Global funds, same tech bet? The diversification trap Indian investors may miss
Many Indian investors mistakenly believe they diversify their portfolios by adding foreign countries. However, numerous global indices often revolve around similar technology narratives, particularly in semiconductors. To achieve real diversificat...
Global diversification is not simply about adding more countries to a portfolio. Most Indian investors already have their core equity exposure at home, through something like the Nifty 500 index, and are now looking to add the United States, Taiwan, South Korea, China and other emerging markets to the mix. On paper, that looks geographically diversified. But the underlying exposure can still be concentrated.
Taiwan is dominated by Taiwan Semiconductor Manufacturing Company (TSMC), while Samsung Electronics and SK Hynix have large weights in South Korea. Emerging market indices also have significant exposure to Taiwan and South Korea, creating overlap across funds. The result is continued exposure to the same themes— semiconductors, technology, Artificial Intelligence (AI) spending and global demand.
As Radhika Gupta, Managing Director and Chief Executive Officer, Edelweiss Mutual Fund, puts it: “International investing should ideally bring exposure to different businesses, sectors and growth drivers —whether that is global technology, semiconductors, healthcare or global consumer companies. It is about adding a different source of return, not just another geography.”
The real question, then, is not which index to add next. It is which earnings driver is still missing from a portfolio built around India.
At a glance: Global equity indices
Know what you own
Before looking overseas, it helps to understand your home portfolio. Financial services make up 31.2% of the Nifty 500 index. Consumer discretionary adds another 14.8%. Industrials account for 12.1%. Information technology is just 6.5%.
Ravi Kumar T.V., Director at Gaining Ground Investment Services, describes the Nifty 500 as a portfolio built around the Indian credit cycle, domestic consumption, infrastructure and capex, manufacturing, energy and financial intermediation. He says its economic engines are materially different from the S&P 500.
Rahul Bhutoria, Co-founder of Valtrust, explains why that gap matters: “The strongest rationale is to combine India’s domestic demand opportunity with the global innovation and technology cycle.”
Gupta highlights that owning several markets does not automatically mean owning different risks. “Investing in companies across five countries does not mean you have five different risks. Businesses in different markets can still be driven by the same technology cycle, interest rates, commodities or global demand. I would, therefore, look at products and services a company offers and what actually drives its earnings, rather than simply looking at where that company is listed.”
The S&P 500 is the inverse of the Nifty 500: information technology at 37.9%, communication services at 9.5% and financial services is just 12.3%. The correlation between the two indices was 0.19 over the five years from August 2021 to August 2026— and the S&P 500 delivered compounded annual growth rate of 11.19% against 10.01% for the Nifty 500. But low correlation is not the same as low risk, and a single long-run number can conceal far more than it reveals. ”Correlation isn’t a fixed number; it moves with the regime. And it moves in exactly the wrong direction for investors: during periods of stress, correlations between the S&P 500 and major Asian markets have jumped above 0.8, well above their historical average,” said Kunal Valia, Founder of StatLane.
How global markets differ from the Nifty 500
ADVERTISEMENT
Biggest drawdowns
US exposure, done differently
The S&P 500 brings a genuinely different economic character—technology versus financials, global platforms versus domestic consumption— but with its own concentration: its top 10 stocks account for 37.8% against 29.1% for the Nifty 500.
Kumar sums up why the combination works: “S&P 500 plus Nifty 500 is genuinely useful diversification. The economic engines in each of these indices are materially different.” The S&P 500’s largest constituents —Nvidia, Apple, Microsoft, Amazon, Alphabet, Broadcom, Meta, Micron and Tesla—are overwhelmingly exposed to technology, digital advertising, cloud computing, AI infrastructure and digital consumption.
Viram Shah, Founder and CEO of Vested, frames the sector gap: “The same label – IT means 6.5% of your portfolio in India and 37% in the US. That gap is the case for adding the S&P 500 to an Indian portfolio, and it is also the reason an investor holding both should track their combined sector exposure rather than the country split.”
