Key takeaways
- Across the ten emerging markets tested here, real GDP per capita growth from 2015 to 2025 correlated -0.11 with MSCI index returns over the decade to 31 July 2026, at a p-value of 0.76.
- China grew real GDP per capita 5.37% a year over that decade, the fastest of the ten. Its MSCI index returned 5.05% a year in dollars, fifth of the ten.
- From 1900 to 2025 the DMS emerging market index returned 6.9% a year in dollars against 8.5% for developed markets. From 1960 to 2025 the order reverses, 10.9% against 9.6%.
- Ritter's 2022 study of 15 emerging markets found real stock returns correlated 0.27 with GDP growth, 0.54 with earnings per share growth and 0.47 with dividends per share growth.
- Since 31 December 1987 the MSCI Emerging Markets index returned 10.43% a year gross against 9.02% for MSCI World, at Sharpe ratios of 0.42 and 0.43.
The growth arrived. The shareholder return didn't follow it
You've been told emerging markets grow faster, so they should return more. It's the reason most people give for owning them at all. Does the growth actually reach the shareholder?
On the evidence, not reliably. We took ten emerging markets, measured how fast each grew real GDP per head over the ten years to 2025, and lined that up against what each country's MSCI index returned over the ten years to 31 July 2026. The correlation is -0.11. The p-value is 0.76, which is another way of saying the number is indistinguishable from noise with ten countries. Rank the two lists instead of pairing the raw figures and the Spearman correlation is 0.03.
That is not a finding about one unlucky decade. It's the same answer the long-horizon academic work has been giving since 2005, and the rest of this piece is about why the mechanism produces it.
The growth gap itself is real, and it is wide
Start by conceding the premise, because it's true. The IMF's World Economic Outlook data put emerging market and developing economies at 4.4% real GDP growth in 2025 against 1.9% for advanced economies, with world growth at 3.4%. That's a gap of 2.5 percentage points, and gaps of that size compound into something enormous over a working life.
So the first half of the claim survives contact with the data. Emerging economies do grow faster. The argument breaks at the second half, where that growth is supposed to turn into a return for someone who owns a slice of the listed companies.
Ten countries, ranked by growth, and the returns go nowhere
Here's the test in full. Real GDP per capita comes from the World Bank's World Development Indicators, in constant local currency units, for 2015 to 2025. Returns are MSCI's own country index figures to 31 July 2026, annualised over ten years, gross, in US dollars.
The chart above lists the ten countries in order of growth, fastest first, and plots what each one's stock market paid. China grew fastest at 5.37% a year and returned 5.05%. India grew second fastest at 4.77% and returned 8.51%. Then it falls apart. Turkey grew 3.68% and returned 2.31%. The Philippines grew 3.40% and returned -1.41%. Indonesia grew 3.31% and returned -3.11%.
Now look at the other end. Brazil grew slowest of the ten, 0.87% a year, and returned 7.54%. Thailand grew 1.77% and returned 4.58%. Korea grew 2.19%, third slowest, and returned 15.79% a year, the best of the ten by a wide margin. The three slowest growers of the ten came first, fourth and sixth on returns.
One number is worth holding onto for the counter-case later. Over that same decade MSCI World returned 13.29% a year gross in dollars. Nine of these ten emerging markets trailed it. Only Korea beat it.
Two caveats on the arithmetic. The windows don't line up perfectly: national accounts are annual and run to December 2025, while the index figures run to July 2026. And the returns are nominal dollar returns while the growth rates are real. Deflating every country's dollar return by the same US inflation rate is a linear transformation, so it moves all ten numbers together and leaves the correlation exactly where it is. Taiwan is missing, because the World Bank does not publish national accounts for it. That omission is worth naming out loud, because MSCI Taiwan returned 23.34% a year over the same ten years and is the single largest country in the index at 26.63%.
A century of evidence on GDP growth and stock returns says the same thing
Ten countries and one decade is a small sample, and a small sample proves very little on its own. The reason to take it seriously is that it agrees with much longer studies.
The version most often cited is Jay Ritter's, published in the Pacific-Basin Finance Journal in 2005. He credits the first sighting to Jeremy Siegel's Stocks for the Long Run and to Dimson, Marsh and Staunton's Triumph of the Optimists, both published in 2002. Over 1900 to 2002, across 16 countries, Ritter found the cross-country correlation of real stock returns and real per capita GDP growth was -0.37. He updated it in 2012 with 19 countries over 1900 to 2011 and got -0.39, with a p-value of 0.10. For 15 emerging markets over 1988 to 2011 the figure was -0.41.
Negative correlations that large invite over-reading, and Ritter doesn't over-read them. Run 21 countries from 1970 to 2011 and the correlation is -0.04, which is nothing. His 2022 paper with Jason Hsu, Phillip Wool and Harry Zhao redid the developed-market test through 2019 and got -0.31 with a p-value of 0.17, then redid the emerging-market test through 2019 and got +0.19, with a p-value of 0.50.
So the honest summary of the academic record on GDP growth and stock returns is not "faster growth hurts". It's that the sign flips with the sample and never reaches significance in either direction. As Ritter's co-authored paper puts it, "there is no statistical relationship between GDP growth and stock market performance."
GDP counts output. A shareholder owns dividends per share
The mechanism is where this stops being a statistical curiosity. GDP measures what a country produces. A share measures a claim on one company's dividends, divided by the number of shares outstanding. Those two quantities can move apart for years, and in fast-growing economies they usually do.
Ritter gives three routes. Growth that comes from more capital and more workers doesn't raise the value of the companies that already exist, because the new capital arrives inside new companies. Growth from technology reaches shareholders only where firms hold the gains rather than competing them away to customers and staff. And growth that everyone can already see gets paid for in advance, in the price.
