Before the financial crisis occurred in 2008, the market share of the main strategies to replicate the underlying indices used by ETF promoters in Europe was nearly evenly split between physical (full/optimized) and synthetic replication. That said, one needs to bear in mind that the vast majority of assets in the European ETF industry was invested in equity ETFs at this point in time. This changed in the aftermath of the financial crisis as European investors realized that a swap or other derivatives used for synthetic replication have a counterparty risk and, therefore, avoided ETFs which were using synthetic replication where possible.
Physical Replication is the Index Replication Method of Choice for European ETF Investors
As a result, physical replication has become the replication method of choice for European investors. Therefore, it is not surprising that ETFs using physical replication held the vast majority (87.02%) of assets under management in the European ETF industry at the end of December 2025.
Graph 1: Market Share of Replication Methodologies by Assets Under Management (December 31, 2021 – December 31, 2025)
Source: LSEG Lipper
In more detail, ETFs using full replication as replication methodology had the largest market share of AUM (€1,278.1 bn, or 49.88%), followed by ETFs using optimization strategies to replicate the risk/return profile of their underlying index (€951.7 bn, or 37.14%), ETFs using a synthetic (swap-based) approach (€302.0 bn, or 11.78%), and ETFs using other replication methods (€30.6 bn, or 1.19%).
Graph 1 shows that the market share of the assets under management for ETFs using full replication has increased (+0.81%) over the course of 2025. In contrast, the market share of AUM for ETFs using optimized replication (-1.02%) has decreased. In addition, the market share of AUM for ETFs using “other” replication techniques has increased (+0.49%) while the market share of AUM for ETFs using synthetic replication (-0.28%) has decreased over the same time period.
Since the movements in market shares are generally caused by the combination of estimated net flows and the market performance, these changes can’t be used to determine investor preferences.
In addition, one needs to bear in mind that the asset type of an ETF often impacts which methodology is used to replicate the respective underlying index. Full replication is the most preferred replication methodology for equity ETFs, while optimized replication is the preferred replication methodology for bond ETFs and synthetic replication is the only way to track blended commodities indices within UCITS ETFs.
More generally speaking, I would expect that the overall trend regarding the preference of physical (full/optimized) replication will continue even as structure terms of the swaps and other derivatives used for synthetic replication have been changed to minimize the counterparty risk within the respective products. Nevertheless, one needs to bear in mind that this replication method is needed to track asset classes or market segments which can’t be replicated with physical replication since the respective assets are not eligible for UCITS products, or if there are other investment restrictions, such as prohibited ownership by foreign investors, in place.
The views expressed are the views of the author, not necessarily those of LSEG.
This article is for information purposes only and does not constitute any investment advice
