ETF tracking errors vary widely

Investors may not be able to work out how well their exchange-traded funds (ETFs) are performing, according to research into variations in their tracking errors. Daniella Acker, professor of finance and accounting, and Nigel W Duck at the University of Bristol studied the tracking errors of ETFs and unit trusts and found that returns could vary by as much as 5 per cent, depending on the day of the month on which underlying monthly returns are based. Their research follows a report in the Financial Times showing that funds differ in how good they are at tracking indices, with those in more illiquid markets more likely to have large tracking errors.

Separate research from Bestinvest found that the FTSE All Share had risen 50 per cent in the two years to the end of March but the Blackrock UK Equity Tracker had only risen 28 per cent after charges were taken out.

Acker and Duck’s figures also found that funds could appear as the best or worst performer, depending on the day selected. Monthly tracking errors for the Prudential UK Index Tracker Trust, for example, varied from 5.8 per cent to 1.5 per cent on an annual basis. The HSBC FTSE 100 Index Retail fund varied from 5.9 per cent to 1.4 per cent, while the Source FTSE 100 ETF varied from 3.1 per cent to 0.2 per cent.

“A lot of newer assets like ETFs don’t have a long returns history – and the longer you look, the less representative it is of the present,” explains Acker.

“Even if you calculate things over five years, it’s still a problem. You need so many years of data for it to become stable that it essentially becomes meaningless.”

ETFs may not track specific indices perfectly as some buy and sell a selection of stocks to replicate the index, which may not match the performance of the index as a whole.

Other “swap-based” ETFs tend to have lower tracking errors as they achieve their return through a swap contract with an investment bank. However, these funds have additional counterparty risk and may hold assets that are unrelated to the index they track.

Acker tests each day of the month when analysing fund performance, running her data 30 times to check she has not chosen a day with an unusual spike.

She says ordinary investors should not ignore tracking errors altogether – but that it is important to look at other information as well. “When factsheets quote beta based on three years’ data, I really don’t take any notice at all,” she says. “Performance data is only historic and that, as we know, is no indication of the future.”

The research comes as financial regulators around the world issue warnings on the risks of ETFs. The Financial Stability Board warned in April that investors might not understand the products and that many contained risks that were not explained in factsheets – such as counterparty risk and illiquidity.

Investors are now being urged by wealth managers to make sure they understand exactly how an ETF achieves its return – and to check they have a range of counterparties backing the ETFs they hold in their portfolios.