Synthetic ETF replication is a way of building an exchange-traded fund in which the fund does not own the securities in its index. Instead, it enters into a total return swap with one or more banks, which agree to pay the fund the return of the index in exchange for the return on assets the fund holds or for cash. The fund's performance therefore depends on the swap, while its own assets serve as collateral or as a substitute portfolio. Funds built this way are also called swap-based ETFs.
Synthetic replication is the main alternative to physical replication. It became common in Europe in the 2000s, where many UCITS ETFs were launched by banks that could act as swap counterparty to their own funds. It is much less common in the United States, where most ETFs hold their securities directly. Synthetic structures are used most for indices that are difficult or costly to replicate physically, including some emerging market, commodity and strategy indices.
How a total return swap delivers the index
A total return swap is a swap in which one party pays the full return of a reference asset or index, including price changes and income, and the other pays an agreed return in exchange. In a synthetic ETF, the swap counterparty pays the fund the index return, net of an agreed spread, and the fund pays the return of its own assets or a floating rate. The counterparty usually hedges its obligation by holding the index securities or using futures, so the bank, not the fund, bears the trading costs of replication.
Two structures are used:
- Unfunded swap model. The fund uses investors' money to buy a portfolio of securities, often called the substitute basket, which may have little to do with the index. It swaps the basket's return for the index return. The fund owns the basket outright, and its exposure to the counterparty is limited to the swap's current mark-to-market value.
- Funded swap model. The fund pays cash to the counterparty and receives the index return, and the counterparty posts collateral to the fund, often held with a third-party custodian, to cover the amount it owes.
Counterparty risk and its limits
The central risk in a synthetic ETF is counterparty risk: if the swap counterparty defaults while it owes the fund money, the fund must fall back on its basket or collateral, which may be worth less than the index exposure investors expected. Several features are used to contain this risk:
- Exposure limits. Under the UCITS framework, a fund's net exposure to a single over-the-counter derivative counterparty is generally capped at 10% of net asset value where the counterparty is a credit institution, and 5% otherwise.
- Swap resets. When the swap's value owed to the fund grows, it is settled and reset, often at thresholds well below the regulatory cap.
- Collateral and haircuts. Collateral is valued daily, and many providers hold more than the amount owed, applying a haircut to riskier assets.
- Multiple counterparties. Some funds spread swaps across several banks to reduce dependence on one.
Concerns also centre on the quality of the substitute basket or collateral. If it consists of illiquid or correlated securities, its value could fall at the same moment the counterparty fails. Conflicts of interest can arise when the counterparty belongs to the same banking group as the fund sponsor and can place hard-to-fund securities in the basket.
Regulatory scrutiny
In 2011 the Financial Stability Board, the Bank for International Settlements and the IMF all published analyses questioning whether synthetic ETFs could transmit stress between banks and funds, with particular concern about collateral quality and transparency. The European Securities and Markets Authority then issued guidelines on ETFs and other UCITS issues in 2012, which required synthetic ETFs to disclose their counterparties, the composition of their collateral or substitute baskets and the swap arrangements. Investor demand moved strongly towards physical funds in the following years, and several large providers converted products from synthetic to physical replication.
Advantages
Synthetic replication can track an index more closely than a physical fund, because the counterparty guarantees the index return less a known spread, and the fund avoids trading costs, cash drag and the sampling error of a physical portfolio. The tracking error of synthetic funds is often very low. They can provide exposure to markets with foreign ownership restrictions or high transaction taxes, to commodity indices that cannot be held physically, and to strategy indices that trade frequently. For some European funds tracking US equity indices, swap structures have also allowed better after-tax results than physical holding, because of the way US dividend withholding tax applies to swap payments on qualifying indices.
Disadvantages
The main drawbacks are counterparty exposure, reliance on the quality of the collateral or basket, less transparency about what the fund economically owns, and swap spreads that are not always visible in the headline expense ratio. A synthetic fund cannot lend its index securities because it does not hold them, though it may earn other income on its basket. Investors comparing synthetic and physical funds usually look at total cost, tracking difference over several years and the fund's published counterparty and collateral reports.