Before the release of all-purpose asset allocation ETFs, many Canadian index investors built their ETF portfolios using three funds, including:
- A Canadian bond ETF
- A Canadian equity ETF, and
- A global equity ETF, that excludes Canadian stocks
Even though you can now conquer the world (or at least invest in it) with a single asset allocation fund, many of you may still prefer the three-ETF approach. Maybe the control freak in you can’t quite embrace investing in just one fund—even if it supposedly does it all for you. Or maybe you’d prefer to tweak your equity home bias. Rather than sticking with the 25%-30% Canadian equity allocation found in many asset allocation ETFs, you may still want to tinker with your domestic vs. foreign stock mix by holding different funds for each.
For whatever reason, let’s say you’re sticking with a three-fund portfolio. Which global equity fund is right for you? Fortunately, there are really only two ETFs that deserve your attention:
- The Vanguard FTSE Global All Cap ex Canada Index ETF (VXC)
- The iShares Core MSCI All Country World ex Canada Index ETF (XAW)
If you’re short on time, here’s today’s compelling conclusion on how to choose from this menu of two: With the exception of one minor detail related to XAW’s U.S. equity holdings, it doesn’t really matter. They’re both solid choices, with such minor differences, you’re unlikely to go horribly wrong either way.
Asset Allocation: VXC vs. XAW
When it comes to asset allocation, either VXC or XAW offer you broadly diversified foreign equity exposure in relatively equal measures, with seemingly similar asset class ratios.
- VXC tracks the FTSE Global All Cap ex Canada China A Inclusion Index, which allocates around 58% to U.S. stocks, 30% to international stocks (including Korean companies), and 12% to emerging markets stocks.
- XAW tracks the MSCI All Country World ex Canada Investable Market Index. It has nearly identical equity region weights to those found in the FTSE index.
Both indexes tracked by VXC and XAW include thousands of large-, mid-, and small-cap companies. We know this, because of the “All Cap” and “IMI,” or “investable market index,” in their names. Both terms suggest you’re investing in the same world of globally diversified, ex-Canadian investments. And each fund allocates this world in a similar fashion.
China A Shares: VXC vs. XAW
We now know VXC and XAW will provide similar market exposure and plenty of diversification. But there are some small differences in how each fund allocates shares of companies located in China: more specifically, China A shares.
China A shares trade on the China mainland, are denominated in Renminbi, and make up just under half of China’s investable stock market. Until very recently, they were excluded from both the FTSE and MSCI global equity indexes. That’s no longer the case today, and there are a couple of seemingly noteworthy differences between how the two funds manage their China A share allocations.
- The FTSE index tracked by VXC includes “China A Inclusion” in its name, and holds nearly 1,500 more China A shares than the MSCI index tracked by XAW.
- The FTSE emerging markets index followed by VXC includes large-, mid-, and small-cap China A shares, while the MSCI index includes only large- and mid-cap China A shares.
At a glance, these differences seem like a big deal. In reality, China A shares end up representing such a tiny portion of each index, I don’t see a compelling reason to prefer either fund at this time.
If we dig a little deeper, we’ll see why.
While inclusion rates are expected to increase over time, MSCI’s emerging markets indexes currently include only 20% of the China A shares market cap. FTSE has a slightly higher inclusion rate of 25% (both MSCI and FTSE inclusion weights are as of December 31, 2020). With these ratios in mind, here’s what happens if we separate China A shares from the general emerging markets allocation:
- China A shares only make up around 0.9% of the FTSE index and only 0.6% of the MSCI index (as of December 31, 2020).
- Even more interesting, the 1,000-ish extra small-cap China A shares tracked by the FTSE Global All Cap ex Canada China A Inclusion Index (but not by the MSCI index) make up less than 2% of the FTSE index (as of December 31, 2020).
In other words, FTSE’s greater diversification into China A shares turns out to be more of a red herring than a distinction worth fishing for.
MERs and Foreign Withholding Tax Ratios (FWTRs): VXC vs. XAW
Lower fees are often another strong argument for favoring one fund over another. But once again, if we crunch the current costs, we find neither ETF has an edge.
For years, VXC had lagged behind XAW on both fees and tax-efficiency. But in November 2019, Vanguard lowered VXC’s management expense ratio from around 0.27% to just 0.21%. This is comparable to XAW’s MER of 0.22%.
VXC also used to be less tax-efficient than XAW. For its international equity allocation, it held two U.S.-based ETFs: The Vanguard FTSE Europe ETF (VGK) and the Vanguard FTSE Pacific ETF (VPL). This combination created an extra layer of foreign withholding taxes across all account types, with an expected unrecoverable foreign withholding tax ratio of around 0.45% per year in RRSP and TFSA accounts.
