Emerging markets are where investors most often get the judgment right and the conclusion wrong: the read on growth is frequently correct, the inference about returns frequently is not. Four gates sit in between — the leakage in per-share value, who the cash flow belongs to, the currency, and the price you pay for the story.
The Principle
You are not buying a country's GDP; you are buying the per-share value of specific companies. Three leaks sit between them: dilution, the unlisted portion, and the valuation you pay on entry.
Origin · Quote
"During that same 17 years, the U.S. GNP — that is, the business being done in this country — almost quintupled, rising by 370%."
— Warren Buffett, "Mr. Buffett on the Stock Market", Fortune, 1999-11-22
A Deeper Reading
Buffett was describing year-end 1964 to year-end 1981: the Dow went from 874.12 to 875.00, seventeen years without moving, while the economy grew 370%. The reason is that growth enters the numerator while share count and valuation enter the denominator. There are three leaks.
Dilution — Bernstein and Arnott called it "the two percent dilution" in 2003: aggregate earnings growth and per-share earnings growth differ by roughly two percentage points year after year, consumed by net new issuance, and the hungrier a market is for capital the wider that gap. The unlisted portion — the fastest-growing part of an economy often sits in private hands. Entry price — once growth is written into the price, what you buy is somebody else's expectation.
A Classic Case
China is the plainest sample: nominal GDP rose from roughly RMB 10 trillion in 2000 to about RMB 126 trillion in 2023, more than twelvefold, while the Shanghai Composite went from 2,073 at the end of 2000 to 2,975 at the end of 2023. The cross-sectional evidence points the same way: Dimson–Marsh–Staunton across nineteen countries over more than a century, and Ritter (2005) across sixteen countries, both find a negative correlation between real per-capita GDP growth and real equity returns.
Limits & Decision Checklist
What fails is only the shortcut from macro growth to index return. At the company level, high ROIC plus reinvestment at high returns is exactly what compounding is (Day 11). And the negative correlation is a long-run cross-sectional result, not a timing tool.
- Does my thesis stop at "this country will grow", or reach "this company's per-share value will grow"?
- How large has net issuance been over the past five years? By how much does per-share earnings growth lag aggregate earnings growth?
- Does this index cover the most dynamic part of the economy, or the part that needs capital most?
- How many years of high growth are already priced in today?
Essence · Reflection
Growth enters the numerator; dilution and valuation enter the denominator — whoever watches only the numerator spends years puzzled that being right did not pay.
Write out the bull case for one emerging-market holding, then delete every sentence about national or industry growth rates. What survives is your actual reason.
The Principle
Governance is not a morality score, it is a valuation input: it determines how much of the cash a business generates actually accrues to outside minority holders — and whether that share can be rewritten unilaterally.
Origin · Quote
"Show me the incentive and I will show you the outcome."
— Charlie Munger, "The Psychology of Human Misjudgment", Harvard Law School, 1995
A Deeper Reading
A DCF carries an assumption so safe in developed markets that nobody states it: free cash flow accrues to all shareholders in proportion to ownership. Where governance is weak it breaks in three ways — related-party transactions move profit to entities the controller owns outside the listed company; discounted issuance dilutes minorities; and the controller's objective function is simply not per-share value. The right response is not a vague "governance score" but two concrete adjustments: a lower multiple you are willing to pay (this is what the emerging-market discount economically means) and a shorter horizon you dare extrapolate. What makes this risk dangerous is that it is discrete — not cash flow coming in 20% light, but discovering one morning that the cash was never there.
A Classic Case
On January 7, 2009, Ramalinga Raju, chairman of India's Satyam Computer, admitted that roughly $1.04 billion of cash and bank balances on the books was fabricated. The stock fell about 78% that day. The company had recently won a global corporate-governance award and was audited by one of the Big Four — the appearance of compliance is not governance. Brazil's Petrobras and the "Lava Jato" scandal is another shape: the 2015 accounts took a write-off of roughly BRL 6.2 billion tied to corruption, its ADR fell from about $20 in September 2014 to under $3 by early 2016, and in 2018 the company settled a U.S. class action for about $2.95 billion and paid a further $853 million to resolve foreign-bribery charges.
