The European ETF market has just posted a record quarter. According to Amundi, EUR 127.4 billion of net new money flowed into European ETFs in the third quarter of 2026, up from EUR 113.6 billion in the second quarter and EUR 89.1 billion a year earlier, taking total assets to EUR 3.2 trillion at the end of September. The word the asset manager uses to describe the move is "diversification": more world ETFs, a return to Europe, a preference for government bonds and, above all, an unprecedented rush into gold.
That is good news for self-directed investors. It is also the right moment to ask an uncomfortable question: when you hold five ETFs, are you five times more diversified, or have you bought the same thing five times?
What the Q3 inflows say about investor behaviour
Amundi's figures, published on 13 October, describe an investor who is cautious but fully invested. Of the EUR 99.3 billion that went into equity ETFs, EUR 29 billion went to all-country world ETFs (ACWI-type), EUR 25.5 billion to developed-world ETFs (MSCI World-type) and EUR 25.5 billion to the United States. Europe took in EUR 9.9 billion after outflows the previous quarter. Emerging markets received just under EUR 6.8 billion.
On the bond side, EUR 26.4 billion came in, of which EUR 11.9 billion went to government bonds, while euro high yield collapsed to EUR 564 million from EUR 3.2 billion the quarter before. In other words, investors want carry and liquidity, not credit risk.
The most striking number concerns gold: more than EUR 10 billion flowed into gold-backed ETCs and ETPs in the third quarter alone, and roughly EUR 10.4 billion since January. The previous annual record, EUR 7.6 billion in 2022, has already been passed.
Morningstar added another layer on 14 October: European active ETFs gathered EUR 28.9 billion in nine months, more than in the whole of 2025 (EUR 26.8 billion), for EUR 117.1 billion in assets. They still account for only 3.4% of European ETF assets, against 13.1% in the United States, but 118 products have been launched since January and there are now 377 of them.
Nothing in these flows is a buy or sell signal. What they show is that a growing number of investors are stacking different building blocks with the idea of spreading risk. The open question is whether those blocks are really different.
The overlap problem: when "world" and "developed" are almost the same ETF
Take a portfolio you meet often among European investors: an MSCI World ETF in a brokerage account with one broker, an MSCI ACWI ETF in a pension or tax-advantaged account with another, an S&P 500 ETF bought "to add some tech", a Europe ETF "to rebalance", and a gold ETC "for protection".
On paper, five lines, five ideas. In practice, three of those lines share most of their constituents.
The MSCI World carried roughly 70% US equities in August 2026, according to figures compiled by the Daily Maverick from MSCI factsheets, up from under 50% in 2006. The MSCI ACWI, which adds emerging markets, is still majority American as well. An S&P 500 ETF is, by construction, entirely American. The ten largest stocks in the S&P 500 account for nearly 40% of that index, and they are exactly the same companies that sit at the top of the MSCI World and the MSCI ACWI.
The result: an investor who thought they held "the world", "the world plus emerging markets" and "the United States" mostly holds the same handful of large US technology companies three times over, with slightly different weights. The Europe ETF does add something different, but its weight in the portfolio is often too small to move the overall exposure.
This phenomenon has a name, overlap, and it has three practical consequences.
1. Your real geographic exposure is not what you think
Adding up the percentages shown on each ETF's page is not enough. You have to recompute, line by line, each country's share of the total portfolio, weighted by the size of each ETF. A "diversified" portfolio can easily end up 65% to 75% exposed to the United States without its owner ever having decided that.
2. Your sector exposure is concentrated
The same exercise by sector often gives an even starker result: technology and communication services, which dominate the major global indices, end up overweight in every equity ETF, and therefore in the total.
3. Fees stack up on redundant exposures
Paying management fees on three ETFs that hold largely the same securities means paying three times for a single exposure. ETF fees are low, but they are not zero, and the differences between products show up over ten years.
How to check for yourself
The method is simple to describe and tedious to carry out by hand.
First, obtain the composition of each ETF (factsheets and holdings files are published by the issuers, usually monthly). Second, multiply each weight by the ETF's share of your portfolio. Third, aggregate by country, by sector and by currency, and look at the result.
Three questions to ask when the table is in front of you:
- What share of my portfolio is really exposed to the United States, once every ETF is added up?
- How many of my lines hold the same top ten stocks?
- What is my foreign-currency share, once unhedged ETFs are taken into account?
An active ETF adds a complication: its composition changes faster than an index ETF's, and two thirds of the flows into European active ETFs now go to "systematic" strategies, meaning quantitative models that depart from the index in a regular but not always legible way. For these products, checking the composition once a quarter is a minimum.
Gold: measuring its true place in your wealth
The record inflows into gold call for a particular kind of check, because "gold" covers very different products.
An ETC backed by physical gold tracks the metal's price, with counterparty risk limited to the product's structure. An ETF of mining companies is an equity line, exposed to company earnings, extraction costs and debt, and it can fall while gold rises. A synthetic product replicates the price through derivatives and introduces a different counterparty risk. These three families react differently to a market shock, and putting them in the same "gold" box distorts the picture.
There is also a question of measurement. Many investors count their gold ETC as a percentage of their financial portfolio but forget the gold they hold elsewhere: coins, bars, or indirect exposure through a multi-asset fund that already owns some. The real share of gold in total wealth is often different from the one shown on the broker's screen.
Finally, currency. Most gold ETCs are quoted in dollars or track a dollar price. For an investor whose home currency is the euro, the Swiss franc or the pound, the performance on screen blends the move in gold with the move in the exchange rate. Knowing how much of the change comes from each is part of proper tracking.
What the 2026 record says is that European investors are looking for an uncorrelated asset. What it does not say is how much gold anyone should hold: that decision belongs to each investor, and inflow figures are not an argument.
Where Tukhe fits in
The calculation described above is exactly what a portfolio tracker should do for you, without your data ever leaving your computer.
Today, Tukhe consolidates positions held with several brokers into a single portfolio, computed and stored on your own machine. Breaking ETFs down into their constituents, and the consolidated exposure by country, sector and currency that follows from it, are part of the analytics module currently in development. The goal is easy to state: an MSCI World ETF, an ACWI ETF and an S&P 500 ETF should no longer appear as three independent lines but as what they are, a cumulative US exposure with the same names at the top, and the share of each asset class, gold included, should be computed on total wealth rather than account by account.
What will not change is the principle: everything happens locally. ETF composition files are public data; your positions stay on your machine. There will be no intermediary server that "sees" your portfolio in order to tell you it is concentrated.
Diversifying means verifying
The third-quarter inflow record shows that European investors have understood the point of spreading risk. The next step is to check that the spread is real: look at what the ETFs contain, not only at what their names promise.
Five well-chosen ETFs can make a genuinely diversified portfolio. Five ETFs chosen for their names can amount to a single position in disguise. The difference only shows when you do the maths.
Tukhe is a local-first portfolio tracking application built for European investors who want to keep control of their data. This article is for educational purposes and does not constitute investment advice.