XAW, on the other hand, gains its international equity exposure from the iShares Core MSCI EAFE IMI Index ETF (XEF), which is a Canadian-based ETF that holds the international stocks directly. This creates one less layer of withholding taxes, resulting in a foreign withholding tax ratio of around 0.32% per year when held in an RRSP or TFSA.
But again, in November 2019, VXC dumped VGK and VPL. It replaced them with the Vanguard FTSE Developed All Cap ex North America Index ETF (VIU), a Canadian-based ETF that holds the international stocks directly.
As a result, VXC and XAW both had the same foreign withholding tax ratio of 0.32% in 2020, which implies similar tax-efficiency for both going forward. It would appear that neither ETF has the edge on product costs or foreign withholding taxes.
Short-Term Tracking Error (2020): VXC vs. XAW
We’ve saved the best anti-climactic comparison for last. When you buy an index fund, you expect it to closely track its index (after fees, of course). Hence the name, “index fund”. But during 2020, XAW lagged its index by more than 2%, while VXC lagged its index by just 0.5% over the same period.
As we’ll explain next, XAW hit some tracking error headwinds in 2020, due to its decision to select slightly more actively managed U.S. equity ETFs. Here’s what happened …
As mentioned earlier, XAW tracks the MSCI All Country World ex Canada Investable Market Index. What we did not mention is XAW is actually a fund-of-funds, holding multiple ETFs to approximately track its foreign equity index.
XAW’s U.S. equity component follows the MSCI USA Investable Market Index. But there are no ETFs tracking this index, so iShares had to get creative. They could have shot for “close enough” with one broad U.S. equity ETF like the iShares Core S&P Total U.S. Stock Market ETF (ITOT). Instead, they have decided to hold four U.S. equity ETFs, hoping to reduce XAW’s index tracking error.
Unfortunately, the attempt had the opposite effect in 2020, because three of the S&P ETFs are not entirely passive. Instead of just tracking every company within the asset class, each index chooses from a representative sample of the whole. Based on numerical restraints, a committee selects which companies’ stocks are included or excluded:
- The iShares Core S&P 500 ETF (IVV) tracks the S&P 500 Index of 500 large-cap companies.
- The iShares Core S&P Mid-Cap ETF (IJH) tracks the S&P MidCap 400 Index of 400 mid-cap companies
- The iShares Core S&P Small-Cap ETF (IJR) tracks the S&P SmallCap 600 Index of 600 small-cap companies.
With their respective 500-, 400-, and 600-company limits, these indexes ended up excluding a number of U.S. stocks that happened to shoot the lights out in 2020. Most notably, there’s Tesla. Even though it returned over 700% in Canadian dollar terms, Tesla was excluded from the S&P MidCap 400 Index throughout 2020, and was only added to the large-cap S&P 500 Index toward year-end. These exclusions caused the U.S. equity component of XAW to underperform its MSCI USA IMI benchmark by around 3.2% in 2020.
Ironically, XAW could have simply held ITOT for its core U.S. equity holding, and significantly reduced its short-term tracking error. Following a passively managed broad-market U.S. equity index, ITOT doesn’t have the same company selection constraints. In other words, rather than trying to choose a representative batch, it simply tracks the entire lot.
Of course, Vanguard faced similar challenges in 2020. There is also no U.S. equity ETF that follows the FTSE USA All Cap Index. However, Vanguard’s selections didn’t suffer from the same S&P committee conundrum. Instead, they’ve combined two existing Vanguard ETFs for a more purely passive approach:
- The Vanguard Large-Cap ETF (VV) tracks the CRSP U.S. Large Cap Index
- And the Vanguard Small-Cap ETF (VB) tracks the CRSP U.S. Small Cap Index (which also includes U.S. mid-cap companies).
Long-Term Tracking Error: VXC vs. XAW
So, we now know that XAW had a brutal 2020 tracking error compared to its index. But is the one-year stumble a good enough reason to dump or avoid the fund? If we look longer-term, we’ll see the jury is still out on who the true tracking-error champ is.
Since the first full month of its launch in March 2015 through December 2020, XAW returned 9.4% annualized. Over that same period, the MSCI All Country World ex Canada Investable Market Index returned 9.7%. That’s a tracking error of negative 0.3%. This seems reasonable, as XAW’s MER would be expected to cause tracking error of at least negative 0.22%.
And what about VXC? Well, it also returned 9.4% annualized over the same period. And its underlying index also returned 9.7%. So, its tracking error was an identical negative 0.3%.
In other words, even with XAW’s recent stumble, both XAW and VXC had comparable returns and tracking error to their index over the past 5+ years.
But what if we calculate the monthly tracking error for each fund between March 2015 and December 2020? Tracking error is commonly used to describe a fund’s annual outperformance or shortfall relative to a benchmark. But in this more technical definition of tracking error, we’ll calculate the standard deviation of the differences in their monthly returns. Just remember, the lower the historical tracking error, the more closely an ETF has tracked its index.
