Tracking Error on International Funds

I recently received an email from a reader, J.W., who wanted to know why the tracking error on some popular Vanguard international equity ETFs were so high in 2014. He noted, for example, that the Vanguard FTSE Developed ex North America (VDU) lagged its benchmark index by 1.62% last year, far more than one would expect.

An index fund’s tracking error is the difference between the performance of the fund itself and that of its benchmark. If the index returns 10% on the year and the fund delivers 9.8%, the tracking error is 0.20%, or 20 basis points. But what could possibly cause a fund to show a tracking error of 162 basis points?

Any time you see a surprising number like this, it’s important to determine the reason: otherwise you risk making a bad decision because you’re working with inaccurate or misleading information. If an index were to lag its benchmark by more than 1.6% because it was badly managed, then you should look for a better alternative. But Vanguard has a long record of tight tracking error, so something else has to be going on here—and indeed it is.

Back on track

To understand VDU’s large tracking error—and why it’s not as bad as it looks—let’s look at the reasons its performance deviated so far from the index.

1. Management fees. As we’ve mentioned, this is the most obvious culprit. You should always expect an index fund to lag its benchmark by an amount equal to its management expense ratio, which includes the management fee, taxes and some small incidental costs. VDU’s management fee is now 0.20%, though the full MER was 0.30% in the fund’s mid-year performance report because the new lower fee only came into effect in October 2014.

2. Foreign withholding taxes. International equities are subject to withholding taxes on dividends, which our recent white paper estimated to be about 7.5%. The yield on international developed stocks is quite high—currently close to 3%—so this would shave another 22 basis points or so from the investor’s return. Moreover, because VDU uses a US-listed ETF as its underlying holding, it is subject to an additional 15% withholding tax by the US. Based on the methodology in our white paper, the total impact of these foreign withholding taxes would be close to 0.70%.

Note that you would only bear the full cost of these taxes if you held the fund in an RRSP or TFSA. If you hold VDU in a taxable account, you may be able to recover the Level 2 tax (the portion withheld by the US) by claiming the foreign tax credit on your tax return.

3. Fair value pricing adjustments. The first two components explain about 100 basis points of tracking error. Now for the more complicated part: fair value pricing. I’ve touched on this idea before, both in the context of tracking error and to help explain why the market price and net asset value (NAV) of an ETF can diverge temporarily.

International equity funds such as VDU holds stocks that trade in Europe, Asia and Australia. These markets are rarely, if ever, open at the same time as North American markets. So when the Toronto and New York exchanges close at 4 p.m. local time, the market prices of the stocks are stale. The idea behind fair value pricing is to adjust the closing net asset value of an ETF to reflect what the stocks would be worth if the overseas markets were open. While this may sound manipulative, it’s actually required by US regulators to protect mutual fund investors from “time zone arbitrageurs.”

The thing is, while each fund’s NAV is subject to fair value pricing, the benchmark is not. On volatile days the change in the index may be quite different from the change in the fund’s adjusted NAV. If one of those volatile days happens to be the start or end date of the period you’re measuring, the result can be a significant—but highly misleading— tracking error.

Why you don’t need to worry

Tracking error from management fees and taxes detract from your returns, but fair value pricing adjustments do not. Unless you purchase your entire ETF holding on December 31 of one year and sell it all on December 31 the following year, the tracking error reported for a single calendar year doesn’t mean much at all.

The impact of fair value pricing is highly time-dependent, because on some days the adjustment is much larger than on others. This means that if you shift your start and end dates by just a few days, there’s a good chance the differential will disappear. Moreover, the adjustment can go in both directions: sometimes it increases tracking error while at other times it reduces it.

To illustrate this idea, Vanguard Canada provided the following performance data for the Vanguard FTSE Developed Markets ETF (VEA). It compares calendar year 2014 with one-year numbers using slightly earlier start and end dates. (These returns are reported in US dollars, and international equities had negative returns last year for US investors.)

+1 day 2014 -1 day -10 days
VEA return -4.79 -6.02 -5.23 -1.83
Benchmark return -4.44 -4.85 -4.60 -2.13
Difference -0.35 -1.17 -0.63 0.30
Source: Vanguard

As you can see, the tracking error for VEA in 2014 was 117 basis points. But if you measure the 12-month return using December 30 (-1 day) or January 2 (+1 day) as your start and end dates, the tracking error for the fund looks much smaller. If you use December 17, the tracking error actually appears to be positive.

The message here is that an international equity fund’s true performance needs to be considered in context. Choosing an arbitrary start and end date can produce highly misleading data. So be aware of your ETF’s management fees and withholding tax implications. But if the reported tracking error is significantly higher or lower than that, chances are it’s a result of a fair value pricing adjustment that has no effect on your investment returns.

 

39 Responses to Tracking Error on International Funds

  1. CheapSkater January 26, 2015 at 11:48 am #

    Interesting piece. So then for a fund that has been around for a while, tracking error since inception should more or less match drag associated with management expenses and withholding taxes only. In other words, fair value pricing adjustments should cancel out in the long term, correct? So then is there something else happening with PXH (PowerShares FTSE RAFI Emerging Markets Portfolio)? I was always under the impression that PXH’s large tracking error since inception was the result of sampling expenses incurred by the fund when its assets under management were (even) smaller in its early days. Any thoughts?

  2. Tyler January 26, 2015 at 12:22 pm #

    Thanks for the very informative article. I had noticed something similar with XEF and wondered what was up, and then noticed the same tracking error also affecting funds from Vanguard and BMO.

    Also regarding “VDU’s management fee is 0.20%, though the full MER was estimated at 0.30% in the fund’s mid-year performance report, probably due to some start-up costs that won’t apply going forward.” I thought I’d note that when that document was released, its fee was actually 0.3%. It only dropped to 0.2% in October, after that report was published.

