Does an S&P 500 ETF diversify an MSCI World ETF? 25 years of data

An S&P 500 ETF does not add a new source of return to an MSCI World ETF: it adds more of the largest one already inside it. From January 2001 to August 2026, in euros, their monthly returns had a correlation of 0.97, and a 50/50 mix of the two moved with MSCI World alone at 0.99.

What the S&P 500 changes is the US share, from 72.8% in an MSCI World tracker to about 86% in a 50/50 mix. That tilt added return after 2010 and cost return before; it is a decision about concentration, and it is worth making deliberately.

Key takeaways

1. The two indexes move almost as one. Over 308 months, MSCI World explained 94% of the S&P 500's monthly variance in euros, and the rolling three-year correlation was never lower than 0.87 (October 2014, rounded).
2. The overlap is structural. At 30 October 2026, 72.8% of MSCI World was invested in US companies, so a World ETF already holds most of an S&P 500 ETF.
3. The US tilt paid off in one era and not the other. The S&P 500 trailed MSCI World in every calendar year from 2002 to 2007, then led in 11 of the 15 years from 2011 to 2025.
4. The mix did not reduce losses. The worst month-end drawdown went from 49.1% for MSCI World to 51.7% for a 50/50 mix and 54.6% for the S&P 500 alone.

The question behind the question

Many investors in Europe hold an MSCI World tracker and an S&P 500 tracker side by side. In the first quarter of 2025, ETFs on US indices, mostly the S&P 500, accounted for nearly 40% of French retail investors' trades in ETFs under the AMF's remit, and MSCI World trackers for 20%, according to the AMF's 2025 market mapping. Two funds look like diversification. Whether they are depends on whether their returns come from different sources.

We measured that directly, using the MSCI World net index from MSCI and the S&P 500 total return index, both in euros, from January 2001 to August 2026. We then checked the result against the two largest iShares trackers listed on Xetra.

How closely the two move

Source: Copio calculation; monthly levels and returns (CSV), rolling 36-month correlation (CSV).

The investable check agrees. From January 2011 to September 2026, the month-end Xetra closes of the iShares Core MSCI World and iShares Core S&P 500 UCITS ETFs had a correlation of 0.97 as well (ETF closes (CSV)).

For comparison, over January 2006 to April 2026 the correlation between MSCI World and EUR-hedged government bonds was 0.01, and with gold -0.09 (correlation matrix (CSV)). Those are different engines. The S&P 500 is not.

Why the correlation is so high

MSCI World is weighted by company size. At 30 October 2026, US companies made up 72.8% of it, up from 52.4% in January 2012, according to MSCI's index factsheet; its ten largest holdings were all American. When the largest US companies move, they move both indexes.

Combining the two funds changes the weights:

The US shares are our arithmetic on MSCI's 30 October 2026 country weight. Our pulse note shows what sits inside the index.

What adding the S&P 500 changed, 2001 to 2026

Source: Copio calculation; mix results (CSV).

Over the full period, each step towards the S&P 500 raised annual growth by a few tenths of a percentage point, raised volatility slightly and deepened the worst drawdown. Two caveats keep this comparison honest:

• The return gap is flattered. Our S&P 500 series reinvests dividends gross of withholding tax, while MSCI World is net. An Irish-domiciled S&P 500 ETF suffers 15% US withholding on dividends, so its return would be lower by roughly 0.3 percentage points a year, before fund charges.
• The full-period result hides two eras. The next chart shows them.

The lead came in long runs

Source: Copio calculation; calendar-year returns (CSV).

From 2002 to 2007 the S&P 500 trailed MSCI World every year, by up to 5.0 points in 2005, as non-US markets outpaced US stocks and a rising euro reduced the value of dollar returns for euro investors. From 2011, the US led in most years, by up to 9.6 points in 2014.

Measured from January 2001, the S&P 500 was behind MSCI World cumulatively from the end of 2003 to the end of 2012, and only moved back ahead at the end of 2013; an investor who added it in 2011 was rewarded much sooner. Neither outcome was knowable at the start, and that 15-year lead is a large part of why the US share of MSCI World rose from about half to almost three quarters.

Did the S&P 500 help in falls?

Not consistently. In MSCI World's 12 worst months since 2001, the S&P 500 fell by more in seven, one of them (February 2009) by less than 0.01 points, and by less in five. In October 2008 it lost 6.7% against 10.2% for MSCI World; in August 2001 it lost 10.4% against 8.3%. Over the whole period, its worst decline on month-end values was 54.6%, against 49.1% for MSCI World.

The two share a single risk: a broad fall in large-company equities. Adding the S&P 500 makes that risk more dependent on one country and its largest companies; it does not add a holding that rises when equities fall.

When adding the S&P 500 can still be a choice

There are reasons an investor may want more US exposure: a view on US companies, liabilities or spending in dollars, or a deliberate preference for the largest market. Those are concentration decisions, and they can be made with eyes open. The data only rules out one reason: holding both funds does not, by itself, diversify.

If the aim is to reduce the impact of an equity fall, the candidates are holdings with a different driver, and our falling-months study shows which ones held up and when they did not. If the aim is to reduce dependence on one country, the S&P 500 moves the other way.

Checks you can run on your own portfolio

1. Calculate your US share. Multiply each fund's weight by its US country weight from its latest factsheet and add them up. A World fund counts at about 73%.
2. Write down why you hold the S&P 500 fund. If the answer is diversification, compare it with a holding whose returns come from a different driver.
3. Check the overlap at the top. List the ten largest holdings of each equity fund. If the same names appear in all of them, the funds are one position in several wrappers.
4. Look at the 2002 to 2007 years. Ask whether you would have kept a US tilt through six years of trailing returns.
5. Count every US-heavy fund. A Nasdaq-100 or US technology fund adds to the same concentration.

Limits of this analysis

• Index returns exclude fund charges, and the S&P 500 series is gross of withholding tax while MSCI World is net. Correlations are barely affected; return comparisons favour the S&P 500 slightly.
• All figures are in euros. In dollars, the currency effects that shaped 2002 to 2007 disappear.
• Mixes are rebalanced monthly without costs or taxes.
• The US share is a snapshot at 30 October 2026 and changes with market values.
• Past relative performance does not predict which market will lead next.

How can copio help you examine your own allocation?

Copio is in private beta across Europe, with early access by application. It looks through your funds into asset classes and places each on a return-correlation map as a main driver, a true diversifier, dead weight or a shock absorber, so you can see whether your holdings differ by driver or only by label.

A correlation matrix shows which holdings move together, and a crash test shows how a mix like yours would have fared in past crises. Copio provides educational analysis, not personalised advice; it does not place trades or choose an allocation for you. The free portfolio checker is a first step.

This article is published for information and educational purposes only. It does not constitute personalised investment advice. Past performance is not a reliable indicator of future performance.