For VXC and its index, we find a very low tracking error of 0.468%. Over the same period, XAW and its index had a tracking error of 0.314%, indicating an even tighter tracking error.
So, by this measure, even though XAW underperformed its benchmark in 2020, XAW actually has had a lower—not higher—tracking error to its index than VXC has had to its own. At least historically, we can’t claim that XAW has had poor tracking error to its index since inception. That’s just not true.
Choosing Between VXC and XAW
Of course, the past isn’t always prologue, and that’s why we still slightly prefer VXC over XAW. XAW’s arguably active shenanigans are still going on under the hood. So, who knows when the next proverbial Tesla will take XAW investors on another ride?
If iShares switched XAW’s underlying U.S. equity ETFs to a single broad-market fund like ITOT, this would create large capital gains distributions to taxable XAW unitholders. But we would then be indifferent between XAW and VXC moving forward.
So where does that leave us today? Well,
- New Investments: If I were adding new cash to my global (ex Canada) ETF position, I would likely purchase more shares of VXC.
- Existing Tax-Deferred Investments: If I held XAW in a TFSA, RRSP, or other tax-deferred account, I would consider switching the position to VXC, since there would be no tax implications.
- Existing Taxable Investments: If I held XAW in a taxable account, I would be reluctant to sell the holding and replace it with VXC, unless at least one of these statements were true:
- I would realize no capital gains on the XAW position
- I was expected to be in a low tax bracket in the year of the XAW sale, or
- I had net capital losses carrying forward that could offset the taxable capital gains on the XAW sale
Could you please explain what you mean by proper market cap weightings from your answer- “VXC would be very similar to holding VUN/VIE (I think you meant VIU though?)/VEE in the proper market cap-weightings” and how to go about calculating this?
@Stephanie – This video discusses “market cap-weighting” concepts: https://www.youtube.com/watch?v=5qv0TXZba4E
Do VXC and XAW rebalance their funds in a similar way that the one ticket funds rebalance? How often does the rebalancing occur?
Would holding VXC be equivalent to holding VIU, VEE and VUN?
I would like more control of my ETFs for asset location purposes than a one fund ETF can provide but would like to still simplify where possible. I hold most foreign equity in my TFSA at the moment so if VXC is just like holding VIU, VEE and VUN then it is a very attractive option.
@Stephanie – VXC and XAW follow market cap-weighted indexes, so the ETF providers continuously rebalance to achieve this objective (they’ll use new cash flows as much as possible to reduce tax implications).
VXC would be very similar to holding VUN/VIU/VEE in the proper market cap-weightings.
Slightly off topic.. can you recommend an equivalent strategy to xaw/vcn for someone holding large us cash (in tfsa/rrsp). Would exchanging to cad and buying xaw/vcn.
@Nip – for a TFSA, VEQT would be similar to a mix of VCN/XAW. For an RRSP, you could hold VEQT, or if you wanted to reduce foreign withholding taxes and product costs, you could hold a combination of VCN/VT.
I thought holding US ETF inside RRSP triggers no withholding taxes? So is it better to hold VT inside RRSP?
@Nip – Correct – I mentioned VT in my response if you were looking to reduce foreign withholding taxes and product fees.
If, on average, XAW outperforms its benchmark, then I guess I can live with its higher relative tracking error =)
Thank you. Just to confirm XEQT doesn’t have the same problem because it holds ITOT. Correct?
@Christina – Correct! :)
Did vxc have a substantial tracking error in 2021? It looked like it lagged xaw quite a bit.
@Christina – VXC outperformed its benchmark by 0.62% in 2021, while XAW outperformed its benchmark by 1.22% (so XAW once again had a higher tracking error to its index, relative to VXC – it’s just that no one complained about XAW because it was positive tracking error).
Another reason to choose VXC: Its position value is significantly higher than XAW. VXC costs ~$55 today, XAW costs ~$35. Some brokers like IB charge per quantity. If you were to invest $10k, IB would charge $1.81 for VXC and $2.85 for XAW in terms of trading costs. Of course, the difference is small and negligent (0.01%).
@Thiago – Another good point. All else equal, the higher-priced ETF is generally the preferred option.
“the higher-priced ETF is generally the preferred option”
Why is that? Is that just a general rule of thumb when trying to choose between two different ETFs? Is there some merit behind that way of thought?
@Vito – All else equal, bid-ask spreads (in percentage terms) tend to be lower with a higher-priced ETF.
Thanks for taking the time to write this article and put the video together. Very well done and useful as always!
I hold XAW (and VCN) in a taxable account, so at this point I guess it wouldn’t make sense to switch from XAW to VXC.
