Investors warm to synthetic ETFs

But their tax and tracking error benefits may be slipping away for some

|

Elena Johansson

“In the first nine months of 2020, we’ve seen $1.5bn (€1.28bn) of net inflows,” says Gary Buxton, head of Emea ETFs and indexed strategies at Invesco. “While the rest of the synthetic market has seen net outflows.”

As of 3 November, the Invesco S&P 500 Ucits ETF has $10.2bn in assets under management and is the largest synthetic ETF in Europe, according to the firm.

The US asset manager says that its synthetically replicated Ucits ETFs have each outperformed the average of their largest physical competitors by 0.24% (S&P 500), 0.31% (MSCI USA) and 0.12% (MSCI World), respectively, over the 12 months to the end of August.

Meanwhile, BlackRock, which previously denounced derivatives-based ETFs, recently made a U-turn.

It launched a synthetic version of its S&P 500 ETF at the end of September, the iShares S&P 500 Swap Ucits ETF.

Cutting costs

Jose Garcia Zarate, associate director, passive strategies, manager research, Europe at Morningstar, explains to Expert Investor that, in the specific case of US large cap equity markets, synthetic ETFs outperform physical ETFs because of their tax advantage.

Under current regulations, swap-based ETFs enjoy 100% of the dividend whereas their physical counterparts have to pay a withholding tax, according to Zarate.

In the case of Europe, Buxton says that synthetic ETFs tracking global and European equity benchmarks can also benefit from a cost reduction.

“For instance, an ETF that physically replicates the FTSE 100 index will have to pay UK stamp duty on the shares it purchases, whereas a synthetic ETF doesn’t [because it doesn’t purchase the shares in the index]. A similar situation arises when replicating a European index, with synthetic ETFs not subject to financial transaction tax,” he explains.

But this could change in future, as a financial transaction tax is currently being discussed at EU level to help finance the planned covid recovery package and boost the economy.

Tracking error 

A key benefit of synthetic ETFs has been their low tracking error, but observers say that this is diminishing.

Zarate explains that the advantage varies depending on the market and asset class: “For example, in the case of developed equity markets you should expect the difference in tracking error between physical and synthetic ETFs not to be meaningful, whereas it should increase as we move into more illiquid or hard-to-reach areas such as emerging markets or some of the least liquid areas of fixed income.”

Synthetic ETFs are able to provide exposure to areas where physical replication is more challenging, but Zarate says that “these hard to reach areas are shrinking as, for example, more developing countries open up their equity and bond markets to international investors, eg China is a clear case at hand”.

Sidi Kleefeld-Von-Wuestenhoff, head of ETF advisory sales at DWS, tells Expert Investor; in certain markets, where the underlying securities are not readily accessible as physical instruments, swap-based structures can be beneficial.

“[But] tracking difference or tracking error is not correlated so much to whether an ETF is swap-based or physical, but instead to how well the product is managed overall. A well-managed physical ETF will track its index very closely,” he says.

At the end of August, synthetic ETFs accounted for only 14.05% of the overall assets under management in the European ETF industry, according to Refinitiv data.

Detlef Glow, head of Emea research at Refinitiv Lipper, recently wrote that synthetic index replication was widely used in the European ETF industry up until 2011.

“But in the aftermath of the financial crisis and the euro crisis, critics argued that ETFs that use a swap to replicate their underlying indices add another layer of risk to their portfolio,” Glow stated.

The products also raised concerns with regulators, such as the Financial Stability Board, over counterparty and liquidity risks.

Risk factors

To hedge swap counterparty risks, “most providers of synthetic ETFs have moved from a single counterparty to a multiple swap counterparty model”, Zarate says.

Another risk factor is the substitute baskets – the basket of securities whose return is swapped by the return of the index.

Zarate explains that, since the financial crisis, there have been substantial improvements in the quality of substitute baskets, “but it’s always a good idea to monitor the contents”.

Kleefeld-Von-Wuestenhoff claims that “it is not the case that there will be higher defaulting risk in volatile markets”, as no swap-based ETF has stopped operations this year – despite extreme levels of volatility.

“Also, the Ucits regulations ensure that all Ucits ETFs meet very high operational standards, including when it comes to aspects such as counterparty risks relating to index swap arrangements,” he notes.

And even if a swap counterparty defaulted, collateral arrangements are in place and the ETF would switch to another swap provider to resume tracking, he adds.

The European Central Bank highlighted, however, in a 2018 article on counterparty and liquidity risks in exchange-traded funds: “In the extreme case of a counterparty default, while ETFs can fall back on collateral assets, investors would face risks associated with the collateral.

“Both synthetic ETFs and ETFs offering securities lending are typically over-collateralised. Collateral baskets often consist of liquid stocks and bonds. However, frictions may arise when dealing with collateral from defaulting counterparties.

“To obtain the original exposure, the ETF issuer might have to sell the received collateral in falling markets given that counterparties are more likely to default when markets are stressed. This may be particularly problematic when collateral exposures differ substantially from the exposure expected by investors […].”

According to Buxton the biggest risk for synthetic ETFs today is arguably single counterparty models.

“The [global financial] crisis highlighted the risks of synthetic models being used by ETF issuers at the time: typically owned by a big bank, which would be the sole counterparty writing the swaps.

“We’ve always had at least four counterparties, which greatly reduces the impact on the fund if any individual counterparty were to default. We take other steps to reduce risks, including only accepting high-quality securities into the basket and resetting the swaps [to zero] whenever certain conditions are triggered; this is often daily, which reduces risk compared to less frequent resets.”

When it comes to choosing between either synthetic or physical replication, Zarate says that it “is all down to personal preferences”.

There has been renewed interest in synthetics for US equity exposure in the last few years, mainly among institutional investors who can monitor risk in all its variants, but, as a whole, physical replication remains the king, he concludes.

MORE ARTICLES ON