Is your emerging-markets fund an accidental bet on AI? Returns for these two ETFs show why it might be.
SK Hynix and other South Korean AI stocks have made some emerging-markets funds much riskier.
The rise of AI turned a dusty debate over South Korea's label into one of the year's biggest drivers of returns.
Suppose you already have substantial exposure to U.S. equities. After another strong year for the American market, you decide to diversify internationally by purchasing an emerging-markets exchange-traded fund. Two obvious choices are the biggest in the category: BlackRock's iShares Core MSCI Emerging Markets ETF IEMG, which trades under the ticker symbol "IEMG," and Vanguard's FTSE Emerging Markets ETF VWO, which trades under "VWO."
At first glance, the funds appear nearly identical. Both hold thousands of companies across emerging markets. Both charge rock-bottom fees. Both manage more than $150 billion. Most investors would expect them to deliver nearly identical long-run returns.
They haven't. Beginning in the second half of 2025, the two funds pulled sharply apart. Measured from a common starting point in January 2025, IEMG outperformed VWO by as much as 25 percentage points at the peak in mid-June 2026.
Although the gap between the two funds narrowed as technology stocks retreated, it still exceeded 15 percentage points in mid-July - a gap that's almost unheard of between two funds that are supposed to do the same job. Over the same period, South Korea's artificial-intelligence champions - SK Hynix (KR:000660) and Samsung Electronics (KR:005930) - experienced a spectacular rally followed by a sharp correction.
The reason had nothing to do with clever stock picking. It came down to a single, seemingly technical decision made years earlier: whether South Korea counts as an "emerging" market or an "advanced" one.
That decision carries a surprising lesson. Index investing is not as passive as it sounds. Every index is built on judgment calls - and those choices can move money.
A classification with consequences
MSCI still labels South Korea 'emerging,' while FTSE Russell calls it 'developed.'
South Korea is one of the most technologically advanced countries on earth, yet where it belongs in global stock indexes has been debated for years. By the usual economic yardsticks - such as income, industry and technology - it looks fully developed, and the Organization for Economic Cooperation and Development and the World Bank treat it that way.
Index providers use a different lens. They also weigh how easily foreigners can trade in a market, how freely money can move and how convertible the currency is. Judged that way, the two big providers disagree: MSCI still labels South Korea "emerging," while FTSE Russell calls it "developed." (In June, MSCI reaffirmed the "emerging" label, citing tight controls on Korea's currency.)
That technical disagreement quietly reshapes the funds. Because BlackRock's IEMG tracks MSCI, it allocates roughly 5% to 10% of its assets to South Korean equities. In contrast, Vanguard's VWO, which follows the FTSE benchmark, has essentially no exposure to South Korea.
For years, this difference attracted little attention. Emerging-market returns were largely driven by China's growth cycle, commodity prices and global financial conditions. Whether a portfolio included South Korean stocks had only a modest impact on overall performance.
That changed when artificial intelligence became the world's dominant investment theme.
South Korea AI play
An ETF that happened to include South Korea suddenly acquired much greater exposure to AI hardware than one that did not.
Today, South Korea no longer fits the traditional emerging-market model of competing through low-cost labor and standardized manufactured exports. Instead, its leading firms compete through innovation, producing highly differentiated, technology-intensive components that occupy critical positions in the global AI supply chain.
Samsung Electronics and SK Hynix are among the world's leading producers of advanced memory chips, particularly high-bandwidth memory, a key component used in AI accelerators and data centers. As investment in AI infrastructure increased, these firms became beneficiaries of the global technology boom.
The consequence for investors was significant. An ETF that happened to include South Korea suddenly acquired much greater exposure to AI hardware than one that did not.
By early 2026, Samsung Electronics and SK Hynix together accounted for around 5.5% of IEMG's portfolio. Combined with Taiwan Semiconductor Manufacturing Co. $(TSM)$, these companies represented one of the fund's largest sources of exposure: TSMC (13.47%), Samsung Electronics (6.30% plus 0.72% nonvoting) and SK Hynix (5.85%) in July 2026.
In effect, buying IEMG was no longer a simple bet on emerging markets. It had become, in large part, a bet that the world would keep spending on AI. Many of the people who bought this ETF never meant to make that bet. They just wanted to own a broad slice of the developing world.
Diversification isn't what it used to be
Investors who own U.S. tech stocks and an emerging-markets fund full of South Korean chip makers may be far less diversified than they think.
This points to a bigger problem for anyone building a portfolio.
Plenty of U.S. investors buy emerging-markets funds precisely to get away from U.S. tech stocks. But the chip industry is borderless. Nvidia (NVDA) designs AI processors in the United States. TSMC builds many of them in Taiwan. Samsung and SK Hynix supply the memory that makes them run.
These companies sit in different countries and different indexes, yet they all feed the same machine. They all rise and fall depending on the same thing: spending on AI. So investors who own U.S. tech stocks and an emerging-markets fund full of South Korean chip makers may be far less diversified than they think. Spreading your money across the map is not the same as spreading your risk.
The real lesson
It pays to know not just what stocks an index contains, but how it was built.
The split between IEMG and VWO shows that index investing is never truly a hands-off endeavor.
Every index reflects decisions such as which countries qualify, how sectors are defined, what gets included and when to rebalance. Most of the time those decisions barely register, which is why investors ignore them. But a big economic structural shift can turn an obscure judgement into a real source of gains or losses. The rise of AI did exactly that, turning a dusty debate over South Korea's label into one of the year's biggest drivers of returns.
The takeaway reaches well beyond these two funds. It pays to know not just what stocks an index contains, but how it was built. Passive investing doesn't remove the active decisions. It simply hands them to the people who design the index.
Most of the time, you'll never notice. But every so often - usually when the ground is shifting fastest - those invisible choices decide your returns.
Daisoon Kim is scholar at the Andersen Institute for Finance and Economics.
-Daisoon Kim
This content was created by MarketWatch, which is operated by Dow Jones & Co. MarketWatch is published independently from Dow Jones Newswires and The Wall Street Journal.
(END) Dow Jones Newswires
July 20, 2026 10:37 ET (14:37 GMT)
Copyright (c) 2026 Dow Jones & Company, Inc.
Comments