Replicating an index without taking the full market impact

Replicating an index without buying every single security in it: the multifactor approach reduces market frictions and recurring costs.

Replicating the performance of an index is often presented as a straightforward mechanical exercise: buy the same securities, at the same weights, and let the market do the rest. The day-to-day reality of managing optimized index portfolios tells a different story. Thirty years of optimized replication in the Swiss market — the method dates back to 1995 — provide enough perspective to distinguish what belongs to theory from what is actually achieved once market frictions, transaction costs, and operational constraints are taken into account.

There are fundamentally three ways to reproduce the performance of an index. The first, full replication, also called “pure,” perfect, or near-perfect replication, consists of holding all, or almost all, of the securities in the index at their exact weights. Synthetic replication, by contrast, relies on the use of derivatives, which introduces counterparty risk. The optimized replication method selects a subset of the index’s securities in order to build a portfolio whose risk profile — rather than its nominal composition — is identical to that of the benchmark. In broad universes comprising several thousand positions, this latter approach allows for efficient implementation that is operationally realistic.

The optimized approach uses a multifactor market model based on the financial theory of APT (Arbitrage Pricing Theory). The principle is to identify the common forces — the risk factors — that explain most price movements, and then to build a portfolio exposed to these factors with the same sensitivities as the index. The model is updated monthly. The replication objective is not to outperform the index based on active conviction, but to achieve zero tracking difference, or a slightly positive one, after fees.

Why not simply buy every security? Because pure replication requires trading in less liquid names, and this is where costs arise: in the Swiss market, for an index portfolio of CHF 300 million, the price impact of transactions can reach 35 basis points per year. Optimized replication avoids this friction by concentrating transactions in a limited number of liquid positions that capture the index’s dominant risk factors.

Beyond the initial market impact, recurring operations — reinvestment of dividends and coupons, periodic index reviews, subscriptions and redemptions — generate a continuous flow of transactions. In broad indices, each operation affects dozens or even hundreds of positions. By reducing the number of holdings and optimizing the transactions required, optimized replication structurally lowers these recurring costs. This is an advantage that does not show up on a three-month chart, but becomes visible in net-of-fees cumulative performance over five or ten years. Experience shows that over a ten-year period, fund management using this optimized indexing method makes it possible to offset part of the fees in Swiss markets, and to generate an average annual net-of-fees performance of 0.5% in international markets.

Two principles also underpin the entire method. First, cost transparency: the fees charged to investors reflect the actual cost of the transactions that need to be carried out on their behalf, while minimizing unnecessary replication activity and without flat fees. Second, the stability of the approach: the optimization model currently in use has remained the same for thirty years. These thirty years of practice in Swiss and international markets confirm over time that optimized index replication, far from being an artificial device, remains a robust and relevant tool for tracking an index while minimizing costs and historically delivering consistent performance.

Thierry Vessereau joined BCGE Asset Management in July 2019, then Synchrony Asset Management in September 2026, as Institutional Portfolio Manager for the passive and active funds in the Synchrony range. Previously, he worked at Pictet & Cie from 2005 to 2019 as Project and Development Manager, then as Investment Specialist in the Indexation team at Pictet Asset Management. Thierry Vessereau holds a Master’s degree in Computer Science and a PhD in Economics.