Emerging markets investors are once again confronting a familiar challenge: market concentration. But while the phenomenon itself is not new, the current iteration is both more extreme and structurally different from previous cycles.
This has important implications for portfolio construction, diversification, performance and risk oversight, particularly for large institutional investors such as superannuation funds.
At the centre of this shift is the extraordinary rise in the dominance of a small group of mega-cap stocks. The top five names in the MSCI Emerging Markets Index (MSCI EM): Taiwan Semiconductor Manufacturing Company (TSMC), SK Hynix, Samsung Electronics, Alibaba and Tencent now account for almost 30% of the benchmark, up sharply from just 18% at the end of 2023. Within that cohort, TSMC alone represents close to 15% of the entire index, a significant jump from 6% at the end of 2023.
This is not just a story about market capitalisation concentration. It is a story about thematic concentration – specifically the growing dominance of artificial intelligence (AI), semiconductor supply chains, and data centre infrastructure within emerging markets.
This raises a critical question for investors seeking broad exposure to emerging markets growth through allocations in the benchmark: are portfolios delivering the diversification they are supposed to, or are they increasingly reflecting a narrow, highly correlated set of drivers?
AI-driven concentration
AI and data centre investment is the distinguishing driver of emerging markets concentration today.
TSMC, SK Hynix and Samsung Electronics are central to this theme. TSMC, as the world’s leading semiconductor manufacturer, plays a critical role in manufacturing advanced chips used by companies such as Apple, Nvidia and AMD. SK Hynix and Samsung Electronics dominate the production of memory chips such as NAND, DRAM and HBM that are essential to AI computing infrastructure.
Together, this cohort is part of a tightly interconnected ecosystem, and their fortunes are linked to the trajectory of AI adoption and data centre expansion.
Their dominance within the MSCI EM index goes beyond individual stock weights. It introduces a concentration that is exposed to the same underlying economic drivers – in this case, the AI investment cycle.
For many investors, this means that what appears to be diversification across multiple stocks within emerging markets may represent a single, highly concentrated exposure.
Exposure to a narrow opportunity set
The MSCI EM Index contains roughly 1,200 stocks, yet its top five names dominate index performance to an extent that is disproportionate even by historical standards.
This dominance has been driven by a combination of strong price performance and significant upgrades to earnings expectations, particularly for semiconductor names such as SK Hynix and Samsung Electronics. Since late 2025, both companies have benefited from surging demand for memory chips used in AI-focused data centres, delivering exceptional returns – SK Hynix is up 610% and Samsung Electronics is up 315%. Upgrades to their earnings forecasts have risen even faster than share prices.
This has resulted in an unusual dynamic. Despite strong returns, forward price-to-earnings (PE) multiples for these companies have declined, making these stocks look cheap on a forward-looking basis.
But this apparent value may be misleading.
Semiconductor businesses, especially those that produce memory chips, are cyclical. Their rapid growth reflects an early stage in the economic cycle for AI, which can make a low forward PE misleading as an indicator of value. Earnings are heavily influenced by demand cycles, capacity constraints, and pricing power. During periods of strong demand, profits can surge and make valuations look cheap, but these conditions are not sustainable.
Markets account for this cyclical nature with stock prices generally reflecting what a company is expected to earn over time, rather than on short-term forecasts. As a result, a low forward PE may indicate where earnings are in the cycle rather than true value.
An illusion of diversification
This has important implications for portfolio construction, particularly for benchmark-aware strategies or those closely tracking indices.
As concentration increases, exposure to this small group of dominant stocks and AI-linked themes also rises, meaning performance is increasingly driven by a narrower set of names.
The result can be a disconnect between portfolio intent and outcome. Investors may believe they hold a diversified emerging markets portfolio, but a large proportion of risk and return is linked to a handful of semiconductor and technology companies. This can lead to unintended exposures, including greater sensitivity to AI investment cycles, semiconductor demand, and geopolitical risks in markets such as Taiwan and Korea.
Implications for super funds
For superannuation funds, the recurrence of emerging markets concentration underscores the need for careful consideration of benchmark construction, diversification, and risk oversight.
Holding a large volume of stocks or investing in a broad index is no longer sufficient to ensure true diversification. Understanding the underlying drivers of returns is increasingly important.
Investors should also consider how benchmark compositions evolve over time. Market capitalisation-weighted indices are efficient and widely used, but they are inherently backward-looking, allocating more weight to companies that have already performed well. In periods of strong thematic growth, such as the current AI cycle, this can lead to increasing concentration and momentum-driven exposures.
Valuation analysis must be contextualised. Metrics such as forward PE ratios can be useful, but they should be interpreted with consideration of cyclical dynamics and sector-specific characteristics. In the case of semiconductors, low forward multiples may not necessarily indicate genuine value.
Finally, there is a broader question about how to balance participation in key growth themes with the need to manage concentration risk. Avoiding dominant sectors such as AI may not be desirable or practical, but neither is accepting benchmark-level exposures without question.
A different approach to emerging market concentration
Low tracking error quantitative investments have a much better chance of outperformance than bottom-up stock picking as the aim is to target maximising active return for active risk taken, with only small tilts relative to the cap-weighted benchmark that matter.
At RQI Investors, our positions in TSMC, SK Hynix and Samsung Electronics reflect our active, systematic positioning relative to the index. We remain underweight in TSMC and SK Hynix, whereas we have a core weight in Samsung Electronics.
This highlights an important point – we do not simply avoid holding the largest market capitalisation stocks to reduce concentration. Instead, our approach is focused on the long-term economic size of companies.
This distinction is important. Market capitalisation can fluctuate significantly based on short-term sentiment, momentum, or thematic trends. Economic size, by contrast, is a more stable measure of a company’s underlying business footprint.
By anchoring portfolio construction to economic size, we aim to achieve a more balanced and diversified exposure – one that is less influenced by transient market dynamics and concentrated themes.
This highlights an important point: reducing concentration risk is not about excluding large companies, but about how they are weighted within a portfolio.
Looking through the market noise
The re-emergence of extreme concentration in emerging markets is a reminder that diversification requires continuous assessment in the context of evolving market structures and themes.
While the AI-driven ecosystem has boosted the rapid growth of these businesses, their dominance also introduces new risks, particularly for investors relying on traditional index-based approaches. The challenge is to navigate the landscape thoughtfully, to participate in this structural growth opportunity while maintaining genuine diversification and robust risk management.
The ability to look through the market noise and focus on fundamental drivers of value has arguably never been more critical.
David Walsh is head of investments at RQI Investors.