Then there is the Nasdaq 100 index, which many Indian investors treat as interchangeable with the S&P 500. That is a mistake. Information technology accounts for roughly 66.4% of the Nasdaq 100. The index excludes financials entirely.
Gupta draws the line: “The S&P 500 gives you broad exposure to US businesses across technology, healthcare, financials, industrials and consumer sectors. The Nasdaq 100 is much more concentrated towards non-financial, technology-led and large-cap growth companies.” For an investor who already holds the Nifty 500 and the S&P 500,adding Nasdaq 100 is not the same diversification decision. It is an additional bet on a theme that is already significant in the portfolio.
2 countries, 1 cycle
After the US, many Indian investors turn to Taiwan and Korea, attracted by their strong recent returns. In the last five years to 31 August 2026, the Taiwan Stock Exchange Capitalisation Weighted Stock Index (TAIEX) delivered a compounded annual growth rate of 24.96%, and Korea’s KOSPI returned 16.75%. The low correlation numbers look appealing too—both the TAIEX and the KOSPI show a correlation of 0.35 with the Nifty 500, which seems to promise genuine diversification. But these headline correlations hide what is really happening inside. As Ankita Pathak, Head of Global Investments at Ionic Asset, puts it: “Geographical diversification asks where the company is listed; economic diversification asks what drives its earnings. The second question is what matters most.”
TSMC alone makes up 41.47% of the TAIEX, and the next largest names— MediaTek and Delta Electronics—are also semiconductor companies. Taiwan’s market has become, as Kumar describes it, “much closer to a semiconductor ecosystem,” highly sensitive to AI demand, chip inventories and US-China tech policies.
Pathak adds a concentration point that goes beyond Taiwan itself: a single chip-making company now carries a 14% weight in the MSCI Emerging Market index—more than all of India’s combined weight in that index.
Korea’s KOSPI is top-heavy too. Samsung Electronics and SK Hynix together make up 50.42% of the index weightage, but Korea also has exposure to autos, shipbuilding, batteries, industrials, defence, and chemicals through names like Hyundai, Kia, and Doosan. Even so, the semiconduc tor engine dominates.
Valia distinguishes two situations in which India and Korea can appear correlated. Before September 2024, both moved together because capital was broadly flowing into Asia— shared macro exposure, not shared economic structure. Then the AI and semiconductor boom took off.
“Korea surged because it’s the home of Samsung and SK Hynix, the chip manufacturers driving the AI hardware cycle. India, meanwhile, doesn’t have a semiconductor industry. The correlation broke because of structural economic exposure,” says Valia. An investor buying Korea today on the strength of that 0.35 correlation is relying on a number calculated across five years that included two very different relationships between India and Korea—one driven by shared regional capital flows, and a more recent one driven by Korea’s specific bet on AI hardware. The 0.35 figure doesn’t tell you which of those relationships is still in play.
Valia’s conclusion: “An investor adding Taiwan or Korea on top of US tech exposure may just be increasing exposure to the same bet, not diversifying away from it.”
Shah illustrates how fast this concentration can move returns: “The KOSPI rose roughly 80-90% in the first five months of 2026 on the AI hardware trade, then fell about 20% from its June peak in a matter of weeks when investors questioned the pace of AI capital spending. That is a semiconductor cycle, not a Korean economy story.”
The maximum drawdown data makes this concrete. Over the last five-year period, KOSPI’s biggest drawdown was -38.6%, happening as recently as June to July 2026—lasting just 38 days. The TAIEX’s worst drawdown was -31.6%, lasting 294 days from January to October 2022. Geography can be very different; the underlying driver, semiconductor and AI capex, remains the same.
China: a risky bet
China is a different story altogether. The Nifty 500-MSCI China correlation over the study period was 0.11—the lowest of any pair in the matrix and well below even the Nifty 500-S&P 500 correlation of 0.19—making China the most statistically distinct market for Indian investors. China’s equity cycle responds to domestic credit policies, property cycles, government regulation, fiscal stimulus and state-driven sectors. Its current market leadership—Tencent, Alibaba, Chinese banks, Xiaomi, Meituan, PDD, and insurers—is structurally different from both India and the US.
