Dual Momentum: Evaluating the International Stocks Component

This post hopes to address a frequent line of questions I get about the Dual Momentum signal I share monthly on this blog.

  • Why did I just post an MSCI EAFE Index ETF as my international signal?
  • Is it a big mistake to invest in the international developed stock market without emerging markets?
  • What about Canadian stocks which are not in the MSCI EAFE Index?

Some of the questions are addressed in an Update on Dual Momentum post. As stated there, I am changing the format and signals of the Dual Momentum update I share each month.

The International Stock Component

We can break the International ex-U.S. equities markets into two broad groups: developed markets and emerging markets. There are also frontier markets, but these are often not investable for various reasons.

The developed markets make approximately three-quarters of the international ex-U.S. market cap. Stock markets include most of western Europe, Japan, Australia, New Zealand, and Canada.

The most popular developed market index is the MSCI EAFE Index which does not include Canada or South Korea. The alternative developed index is the FTSE Developed ex-U.S. Index which does include Canada and South Korea.

Depending on the index used, emerging markets currently form approximately 20 to 25 percent of the global ex-U.S. markets. This is set to grow rapidly in the future as the second largest market in the world—the Chinese domestic market—opens up to foreign investors.

The emerging markets includes 24 developing economies on every continent with reasonably investable financial markets. The most popular indices are the MSCI Emerging Markets Index and the FTSE Emerging Index.

There are slight variations in the MSCI and FTSE indices. The big ones are the inclusion of Canada and South Korea in developed markets, and the degree of inclusion for the Chinese domestic market (A-shares) in the emerging markets.

In the total accessible stock market space right now, U.S. stocks form about one-half of the global market cap and the rest of the world makes up the remaining half.

Do Emerging Markets Matter to Dual Momentum

As stated above, for 2018, I used the MSCI EAFE Index to measure International equities in my TADM model. One of the questions I get asked is if emerging markets should be included.

Before we jump to any conclusions about this, we should do a proper evaluation on precisely how much of an impact emerging markets have on our Dual Momentum model.

For this backtest, we used the MSCI USA Index, the MSCI World ex-USA Index, the MSCI ACWI ex-USA Index, the FRED 3-month Treasury Bill data for cash, and the NYU calculated 10 Year Treasury Bond data with for our out-of-market investment.

Emerging markets have only been tracked since 1987, so we will look at performance from that time forward.

Sources: TheRichMoose.com, MSCI Inc., FRED Federal Reserve St. Louis

Based on the graphs, emerging markets do matter. But maybe not as much as you think.

From 1987 to 2017, the backtest which included emerging markets generated a 14.76 percent compound annual return. This compares to a 14.08 percent compound annual return for the backtest using developed markets only.

The advantage was negligible for the first few decades, but in the mid-2000s it significantly tilted towards emerging markets.

We don't know the future, but a potential 0.7 percent annual advantage adds up as time works its magic.

Given the information we know with current data, you should invest in a global market ETF which includes emerging markets when the signal is in International stocks if that is feasible.

In Canada, this leaves us with two viable options:

  1. We could convert our accounts to U.S. dollars and invest in U.S. listed ETFs where several low-cost global ex-U.S. ETFs exist; or
  2. We could invest 75 percent of our portfolio in a developed market ETF and 25 percent in an emerging market ETF.

Both of the options add further complication to our investing process. It does not make sense for all Canadian investors to take these extra steps.

If you have more than $10,000 per account and you invest with a brokerage that charges no fees on many ETF transactions like Questrade or National Bank Direct, you should consider buying a developed market ETF and an emerging market ETF as in option 2 above.

If you have an account balance of at least $50,000 per account, you could consider switching to a U.S. dollar account and investing in U.S.-listed ETFs for all of your Dual Momentum holdings.

You can keep costs low by using Norbert's gambit with DLR/DLR.U, or Interactive Brokers to change currencies. Remember, not all brokerages offer U.S. dollar accounts, particularly for registered accounts.

Does Canada Matter to Dual Momentum

Since in the TADM model I have shared an ETF which tracks the MSCI EAFE Index (which does not include Canada), another common question I get relates to the impact of not having Canadian stocks in the Dual Momentum model.

Before we jump to any conclusions about this, we should do a proper evaluation on precisely how much of an impact Canadian markets have on our Dual Momentum model.

For this backtest, we used the MSCI USA Index, the MSCI All Country World ex-USA Index, the MSCI EAFE+EM Index, the FRED 3-month Treasury Bill data for cash, and the NYU calculated 10 Year Treasury Bond data for our out-of-market investment.