Hopefully XAW doesn’t have the same tracking error issue in 2021. But I am sure you will let us know in 2022!
I also like the idea of having a iShares ETF and a Vanguard ETF (as opposed to having my 2 ETFs with the same provider).
I used to have the BMO Discount Bond ETF but got rid of it during the “pandemic” crash of March 2020 and invested in more XAW/VCN instead (the 30% market dip didn’t bother me at all and I figured I was comfortable increasing my Equity allocation).
@Jacky: What what it’s worth, YTD (as of October 31, 2021), VXC is outperforming its benchmark by 0.25%, and XAW is outperforming its benchmark by 0.57%.
Thanks for this information – it will be helpful in my upcoming portfolio adjustment. I do have a question in terms of the best way to do that ajustement.
To further simplify my taxable account portfolio (currently based on your Ridiculous portfolio!), I am thinking of replacing my US/International/emerging component (VUN, VIU, VEE) with VXC. However, this will generate significant capital gains. Any insights on the best strategy for the adjustment?
All at once (immediate tax impact)?
Over time (through re-balancing – which could likely make my portfolio more complex to manage up until the transition is completed (as it would add VXC to the mix)?
Following a market correction (which may not necessarily offset the full capital gains)?
Continue with current strategy (and accept that it is slightly more complex)?
Currently leaning either through regular re-balancing or status quo!
@Mike – It’s really a personal choice. If you’re holding VUN/VIU/VEE in the proper market-cap weights, you could just start adding to VXC instead. If there are some tax-loss selling opportunities in the future, perhaps that would be a better time to sell VUN/VIU/VEE and switch to VXC. Again, all of this would depend on your tax rate, and your comfort level with managing a more complex portfolio.
Thanks for the great advice.
A took a moment to better understand my pain point and it was on my re balancing spreadsheet and the fact that I was entering my portfolio manually… I took a bit of time to automate my excel spreadsheet and with that I will be able to continue with the status quo.
@Mike – I’m glad you enjoyed the post – best of luck with your portfolio rebalancing :)
Thanks for the article Justin. I made the switch from D Series MFs to ETFs in early 2016 after reading Dan’s CCPP Blog. I followed his recommendations and bought VAB, VCN and VXC. A couple of years later Dan switched his Global ETF recommendation from VXC to XAW, so I switched all the VXC to XAW. Now a couple of years after that you’re recommending doing the reverse switch. I am with RBC DI, so it’ll cost $9.95 per trade plus bid-ask.
I understand your reasoning of course, but am just a little hesitant to switch back again.
@Brian – It’s not absolutely necessary to switch back to VXC. We still expect both ETFs to have similar long-term returns going forward (even if their short-term returns will sometimes deviate, due to the differences discussed).
There is another reason to own VXC which is the fact that most of the low costs quality products and amazing all in one ETFs are because of Vanguard. If it wasnt for Vanguard, canadians would not have access to these and for this simple reason, i choose VXC to reward the company that really helps investors instead of profiting from investors. Even if XAW became a bit cheaper, i would still choose VXC.
The fact that There are actively managed funds in what is supposed to be an index tracker is just another indication that you should be careful when buying investiments products from Blackrock.
Blackrock evolves because Vanguard forces them to. Otherwise they would still offert expansive and crappy products for speculators and trend followers. Vanguard all the way!
This is the article is amazing! and is the same video you made in on YouTube! So thank you for both!
I have been debating about making the switch from XAW to VCX but just didn’t pull the trigger, but after watching the video again and reading your post, I think I will make the switch.
I do have one question based on your last 3 scenarios.
I am in the #2 scenario category, I have XAW in my TFSA, making the switch to VXC would only incur a commission and ECN fees which are not significant, I calculate around $15 all-together. Now you say that you’d consider switching from XAW to VCX, but does that mean, simply start a new position to VXC or sell my current XAW position and then buy VXC?
As I said, the $15 is not significant, but considering I still have 2 – 2.5 decades of employment left in me, based on this article I think it makes sense to pull that trigger.
Thanks again Justin for all your contribution and work, your videos are so easy to follow and comprehend, I very much appreciate it.
@Vito: I’m so glad to hear you’re finding the videos helpful :)
If your XAW holdings are in a TFSA (and you are comfortable with the switch to VXC), I don’t see any reason to continue holding your existing XAW units in the TFSA.
Thanks for the deep dive into the two. I don’t know if you’ve mentioned this in the past but how does trading volume and bid-ask spread factor into your analysis? I’ve been burned in the past before with some lower volume ETS from some of the smaller providers.
@Leon YYC – The bid-ask spreads on VXC and XAW tend to be between 1-3 cents (so fairly tight). Trading volume is adequate as well (always use limit orders placed at the current bid/ask price when selling/buying).