There’s a currency layer to the China story too, Bhutoria points out. China’s currency doesn’t move in lockstep with the dollar, so an unhedged investment there depends not just on how Chinese stocks perform, but on how the rupee moves against the yuan — a very different, less predictable relationship than the rupee dollar link that has generally worked in Indian investors’ favour when they invest in dollar markets like the US. But Bhutoria is careful to frame this correctly: “Currency should be viewed as a risk factor, not a one-way return enhancer. Investors should not assume that currency gains will always supplement equity returns.”
But the diversification benefit comes packaged with serious risk. China’s maximum drawdown was -52.1%, lasting 419 days from September 2021 to October 2022. Its five-year return was -4.39%—the only negative return in this study.
As Kumar notes, “You cannot take the diversification benefit without accepting those risks.” Bhutoria argues that selective exposure to Chinese businesses makes sense when valuations become sufficiently attractive, given the unique exposure to manufacturing, robotics and e-commerce. But this is a considered allocation, not a passive diversification move.
Correlation matrix
(A value of +1.0 means they move in perfect lockstep; 0 means no relationship at all; −1.0 means they move in exactly opposite directions. Data based on daily closing prices from 1 Sep 2021 to 31 Aug 2026. Source: Investing.com; iShares ETF prices used as proxies for MSCI Emerging Markets ex China and MSCI China.)
How to invest overseas
Note that almost all overseas mutual funds in India are currently closed for fresh investments. For those looking to invest now, options include direct overseas exchange-traded funds (ETFs) listed on Indian exchanges, funds available through the GIFT City route, or direct equity and ETF investing abroad under the Liberalised Remittance Scheme, which allows up to $250,000 per individual per year for overseas investments.
The diversification illusion
For investors who want broad international exposure, a broad emerging market fund—something like MSCI EM ex-China- seems like the simplest answer. It has 602 constituents and promises access to Brazil, South Africa, Indonesia, Mexico and dozens of other economies.
The reality is narrower. In the MSCI Emerging Market ex-China index, the top-10 weight is 42.64%, and Taiwan and South Korea alone account for 60.83%. India is next at 14.17%.
As Shah notes: “More stocks do not automatically mean more diversification.” Kumar adds: “An Indian investor buying an EM fund may think he is buying Brazil, Mexico, South Africa, Indonesia, Thailand, Middle East and Eastern Europe. But economically he is largely buying Taiwan plus Korea plus China plus some more India.”
Valia’s data on what has driven EM returns makes the point sharper: three chip companies —TSMC, Samsung Electronics and SK Hynix— drove roughly 57% of the MSCI EM index’s gains over the past year, and technology accounts for approximately 45% of the index. Information technology in the MSCI EM ex-China index runs 42.9 percentage points higher than in the Nifty 500—more technology-heavy than even the S&P 500. A country-level correlation table would never surface this. Valia puts the point directly: “Two countries can look completely different on a map and still be the same trade underneath.”
Shah argues that the right measure is not correlation alone but a combination of factor exposure, sector overlap, concentration and downside behaviour. “Factor analysis is particularly useful because it identifies whether different markets are responding to the same drivers—such as growth, value, momentum, interest rates, commodities or technology.” By that test, much of the emerging market index is responding to the same technology driver as the US and the Nifty 500’s own IT services sector.
For a Nifty 500 investor, a broad EM fund duplicates India exposure, semiconductor exposure and Asian market dynamics, while adding far less new diversification than the label suggests.
Away from the AI trade
If the goal is to avoid adding more of the same technology and AI cycle, two markets stand out: Europe and Japan. The MSCI Europe index has 396 constituents and a P/E of 17.52. Its largest names—ASML, HSBC, Roche, Novartis and Shell— span industrials, healthcare, financials and consumer staples. Europe runs +6.8 percentage points above the Nifty 500 in industrials and +5.3 percentage points above it in healthcare.
