SINGAPORE (Feb 13): Those who have been following this column will notice that I often make reference to investing in an entire stock market. As preposterous as it may initially seem to buy a whole stock market’s worth of shares, this can actually be very simply done through a class of financial products known as exchange-traded funds (ETFs).

At its simplest, an ETF pools money from its investors in order to buy every stock in the stock index that it is supposed to replicate. This stands in contrast with actively managed funds (also called unit trusts), where the fund manager does not buy the whole index, but only stocks that he thinks will do well. An active fund can do much better than the stock index, but it can also do much worse. In contrast, an ETF aims to return the same profits as the stock index, less any fees and expenses charged by the ETF provider.

In recent years, ETFs have gained popularity at the expense of active funds for one key reason: cost. Active equity funds often charge 1.5% per year in management fees, and about 1.7% to 2% once other expenses are included (although these figures vary widely). In contrast, the world’s largest ETF by assets, the SPDR S&P 500 ETF which tracks the US Standard & Poor’s 500 stock index, charges a miniscule 0.09% per year, including ex penses. An active fund manager’s US stock picks would thus have to do much better than the S&P 500 index to make up for the difference in fees and be more profitable for an investor than the ETF.

But it is dangerous to assume that ETFs always have a massive cost advantage and are therefore automatically the better choice compared with active funds. Here are three things you should be aware of.

ETF fees are not always that low

Numbers such as SPDR S&P 500 ETF’s 0.09% expense ratio are headline-grabbing, but there are lots of ETFs that charge much more. For example, the db x-trackers FTSE Vietnam ETF listed on the Singapore Exchange has an all-in fee of 0.85% per year based on its Dec 30, 2016 fact sheet. The iShares MSCI India Index ETF has an even higher management fee — 0.99% per year. SGX publishes the expense ratios of all Singapore-listed ETFs on its website. This is a useful resource that should be consulted and compared with the expense ratios of active alternatives before deciding which to invest in.

Tracking error often adds a further implicit “cost” to the ETF

For various technical reasons, many ETFs are unable to fully replicate the performance of their benchmark indices. As a result, investment performance may deviate significantly from the index. This deviation fluctuates over time and may be negative or positive, but it is important to understand the overall trend to see if there is a consistent underperformance that needs to be taken into account.

For example, the SGX- listed Lyxor MSCI AC Asia Pacific Ex Japan ETF aims to replicate the performance of the MSCI AC Asia Pacific Ex Japan stock index. Since the total expense ratio from the fact sheet is 0.6%, one should expect the ETF to underperform the index by 0.6% per year. The table shows the actual yearly performance, and the corresponding five-year average based on the November 2016 fact sheet.

As the table demonstrates, the five-year index return was 4.94% per year, but the five-year ETF return was only 4.03% per year. The average annual underperformance (known as tracking difference), therefore, was 0.91% — 1.5 times what it should be. Moreover, the tracking difference is consistently greater than 0.6% every year, suggesting that this is an effect that might persist. At the least, it is something that requires further investigation with the ETF provider. If we expect the deviation from the index to remain negative in the future, we should consider the cost of the ETF to be closer to 0.91% than 0.6%.

Overall, the average tracking difference is likely to be a better predictor of ETF cost than just the expense ratio alone. This matters because a high tracking difference narrows the ETF cost advantage compared with active alternatives.

The active fund will almost always still have higher costs. But if the margin of difference is small and the active fund manager has a good track record of consistently outperforming the benchmark index even after its relatively higher costs are taken into account, then it might make more sense to choose the active fund. At the highest level, the question to ask is: Compared with the baseline expected underperformance of the ETF relative to the index, is the active fund manager expected to do better or worse?

Overseas ETFs may have tax implications that reduce returns

It is possible to invest in ETFs in overseas stock exchanges. The US, in particular, offers an unparalleled range of ETFs tracking almost any asset class and geography you can think of, and costs are often cheaper owing to competition and economies of scale. Unfortunately, the US applies a withholding tax of 30% on dividends for non-US investors. If the dividend yield for an ETF is 3% per year, then 30% tax on 3% is a 0.9% yearly cost on top of the ETF’s tracking difference. That can quickly make an overseas ETF uncompetitive.

All this is not to say that ETFs are bad; they are a welcome financial innovation and in many cases, an ETF is indeed the most efficient solution for gaining access to a particular market or geography. But a more nuanced and detailed analysis is needed before jumping to that conclusion. As with any investment, be sure to do your research and evaluate alternatives first, and enter with your eyes wide open.

Herbert Lian is an independent financial adviser at IPP Financial Advisers. The views expressed here are solely his own. This article should not be regarded as professional in vestment advice or a recommendation regarding any particular investment.

This article appears in Issue 766 (Feb 13) of The Edge Singapore which is on sale now