Synthetic vs. Physical ETF Replication

See the two different ways an ETF can actually deliver an index's return.

Physical replication

Physical replication is the traditional ETF method of directly buying the actual securities in an index. Full replication buys every constituent at its index weight, while optimized replication (sampling) buys a representative subset when an index has too many holdings or includes illiquid securities to fully replicate cost-effectively.

Synthetic replication

Synthetic replication doesn't hold the underlying securities directly. Instead, the fund manager enters into a swap agreement with a counterparty, typically a large investment bank, who agrees to pay the fund the index's return in exchange for a fee. This is often used to access markets or indexes that are difficult or costly to hold physically.

Counterparty risk

A physically replicated ETF holds real assets that retain value regardless of the fund manager's own financial health, but a synthetic ETF is exposed to the credit risk of its swap counterparty. Because of this, most synthetic ETFs post collateral and cap collateral exposure under regulatory limits to manage the risk.

Where synthetic replication is common

Synthetic replication tends to be used for markets that are hard to access directly, such as emerging markets with foreign-investment restrictions, commodities that are impractical to store physically, or niche indexes with illiquid constituents. Europe's UCITS ETF market is known for having a relatively higher share of synthetic ETFs for these reasons.

Physical replication still dominates in many markets

In many domestic ETF markets, physical (or optimized) replication remains the dominant structure, with synthetic ETFs a smaller share overall, though some ETFs tracking hard-to-access commodities or foreign indexes do use swaps or other derivatives partially. The exact method always varies by product, so check the fund's prospectus.

How to check your ETF's replication method

You can find out whether an ETF uses physical or synthetic replication, and how much it relies on derivatives, in the fund's prospectus or monthly fact sheet, typically under a section describing the fund's structure or strategy. Checking this before buying is worthwhile, especially for ETFs tracking less conventional markets.

Replication method isn't a verdict on quality; it's a structural trade-off

Neither replication method is inherently 'better': physical replication carries tracking error from timing differences in reinvesting dividends and interest, while synthetic replication carries counterparty risk from its swap structure. Each is a different way of achieving the same goal, matching an index's return, with its own structural risk to understand.

For general education only, not investment advice

This page is a general educational explanation of ETF replication methods and does not recommend any specific ETF. Replication structures, collateral requirements, and disclosure practices vary by fund, issuer, and jurisdiction and can change, so confirm the current prospectus before investing.

Frequently Asked Questions

Are synthetic ETFs automatically risky products?

Not automatically. Most synthetic ETFs post collateral and limit collateral exposure under regulation specifically to manage counterparty risk. That said, they carry a structural risk factor that physically replicated ETFs do not have, which is worth understanding.

Would an ordinary investor ever encounter a synthetic ETF?

Yes; while many domestically listed ETFs use physical replication, investing directly in certain overseas-listed ETFs (especially European UCITS funds) or in ETFs tracking niche commodities or hard-to-access indexes can mean holding a synthetically replicated fund.