases. While some sectors, such as power equipment and substrates, are seeing benefits from investments in AI infrastructure, their market share is minimal compared to Samsung Electronics and SK hynix, which raises concerns about the wider market's ability to keep pace.

Analysts emphasize that the sustainability of earnings among South Korean semiconductor firms will be crucial for a potential rebound in the stock market as the focus shifts toward AI monetization in a high-interest-rate environment. Kim Junyoung, an analyst at iM Securities, noted, "As AI transitions to application phases in the first half of 2027, fears of the South Korean stock market lagging behind could peak, especially since the country lacks exposure to sectors like software and cybersecurity that are likely to benefit." He also mentioned that if a surge in AI application consumption leads to another shortage of inference capacity, it could create a positive cycle benefiting memory and AI infrastructure in the latter half of the year. A strategy focusing on reducing memory concentration and emphasizing high-bandwidth memory (HBM) and power infrastructure could be effective in preparation for anticipated challenges in the first half of next year, with an eye toward a potential rebound later on.