Monday, 23 September 2013

Emerging Markets ETFs Comparison Update - Which is best?

It has been almost three years since we first compared the main broadly-diversified ETFs for US-traded and Canadian-traded Emerging Markets equities. Things have changed with some new entrants and evolution of existing options. Emerging markets like China, Russia Taiwan, Brazil and South Korea are ever more important economic players. It's time for an update.

At the same time we'll integrate information on the pros and cons of buying ETFs on US markets, especially the effect of foreign withholding taxes, that we wrote about here. We'll focus on broadly based ETFs that aim to diversify across all emerging markets countries and all industry sectors and pick the biggest ones amongst the many listed in the USA - see the scores of offerings in the ETF Database list of emerging market ETFs here.

Two sets of ETF choices ...
The largest divide in the options is between traditional passive, cap-weighted index portfolios and those based on a variety of alternative strategies that have been gaining popularity - such as fundamental accounting data weighting, high dividend payout and low portfolio volatility.

1) Traditional Cap-Weighted Index ETFs
Some are traded in US dollars on US stock exchanges, others in Canada in Canadian dollars, though they all ultimately hold equities of emerging markets countries. As our comparison table below shows, the asset base of US ETFs dwarf those of Canadian ETFs.

 2) Alternative Strategy ETFs
There is also a mix of Canadian and US-traded ETFs in this group.

Costs - MER and Bid-Ask Trading Spread
Management Expense Ratios (MER) are very important because they come right off the bottom line of returns to the investor. Since they get charged every year over the long term they can seriously undermine returns. In other words, the lower MER the better. SCHE at 0.15% is the lowest/best of all, followed closely by VWO and IEMG at 0.18%, with EEMV the best amongst the alternative strategies at 0.25%. Unfortunately, the best Canadian ETF VEE is a fair bit higher at an MER of 0.37%.

The gigantic size of the US ETFs ensures plenty of trading volume, which in turn means the spread between bid and ask prices is always very tight. That is a cost to the investor but is a relatively minor factor for long term buy and hold investors.

Cash Yield - a large range and effects of MER
It isn't surprising to see DEM with highest percentage of cash distributed to shareholders over the past twelve months since that is the ETF's strategy. However, it is surprising how little ahead - 3.9% vs 3.7% - it is over VWO, which pays no attention to this factor. It is also surprising to see the big spread of results across the ETFs, even those that use the same cap-weight approach.

All the Canadian clone ETFs have a significantly lower yield than thier respective US holding, illustrating that the higher MER needs to be paid from somewhere. PXH, which weights in part according to dividends, also loses ground from its higher MER.

Sharpe Ratio and Volatility - foreign currency effects have helped recently
Sharpe Ratio (see Investopedia explanation) is a measure of return per unit of risk/ volatility. The higher it is, the better - more bang for the buck risked in effect. XMM with its much lower volatility and higher return wins handily on this measure.

The reason XMM achieved a higher return than its US holding EEMV, is that a falling Canadian dollar over the past year has boosted the value of EEMV within it. Since our table shows returns in the local currency of trading, EEMV's returns are lower in terms of US dollars. But a Canadian holding EEMV would see a higher return once the value of EEMV was translated into Canadian dollars. We discussed the beneficial effect of foreign currency in this long ago post and illustrated how it would have affected a diversified portfolio in this post. Of course, the effect can go the other way and hurt returns if our dollar strengthens against the currencies of emerging countries. The yellow-outlined cells in our table show for one clone ETF pair where foreign currency effects are influencing the differential USA vs Canada ETF figures.

Attractive prices? Weak returns but valuation ratios vary
The trailing one- and three-returns for all of our ETFs look quite low, which would normally make one think a turn-around is in the offing. But valuation ratios of price to earnings at 17+ and price to book value are not that low for most of the ETFs. PXH, DEM, VWO and SCHE are the exceptions with attractive ratios. Some people believe some markets like India are good buy at the moment. Long term investors with a defined asset allocation will know when their portfolio rebalancing threshold is breached.

Foreign withholding taxes - beware which account to hold which ETF
We have copied the results of the discussion in this post about the subtle and complicated effects of foreign withholding taxes on a Canadian investor, depending on which account the various types of ETF are held. Ironically, many ETFs traded in Canada are worse for a Canadian investor! In particular, the clone ETFs which only hold a US ETF get dinged by US 15% withholding taxes that cannot be recovered or avoided when the ETF is in any type of registered account. TFSAs and RESPs are not good places from a withholding tax perspective to hold any emerging market ETF.

Portfolio holdings differ markedly - mind South Korea, Russia, India and sector weights
FTSE does not consider South Korea to be an emerging market country while MSCI does. Depending on the combination of developed market ETFs in a portfolio, an investor might end up having South Korea doubled up or worse, absent, which would defeat the aim of worldwide diversification considering the country is such a large economy. We wrote a guide to figuring out ETF combinations for country coverage when Vanguard decided to change from MSCI to FTSE as its index. In the comparison tables below blue highlighted text indicates where an ETF departs a lot from the weights amongst the group.

Similarly, Russia and India are also major economies, but due to their respective index rules, most of the ETFs have little or none of one or both countries represented. VWO, PXH and CWO are the exceptions with both India and Russia occupying major places, which we believe make them better for doing so. However, PXH and CWO are very concentrated in the six biggest countries. All the market cap ETFs are much less concentrated in terms of top ten company or top six country exposure. Overall VWO wins on balance for portfolio characteristics.

Other cost factors for the US vs Canadian ETF decision
There are other important cost factors that are not incorporated in the table e.g. the cost for a Canadian investor convert Canadian into US dollars or vice versa, which may cost up to 1.5% depending on the broker; the size and frequency of contributions or rebalancing, the level of yield interacting with withholding taxes, ETF tracking error. Canadian Couch Potato developed a spreadsheet to enable investors to crunch numbers with various scenarios but concluded that long term buy and hold investors are better off with US ETFs. We built a similar spreadsheet tool to calculate the cost of US vs Canadian ETFs and we found that the best choice might vary with assumptions.

Bottom LineVWO is the default overall emerging market ETF of choice at the moment due to its combination of low MER, low trading (bid-ask spread) costs, tax characteristics for Canadian accounts, good yield, attractive P/E and P/B valuation and portfolio diversification.

Disclaimer: this post is my opinion only and should not be construed as investment advice. Readers should be aware that the above comparisons are not an investment recommendation. They rest on other sources, whose accuracy is not guaranteed and the article may not interpret such results correctly. Do your homework before making any decisions and consider consulting a professional advisor.


Be'en said...

I find your blogs to have a lot of value. Like the presentation which is always to the point and very well researched and analysed.

Do you have a similar analysis of ETFs of developed countries outside of North America? I am looking to bye one in an appropriate account.

CanadianInvestor said...

Good question Be'en! Looking back through my posts I realize I've never done such a review despite the fact that such an ETF should be a core part of any investor's portfolio. I'll do one next week.