The only difference between the MSCI ACWI ex-USA Index and the MSCI EAFE+EM Index is the latter does not include Canada.

You will notice we have to run the test from 1987 forwards given that emerging markets were included at that time and we need to get an effective comparison test of Canada's impact in our broad investment options today.

Sources: TheRichMoose.com, MSCI Inc., FRED Federal Reserve St. Louis

When isolated as best as I can, the impact of Canada is very real.

The return of the MSCI ACWI model, which includes Canada, was 14.76 percent compounded annually over the 30 year period.

If we just used the MSCI EAFE+EM index, the return was 14.19 percent compounded annually.

We could say that inclusion of Canadian stocks translates to a 0.57 percent annual return advantage over the entire period.

However, until the mid-2000s there was no real difference. As with emerging markets, the impact of Canada is strongly related to the natural resource boom.

We do have a lower cost ETF on the Canadian exchange which tracks a developed market index that includes Canada. It is the Vanguard FTSE Developed All Cap ex-U.S. Index ETF (VDU.TO).

The problem with this VDU.TO compared to the ETF I share—XEF.TO—is the structure behind it.

While iShares XEF.TO holds stocks directly, Vanguard's VDU.TO holds the U.S.-listed counterpart ETF: VEA. This causes VDU.TO to have an unnecessary extra layer of withholding taxes, taking between 0.30 and 0.35 percent off the overall return each year at current dividend yields.

Another advantage to XEF.TO is its superior liquidity. There are about 6 times as many shares outstanding and about 4 times as many shares traded each day compared with VDU.TO. Given that Dual Momentum calls for complete account turnovers, this can have an impact, especially on larger portfolios.

If we add the impact of the withholding taxes alone, the return advantage of investing with VDU.TO compared to XEF.TO would have shrunk to approximately 0.25 percent. If you add the liquidity costs, this could conceivably shrink down quite a bit more.

In Canada, this leaves us with two viable options to tackle this issue:

  1. We could choose VDU.TO instead of XEF.TO, in addition to our emerging markets ETF; or
  2. We could convert our accounts to U.S. dollars and invest in U.S. listed ETFs where several low-cost global ex-U.S. ETFs exist which include Canada in their holdings.

Again both of these options complicate our investment process for Canadians.

If you have a smaller account where liquidity is not a major problem, I would recommend using Vanguard's ETF pair. This means a 75 percent allocation to VDU.TO and a 25 percent allocation to VEE.TO when the signal calls for International stocks.

For larger accounts, we are back to switching our accounts to U.S. dollar accounts and buying U.S.-listed ETFs.

You will notice I do not think adding a Canadian ETF to the iShares ETFs is a good option.

Canadian stocks make up less than 6 percent of the All-World ex-U.S. markets. In my view, it would not make sense to hold three ETFs when the signal is in International stocks.

Comments & Questions

All comments are moderated before being posted for public viewing. Please don't send in multiple comments if yours doesn't appear right away. It can take up to 24 hours before comments are posted.

Comments containing links or "trolling" will not be posted. Comments with profane language or those which reveal personal information will be edited by moderator.

5 Replies to “Dual Momentum: Evaluating the International Stocks Component”

  1. Great summary. Thanks.

  2. Nice article! Could you please comment on the reason why you don’t think that adding a Canadian etf for CA exposure makes sense? In my opinion, this wouldn’t add significant complexity or cost for a portfolio of considerable size, even if the Canadian portion is only 5%. Furthermore, did you consider the potential tax loss harvesting opportunities made possible when the 3 components are distinct (eg XEF+XEC+XIC) instead of combined in a single etf?

    1. Daren (Editor) says:

      When you consider that U.S. dollar accounts are optimal, it wouldn’t make sense to transact in an extra ETF holding worth no more than $2,500. The tax loss harvesting in a non-registered account would be minimal on a holding that size, especially when DM boots you out of the position when 12-month returns are negative (or earlier).
      There could be some tax harvesting opportunities by splitting developed and emerging markets as these do not always track the same. If you’re account is very large, you could even split into Europe, Japan, and emerging.

    2. I agree with you. By considerable size, I guess I had in mind something more like >300k. In my case, I tend to favour Canadian listed securities in taxable accounts to avoid the hassle of having to declare foreign property above 100k on my tax returns.

  3. It’s worse for us Brits. Apart from individual country ETFs (so US is fine) I have a choice of World with or without UK, Europe with without UK, Asia and Emerging Markets. Thankfully the Dual Momentum signal continues to be US while I wonder what to do about my non-US options!

Leave a Reply

nine × one =