East Capital Global Emerging Markets ex China

Q2 was another very strong period for emerging markets outside of China, with the benchmark returning 33.7%. East Capital Global Emerging Markets ex China generated 10.4% of alpha, returning 44.1%. For the first half of 2026, our fund has returned 49.4%, outperforming the benchmark by 12.1%. South Korea and Taiwan were the main contributors in both absolute and relative terms, with the markets up 88% and 49%, respectively during the quarter.

The quarter marked something of a new era for emerging markets. Strong performance meant that Taiwan and Korea overtook China to become the two largest markets in the main MSCI Emerging Markets Index. As our ex-China fund uses a limited index of 10-40 names, the weight of Taiwan in particular is constrained, as TSMC’s weight is capped at 10%. Consequently, Korea and Taiwan now represent 55% of this fund’s benchmark. Such strong movements would have been difficult to imagine at the start of the year, which reinforces our stance to maintain a country-neutral approach since the fund’s inception. We do not believe that making macro calls on relative country winners is a reliable way to generate alpha. Instead, we focus on constructing portfolios within each major market that are aligned with the benchmark’s country weights.

As a result, we did not spend time this year debating the appropriate overall allocation to Korea. Instead, we noted the Korea weight in the benchmark, replicated this and focused on building a portfolio designed to outperform the Korean part of the index. This country-neutral approach is often viewed by investors as a risk management tool, and rightly so. However, it has also driven upside, both relative to peers who were underweight in these markets and relative to the benchmark, as our stock selection outperformed and we avoided diluting alpha by reducing overall exposure.

The main driver of returns has been the significant upward revision in earnings expectations in Korea and Taiwan. Markets are increasingly reflecting successive price increases in the underlying hardware supporting the build-out of AI data centres, particularly memory chips, which we have written about in detail in this article. Consequently, despite a year to date return of 119%, Korea has become cheaper, as have emerging markets as a whole, with India having derated.

An interesting observation in Korea is that foreigners have been net sellers, trimming around USD 75 billion of equities year to date as of early June. In contrast, retail investors have been pouring in money, particularly through leveraged ETFs. Even in Hong Kong, the fastest growing ETF of all time is a double-leveraged Hynix ETF, which now represents 13% of the Hong Kong ETF market.  This is clearly not ideal, as the Korean regulator belatedly acknowledged. As such, we expect volatility to remain high and have therefore reduced our positions in these stocks. However, we remain slightly overweight as we believe that the earnings upgrades will continue as the memory tightness persists into 2027 and valuations remain reasonable. Hynix and Samsung together will generate almost USD 1 trillion of free cash flow over the next three years, with a current combined market cap of around USD 2.6 trillion, a stark contrast to the US hyperscalers, where free cash flow is falling steeply as they invest in data centres. 

Looking ahead, our thesis on AI is that we have reached the bottom of a steep S-curve in terms of token consumption, as agentic AI accelerates throughout enterprises worldwide. We are experiencing this first hand, as we regularly max out our Claude tokens when using multi-chained agents rather than simple queries. This will mean that the AI build-out will have to continue and the hyperscalers will have no option but to invest heavily in data centres to meet this insatiable demand for tokens. According to some sources, approximately 30% of their capex will go on memory in 2026 , compared to just 8% in 2024, with more flowing to Taiwan for chip production and components. This will lead to continually rising prices. One industry expert sees memory prices rising by 40-50% quarter on quarter in 3Q 2026, whereas market consensus is closer to 15-20%. Capacity will remain constrained for memory and most other components well into 2028. 

Given the exceptionally strong market environment and our focus on identifying growing but underappreciated companies, it is unsurprising that the majority of our alpha was generated in Taiwan and Korea. Most of this alpha came from smaller companies operating in key bottlenecks in the AI value-chain (five were added this year), with the largest contributor being Samsung Electro-Mechanics (SEMCO).