Kumar identifies European exposure to “industrials, aerospace and defence, health care, financials, luxury brands, energy and consumer staples, with far less dependence on mega-cap technology” as its key contribution. Its five-year return was 6.96%, correlation with the Nifty 500 was 0.39, and its maximum drawdown was -32.4%. The numbers are modest, but the earnings drivers are different from both India and the US.
Japan adds precision manufacturing, robotics, automobiles, machinery and trading houses. As Kumar notes, “While parts of Japan benefit from the AI and semiconductor cycle, its market is much less dominated by digital platforms and mega-cap growth companies.” For an investor who has already loaded up on US technology and Asian semiconductors, Europe and Japan offer the cleaner counterweight.
According to Bhutoria, for most investors starting out with international exposure, a broad allocation with the US as anchor makes more sense than building a portfolio of country-specific funds immediately. But as the portfolio grows, “selective allocations to other markets can make sense where they add a genuinely different exposure. Europe and Japan are the clearest candidates for that second layer,” he says.
When markets fall together
The drawdown data from the last five years is humbling. The Nifty 500’s worst drop was -18.8% between September 2024 and February 2025. The S&P 500’s worst drop was -25.4% from January to October 2022. MSCI Europe fell -32.4%. Taiwan fell -31.6%. MSCI EM ex-China dropped -30.3%. MSCI China fell -52.1% over more than a year.
During periods of global stress, correlations rise precisely when protection is most needed. As Pathak explains: “Cross-market correlations spike in every major risk-off episode, whether it’s the 2008 global financial crisis, the 2013 taper tantrum, March 2020, or even 2022. Every single equity market fell by about 25 to 35% in the middle of Covid-19, regardless of fundamentals.”
Shah provides the historical numbers. In 2008, the Nifty 50 fell about 52% over the calendar year while the S&P 500 fell about 38%. In early 2020, the Nifty fell roughly 38% from its January peak to the March low in about six weeks; the S&P 500 fell about 34% over a similar window. Two different markets, two different economies—and near-identical behaviour when liquidity dried up.
This is why long-run average correlation can mislead. Pathak notes that India-US correlation “was very low in the 2000s, when the US did well, but India did even better. But as we speak, in the last few years we’ve had patches of very high correlation between India and the US.” In mid-2022, during the global rate-hike sell off, the 30-day correlation between the Nifty 50 and the S&P 500 touched 0.68— the highest in over a year, according to Shah. A 20-year average hides that range entirely. Different return drivers reduce the tendency to fall together—but they do not eliminate it.
Five questions
Rather than chasing the market with the lowest historical correlation, investors would do better to ask five questions before adding any overseas market.
First: what do you already own? The Nifty 500 is a financials, consumer and industrial story. Anything that replicates those exposures adds little. “Understand what you already own, and then identify what is missing. That could be a different sector, business model, economic cycle or currency exposure. The objective isn’t to diversify across every possible dimension. It is to build a portfolio where the underlying drivers are meaningfully different,” says Gupta.
Second: what new earnings driver does this market bring? The US adds global technology and AI; Taiwan adds semiconductor manufacturing; Korea adds semiconductors and industrials; China adds a domestic policy-driven cycle; Europe adds industrials, healthcare and consumer staples; Japan adds manufacturing and machinery. These are the true diversifiers, not the flags.
Third: how concentrated is the index? Counting constituents misleads. A 1,000-stock index where one company is 40% of the weight is functionally a single stock bet. Look at top-10 concentration and the largest stock’s share. Pathak suggests a test: decompose an index into its earnings drivers—domestic consumption, global trade, commodity prices, interest rates, technology capex—and check the overlap with what you already own.
Fourth: what happens during stress? Rolling and downside correlations are more honest guides than a long-period average that hides regime changes. Experts recommend looking at the marginal effect of adding a market: does it change the portfolio’s volatility, maximum drawdown or conditional drawdown? Diversification should reduce the downside while capturing the upside, and that trade-off needs to be visible at the portfolio level, not just in a correlation table.
Fifth: are you adding a new market, or an existing theme at another address? A portfolio spread across five countries that are all secretly betting on the same AI capex cycle isn’t diversified at all.