That third route is the one that bites hardest in emerging market equities. Ritter's own framing is blunt: "Apparently investors overpay for expected growth, and this overpayment more than offsets the benefits of unexpected growth." A market priced at a high multiple hands the investor a lower dividend yield for the same stream of dividends, and dividend yield is a large share of long-run return.
There's a fourth route the arithmetic makes obvious once you see it. If a fast-growing economy funds itself by issuing shares, earnings per share can crawl while total earnings sprint. Ritter notes that China's market capitalisation grew to roughly $4 trillion by the end of 2011 largely through hundreds of new listings, including its four largest banks. Existing shareholders paid for that growth through dilution.
Earnings per share tracks returns where GDP does not
The 2022 paper does the thing that makes the whole argument land. It takes the same 15 emerging markets and runs the correlation against three different growth measures rather than one.
Real per capita GDP growth correlated 0.27 with real stock returns, at a p-value of 0.34. Real earnings per share growth correlated 0.54, at a p-value of 0.04. Real dividends per share growth correlated 0.47, at 0.08. Only the per-share measures clear a conventional significance threshold.
China is the case study inside its own table. Over 2000 to 2019 China grew real GDP per capita 8.4% a year. Earnings per share grew 7.9% and dividends per share 8.0%, and the market returned 5.5% a year in real local-currency terms. That is a market where per-share growth came close to the headline growth, and the return still landed below both.
The practical reading is narrow and worth keeping narrow. It isn't that growth is irrelevant. It's that the growth an equity investor holds a claim on is per-share growth, and the gap between a country's GDP growth and its listed companies' per-share growth is the part nobody quotes.
The strongest case on the other side, stated properly
There's a serious counter-argument and it deserves the real numbers rather than a caricature.
Emerging markets have, over some long windows, simply won. MSCI's own factsheet shows the MSCI Emerging Markets index returning 10.43% a year gross in dollars since 31 December 1987, against 9.02% for MSCI World. That's 1.41 percentage points a year over 38 years, which is a very large edge. The 2026 UBS Global Investment Returns Yearbook, built on the Dimson, Marsh and Staunton database, finds the same thing from a 1960 start: emerging markets 10.9% a year against 9.6% for developed markets to 2025.
Push the start back to 1900 and it reverses. The same Yearbook puts the emerging market index at 6.9% a year against 8.5% for developed markets over 1900 to 2025. The authors are careful about why this is hard to measure, noting that "several countries that we today regard as developed would once have been classified as EM, including Finland, Japan, Portugal, and Spain." A country that succeeds gets reclassified out of the emerging index, which is not a neutral thing to do to a long-run average.
There's also a time-series version of the growth claim that the cross-sectional test doesn't touch, and the Yearbook supports it. Sorting years rather than countries, the authors find that "for both equities and bonds, real returns tend to be higher when economic growth is higher and inflation is lower." A good year for the economy tends to be a good year for the market. That is a different claim from "a country that grows faster over 30 years pays its shareholders more over 30 years", and only the second one is the allocation argument.
Then there's risk. Since 1987 the emerging index earned its 1.41-point return edge at a Sharpe ratio of 0.42 against MSCI World's 0.43, so per unit of volatility the two are the same investment. Ten-year annualised standard deviation runs 17.44% against 14.85%. The worst drawdown was 65.14%, from 29 October 2007 to 27 October 2008, against 57.46% for developed markets.
What this test cannot tell you
Ten countries is ten data points. A correlation of -0.11 on ten observations has a confidence interval wide enough to contain almost any modest true relationship, and that is precisely why the p-value of 0.76 is the honest headline rather than the -0.11. The finding here isn't "growth hurts". It's "you can't see a relationship in this data".
The decade to 2026 is also one regime. It contains a pandemic and a semiconductor cycle that lifted two north Asian markets a long way. The index numbers are gross returns, before the withholding tax and platform costs a real investor pays on foreign holdings, and those costs land harder on emerging market equities than on domestic ones.
Country indices are not economies either. MSCI Emerging Markets holds 1,178 companies across 24 countries, and its four largest countries are Taiwan at 26.63%, China at 21.38%, Korea at 20.33% and India at 11.66%. An index that concentrated is a bet on a handful of listed sectors, not a claim on emerging-world output. This is the same measurement problem that shows up in fund overlap and mega-cap concentration, and it applies to a country index as much as to a fund.
Finally, none of this speaks to how much of a portfolio sits abroad in the first place. That question turns on home bias and international equity allocation, and on whether the currency exposure is hedged, which is its own argument covered in currency hedging and hedge ratios.
What would change the conclusion
Three things would move this, and each is measurable rather than rhetorical.
The first is dilution closing. If emerging market earnings per share started compounding at something near the pace of GDP per capita, the mechanism that breaks the link would be gone, and the 0.27 correlation would have a reason to climb toward the 0.54 that per-share growth already achieves. That's an observable series, not a forecast.
The second is valuation. On 31 July 2026 the emerging index traded at a forward price/earnings ratio of 10.35 against 18.76 for MSCI World, with a dividend yield of 2.03% against 1.53%. Ritter's 2005 abstract ends on exactly this point, that countries with high growth potential "do not offer good equity investment opportunities unless valuations are low." A wide enough discount changes the arithmetic without anything changing about growth.
The third is the index itself. If the 26.63% Taiwan weight and the 40.79% information technology weight keep rising, the question stops being about emerging economies and becomes a question about a few semiconductor supply chains. At that point the growth-versus-returns test we ran here is measuring something other than what its name suggests, and the answer would need re-deriving from the constituents up. LedgerTouch tracks country and sector weights on holdings you already own, which is where that check starts.