SEMCO has traditionally been viewed as a relatively mature passive component manufacturer. However, it produces two critical components required for AI data centre build-out: multilayer ceramic capacitors (MLCCs), which regulate and stabilise the power supply to chips, and FC-BGA package substrates, the high-density boards that connect chips to circuit boards.

Demand for MLCCs is being driven by the rapid expansion of AI infrastructure. While a standard server typically uses around 2,000 MLCCs, a single NVIDIA GB200 or GB300 NVL72 rack requires approximately 440,000 units. Importantly, only two companies are currently capable of manufacturing these high-end, AI-server-grade MLCCs at scale: Murata in Japan and SEMCO.

In 1H2026, the company delivered a return of 646% to the fund, as investors bought into our thesis that supply and demand dynamics for these components were incredibly tight, giving the company considerable pricing power. Towards the end of the quarter, we took profits as we felt that valuations looked quite stretched, even when taking into account our fairly bullish earnings expectations. At its peak, SEMCO represented just over 4% of the portfolio, but currently accounts for only around 0.8%.

Aside from SEMCO, the main trades were driven by risk management. We took profit in positions that exceeded our active weight limit of 4%, and actively managed factor exposures. For example, we began the year with an overweight position of 4-5% in Korean memory names, Samsung Electronics and SK Hynix, which we subsequently reduced to around 1%.

Amid the AI hype, we dug deeper into alternative opportunities in Korea; K-Beauty is one area that caught our attention. We travelled to Korea to meet with a range of companies in the sector. We initiated positions in names such as APR, which delivered 174% year-on-year growth in Q1, supported by strong sales growth in both Europe and the US.  We also added Coway, a water purification company and stable compounder, which is trading at single-digit P/E multiples. 

Looking ahead, we believe that the backdrop for emerging markets remains positive. Micron’s results and guidance, published just before the end of the quarter, indicate that the market continues to underestimate the substantial pricing power of component and semiconductor manufacturers. At the same time, we remain disciplined in our positioning and are not taking material risks relative to the benchmark, either at the sector or country level.

Valuations are even more attractive than at the start of the year, particularly in markets such as India. While the broader narrative in these countries is less compelling than in Korea and Taiwan, there are numerous individual investment opportunities where valuations have corrected despite strong fundamentals. A similar dynamic is evident in Brazil, where we recently visited and found that local investor sentiment appears overly pessimistic. Valuations remain attractive for the high-quality compounders that we believe can grow regardless of the political backdrop.

Another important factor that we believe is being overlooked is the current low level of oil prices. The International Energy Agency forecasts a surplus of 5 million barrels per day by 2027, which could place further downward pressure on prices. This would benefit most of our markets, particularly India and the ASEAN region. Indeed, having spent the last few weeks meeting Indian CEOs in London, we have concluded that the fundamental situation on the ground in India is not nearly as bad as the common market narrative and weak market performance would suggest.

Based on current consensus estimates, the fund trades at a 2027 P/E of 9.7x, with expected earnings growth of 28%. By comparison, the benchmark trades at a 2027 P/E of 9.4x, with expected earnings growth of 26%.

Performance in USD net of fees.

The information in this document should not be considered investment advice and should not be used as the sole basis for an investment decision. Please read the Prospectus and the KID, which are available on the fund page. This publication is not directed at you if we are prohibited by any law in any jurisdiction from making this information available to you and is not intended for any use that would be contrary to local laws or regulations. Every effort has been made to ensure the accuracy of the information, but it may be based on unaudited or unverified figures or sources.

East Capital Global Emerging Markets ex China is classified as an Article 9 fund under SFDR. ESG factors are fully integrated throughout the investment process, with the team using proprietary tools to identify and assess sustainability-related risks and opportunities. This enhances the understanding of long-term value drivers and strengthens the quality of the investment case for each company.

You can find key documents explaining the sustainability process in the "Sustainability" section above.