MSCI World vs FTSE All-World vs S&P 500: which index should your core ETF track?

The S&P 500 holds roughly 500 large US companies — one country, about 80% of the US market's value. MSCI World holds around 1,400 large and mid-cap stocks across 23 developed markets; despite the name, it excludes emerging markets entirely, and the US makes up roughly 70% of it. FTSE All-World is the broadest of the three: about 4,000 stocks across developed and emerging markets, with the US near 60% and around 10% in emerging economies like India, Taiwan and Brazil.

So the real choice is a diversification ladder: US only, developed world, or (nearly) everything. Each step down the ladder dilutes the US weight and adds countries — at the cost of holding more of the world's slower-growing markets, which is precisely the point of diversification and precisely what makes it uncomfortable in decades when the US wins.

Choosing a broader index diversifies you less than the stock counts suggest. Because all three are market-cap weighted, the same handful of US technology giants sit at the top of each: the top ten holdings are near-identical, representing roughly a third of the S&P 500, a quarter of MSCI World, and a fifth of FTSE All-World. Buying "4,000 stocks" still means your returns are substantially driven by the same few companies as someone holding 500.

This is not a flaw to fix but a property to understand. Cap weighting concentrates in whatever has done well; it also automatically rotates when leadership changes, with no action required from you. Investors who find today's US-tech concentration genuinely uncomfortable can pair a global fund with a small-cap or equal-weight satellite — but should recognise that this is an active bet against the market's current pricing, with the tracking regret that entails when the giants keep winning.

Over the past 15 years the ranking has been simple: S&P 500 first, MSCI World second, All-World third — a direct consequence of US outperformance. It is tempting to read that as a verdict. It is actually a single historical draw: in the 2000s the ordering was reversed, with the S&P 500 delivering a lost decade (negative real returns from 2000 to 2009) while emerging markets tripled.

The honest framing: nobody knows which regime the next 30 years hold, and the three indices are different-sized bets on the answer. The S&P 500 is a concentrated bet that US exceptionalism continues. All-World is the closest thing to "no view" — you hold the world roughly as the market prices it. MSCI World sits in between, with the notable quirk that excluding emerging markets is itself an active choice hiding inside a passive-sounding name.

In practice the decision is often settled by implementation details. S&P 500 UCITS ETFs are the cheapest funds in Europe — TERs from 0.03% — and among the most liquid instruments on any exchange. MSCI World trackers cluster around 0.12–0.20%, with the largest funds (like the €90bn+ iShares Core MSCI World) offering excellent spreads. FTSE All-World funds run 0.15–0.22%, with Vanguard's VWCE the category's flagship; newer entrants have pushed costs down.

Check three things beyond TER: whether your broker executes the fund free in a savings plan (this often decides the question by itself for monthly investors), the share class (accumulating for most builders), and fund size. And resist combining them: MSCI World plus an S&P 500 fund is not diversification — it is doubling your US weight while feeling diversified. One core fund, chosen once, is the entire job.

If you want a defensible default: FTSE All-World (or its MSCI equivalent, ACWI) for maximum breadth with zero maintenance, MSCI World if you prefer developed markets only, S&P 500 only if you have thought about the 2000s and accept single-country concentration with open eyes. All three, held for decades at low cost, have been excellent; the biggest risk is not picking the "wrong" one but switching between them after every regime change.