Limits & Decision Checklist
Governance is a variable, and treating it as a fixed label forfeits an entire class of opportunity: the Tokyo exchange asked companies trading persistently below book value to publish improvement plans in 2023, Korea launched its "value-up" program in 2024, buybacks and dividends rose in both, and multiples re-rated — improving governance is itself a revaluation. The opposite misuse is treating "weak governance" as "uninvestable": the correct conclusion is demand a lower price and set a lower position cap.
- Is the controlling shareholder's objective function per-share value, or something else — and can I name it?
- Over five years, any discounted placements, unusual related-party deals, or frequent auditor changes?
- Are dividends and buybacks a standing policy, or a one-off gesture?
- If this company's cash were fabricated, at what step would I catch it? If never, cut the position.
Essence · Reflection
Valuation tells you what a business is worth; governance decides how much of it is yours — get the first wrong and you lose once, get the second wrong and you lose everything.
Pick one emerging-market company you own and write down who controls it and how that party makes money. If you cannot, you hold a lottery ticket, not an ownership stake.
The Principle
Buying an asset priced in a foreign currency means simultaneously opening a currency position you never chose. It does not stack independently but in the same direction — it tends to break exactly when you most need it not to.
Origin · Quote
"Original sin is a situation in which the domestic currency cannot be used to borrow abroad or to borrow long term, even domestically."
— Barry Eichengreen & Ricardo Hausmann, "Exchange Rates and Financial Fragility", NBER, 1999
A Deeper Reading
The risk has three layers, each worse than the last. Translation: local-currency returns must be converted into your accounting currency, and the exchange rate takes a slice. Fundamentals: a company that borrows in dollars and earns in local currency sees depreciation magnify its liabilities directly — this is original sin, where a currency mismatch turns exchange-rate risk into solvency risk. Correlation: depreciation usually arrives alongside falling equity prices, capital flight and forced rate hikes — three faces of one event, which is precisely where diversification fails.
Hedging addresses only the first layer, at an explicit cost: the forward cost is roughly the interest-rate differential. Hedge a high-yielding currency for the long run and you pay a certain few percent a year, which is often the entire excess return you were hoping for. The mainstream stance is therefore to accept the first layer and avoid the second.
Turkish equity index, 2018 · in lira
about −20%
Turkish equity index, 2018 · in dollars
about −40%
Almost the entire gap is lira depreciation. Same companies, two currencies, two conclusions.
A Classic Case
On July 2, 1997 Thailand abandoned its dollar peg and the baht fell from about 25 to roughly 55 per dollar within six months; the Indonesian rupiah went from about 2,400 to more than 16,000 by early 1998, and Indonesia's real GDP contracted about 13% that year. Companies there routinely funded local-currency assets with short-dated dollar debt — this was not an earnings decline, it was a balance sheet detonated by an exchange rate. A milder recent example: in 2018 the lira lost about 30% against the dollar and the Argentine peso about half.
Limits & Decision Checklist
Depreciation is not always bad: an exporter with local-currency costs sees margins widen, and mean reversion from a deeply undervalued currency has produced real returns. Two misuses matter — systematic long-term hedging (paying a certain carry cost for uncertain protection) and ignoring currency entirely (importing dollar-asset position discipline unchanged).
- Do this company's revenue and liability currencies match? How large is the mismatch?
- Are the historical returns I am looking at in local currency or in my accounting currency?
- If the local currency fell 30%, would my loss come from translation or from the company's ability to service debt?
- If I hedge, how many percentage points of carry am I willing to pay each year — and is it worth it?
Essence · Reflection
Every foreign-currency asset comes with a currency position attached; it does not diversify your risk, it stands on the same side as your risk at the worst moment.
List your holdings again, this time by currency. If exposure to one non-accounting currency exceeds what you expected, was that a decision — or the absence of one?
