Synthetic replication

Synthetic replication was first introduced in Europe in 2001. Synthetic replication is done through a type of exchange traded fund (ETF). An important attribute of this specific type of fund is that it does not hold any underlying securities featured on its benchmark. Instead of holding these securities synthetic ETF’s use derivatives such as swaps to track the underlying index in the process of replication. In replication of these synthetic accounts the return is 100% tied to the ETF it represents. Therefore, the ETF “replicates” the fund’s it is tied to performance. In this process the ETF manager enters a swap contract with an investment bank that agrees to pay the index return in exchange for a small fee.

The biggest argued benefit of synthetic ETFS is that they seem to do a more accurate job of tracking indexes, and when used in full replication can allow for less risk/higher return investments. Those that argue against synthetic replication says that it adds counterparty risk, is not fully transparent, and could mislead less experienced investors.

For instance Black Scholes theory claims vanilla option pricing can be achieved through the use of and zero-coupon bond. A simple example would be if you went to a bank and purchased a synthetic ETF for the purpose of replication. That fund may represent a certain group of stocks in Apple. However, the synthetic ETF is in no way physically attached to Apples stock. The bank you purchased the fund from would largely mirror Apples stock performance 100% through the use of derivatives and swaps in the process of replication and would charge you a fee for it. This is an extremely simple example but does describe the base process of synthetic replication.

...
Wikipedia