The Principle
The payoff from diversification comes from independent, weakly correlated sources of return, not from owning more country names. And any country's story is worth buying only while the price has not finished telling it.
Origin · Quote
"Investment success doesn't come from buying good things, but from buying things well."
— Howard Marks, The Most Important Thing (2011)
A Deeper Reading
"Good things" are the visible part: younger populations, lower penetration rates. Those judgments are often right — and often already in the price. Marks insists on separating two questions: is the country good, and is the price good? Most people ask only the first.
The diversification half is eroding too. Correlation between emerging and developed markets has risen steadily over three decades — global supply chains, a shared dollar liquidity cycle, and the same pool of cross-border capital mean another continent no longer means another source of return. When correlation rises and volatility does not fall, the trade-off gets worse.
A Classic Case
India is the best two-sided sample. The positive side: the Nifty 50 rose from roughly 1,100 in 2003 to about 24,000 in 2024, roughly 16% a year in rupees. Over the same span, the rupee went from about 46 to about 83 per dollar, leaving a dollar investor with roughly 12–13% a year — and that gap of three or four points is precisely the second position from the previous principle.
The other side is price. As the "next China" narrative spread, India's weight in the MSCI Emerging Markets index rose from under a tenth around 2020 to roughly a fifth by 2024, with valuations well above the emerging-market average. The reference point is 2011–2020: MSCI Emerging Markets returned only low single digits a year in dollars while the S&P 500 compounded near 14% — and that decade opened at the loudest moment of the BRICs story.
Limits & Decision Checklist
None of this justifies "every story is a trap": India's last twenty years show some stories are true, and staying absent has a cost of its own. Nor is a valuation premium automatically an error — better governance and higher earnings quality deserve a premium; the question is how much. One boundary is actually executable: write the allocation as a rule rather than a mood — a weight cap plus scheduled rebalancing, so you trim automatically when the story is loudest and add when it is coldest.
- Am I allocating because "it will grow", or because "the price has not finished counting the growth"?
- What is this sleeve's actual correlation with what I already own — did I check, or assume?
- Do I have a weight cap and a rebalancing rule set in advance, or do I decide each time?
- If this market lags for a decade, can I hold long enough for rebalancing to work?
Essence · Reflection
Being right about a country and making money in it are two things separated by price — most people's error is not in the forecast but in what they paid.
Recall the last time you bought because "this country has a future". Had you estimated how many years of growth the price already contained? If not, you bought a narrative, not a judgment.
Going Deeper
If GDP growth and equity returns are negatively correlated over the long run, why does capital keep flowing to fast-growing economies?
Because the negative correlation is a long-run cross-sectional statistic, while capital faces short-run opportunities in the time series. Periods of fast growth bring earnings upgrades and spreading narratives, which genuinely lift prices in the near term — the negative correlation is the consequence of that lift: valuation is pulled forward, and the return that remains is pushed into the future as a negative. The inference is not to avoid fast-growing markets but to treat entry price as a second decision, independent of the growth judgment.
If governance risk can be priced through a lower multiple, why is a position cap still necessary?
Because a multiple handles continuous risk — cash flow 20% lower, growth two points slower. Governance failure is usually discrete: the cash is fabricated, the assets are moved. For risks whose probabilities cannot be estimated reliably and whose loss distribution is fat-tailed, adjusting the discount rate is the wrong instrument. A position cap does not raise expected return; it guarantees no single governance event ends your ability to keep investing.
Will AI and automation erode the traditional comparative advantage of emerging economies?
The pressure is real: the path of absorbing offshored manufacturing with low-cost labor may be compressed by automation, weakening the link from factories to employment and income — what is often called premature deindustrialization. But new variables appear too: digital infrastructure lets some economies skip physical stages, and domestic digital consumption becomes an independent source of growth. For the framework this means re-reading emerging markets as owners of domestic demand rather than exporters of cheap inputs, shifting the analytical weight from export competitiveness toward the depth of domestic consumption and the quality of institutions — which lands right back on the first three principles: governance, currency, price.