  3. Scotty January 26, 2015 at 12:36 pm #

    A bit off topic, but I’ve been noticing that every time the FWT tax credit is discussed it seems as if it not a for sure thing. In this article:

    “If you hold VDU in a taxable account, you MAY be able to recover the Level 2 tax (the portion withheld by the US) by claiming the foreign tax credit on your tax return.”

    What situation would stop you from getting the tax credit?

  4. Constantine January 26, 2015 at 1:49 pm #

    While we can simply calculate the net effect of management expenses and tax withholding, is there any way to correct for the effects of fair value pricing adjustments to judge an ETF’s ability to reproduce an index?

  5. JohnR January 26, 2015 at 3:42 pm #

    Hello
    Im looking to rebalance my portfolio. I switched from my financial planner a few years ago and went to being a couch potato. . I was in all Canadian hedged dollars at the time. I switched approx. 350K from cdn accounts to vti and vxus in us dollars when our dollar was around par to usd.
    Im thinking of selling vti and vxus, using norberts gambit and converting to cdn dollars and rebuying my us and international etf’s in hedged accounts.( taking a great gain from us markets and also approx. 25% gain on the currency). Then possibly down the road if our dollar comes back close to par switch back again. By the way this is in an rsp account.
    Does this make sense to try and play with the currency and try and solidify my gain on the us dollar? I know the us dollar could get stronger but Im pretty happy with the 25% gain.
    Thanks

  6. Canadian Couch Potato January 26, 2015 at 3:44 pm #

    @CheapSkater: Representative sampling can indeed be a problem, though it is not a factor with VDU or any other Vanguard fund I am aware of. It used to be a huge issue with the RAFI Fundamental ETFs, but I think they have corrected that in recent years:
    http://canadiancouchpotato.com/2010/04/28/international-tracking-error-part-2/

    @Tyler: Thanks, you’re right. I have updated the post accordingly.

    @Scotty: For starters, the foreign tax credit is non-refundable, so if you don’t pay tax you won’t be able to recover the FWT. It can also get complicated if you have foreign business income, or if you invest in a corporate account. I’m not a tax specialist, so I like to use tentative language when I write about these issues:
    http://www.taxtips.ca/filing/foreigntaxcredit.htm

    @Constantine: The best thing to do is look in the year-end Management Report of Fund Performance, where the FVP adjustment will be disclosed. See this post for an example:
    http://canadiancouchpotato.com/2013/04/25/when-you-can-ignore-tracking-error/

  7. pjb January 27, 2015 at 1:06 pm #

    @JohnR,

    As Dan has pointed out before, you won’t actually gain anything by moving your investments into Canadian dollars: http://canadiancouchpotato.com/2014/01/16/currency-exposure-in-international-equity-etfs/

    With VXUS, you have no exposure to US$, even though it’s denominated in US$. With VTI, your exposure to US$ is the same whether you hold VTI or its Canadian equivalent. The only way you can “win” is if you sell your VTI and never invest in US equities again. That doesn’t sound like much of a bargain in the long run.

    I realize that this is counterintuitive; Dan does a good job of explaining it in the Jan 2014 post, along with a few others.

  8. Jake January 27, 2015 at 1:55 pm #

    A bit off topic but why did gic’s rates drop way more than the 0.25% that was reported in the news? first of january i got a 2 yr gic at 2.15 %, i have another gic maturing in a few days and was going to put it in a 3 yr gic but the best rate for 3 yr gic on bmo investoriline today is 1.85%, a 5 yr best rate is 2.15%.
    best 2 yr gic today shows 1.75% which is 0.40% below what i got just 3 weeks ago.
    i don’t get the extreme decline.

  9. Andrew January 27, 2015 at 2:34 pm #

    @pbj JohnR can make a currency bet by selling VTI, converting the USD to CAD and then subsequently buying VUS (VTI hedged).

    @JohnR In the spirit of this blog, the true couch potato doesn’t try to time the market. The Canadian dollar could go down to 0.7 and reduce your future gains with the above move or go back to parity and increase your future gains by the above move. Either way it’s a gamble and none of us can predict the future dollar value or pick a market bottom or top. All the previous articles here have shown the unhedged approach to have lower volatility over the long term in a balanced portfolio.

  10. Canadian Couch Potato January 27, 2015 at 4:06 pm #

    @Jake: GIC rates are not tied directly to the overnight rate (the one the Bank of Canada recently cut). GIC issuers choose their own interest rates based on their specific needs: for example, sometimes you will see an issuer offer a very good rate on a 5-year GIC but a lousy rate on a 3-year GIC. The reason is that they have some specific need for five-year money and they make an offer in the market accordingly.

  11. Wilkster January 27, 2015 at 5:08 pm #

    Great information here at CCP. I recently reviewed your new CCP model portfolios and have a question regarding the Vanguard bond ETF’s. Previous model portfolios recommended VSB (short term bond index) whereas the new ones recommend VAB (aggregate bond index). I’m in the process of rebalancing my RSP/TFSA and was wondering for the bond rebalance if it makes more sense to add the new money to my existing VSB or to purchase VAB as an alternative. Note that I am not planning on selling my existing VSB. Whats the major difference between the two other than the MER. Thanks.

  12. phil January 27, 2015 at 5:24 pm #

    Hi dan, it seems to me that tdb911 is lagging as well. Could the ideas you presented for the drag on VDU be this e series fund? Or should I be concerned?

  13. Canadian Couch Potato January 27, 2015 at 7:00 pm #

    @phil: The same principles will apply, though the details will be different. No need to be concerned about TDB911.

    @Wilkster: Only one of the old model portfolios (the Uber-Tuber) used short-term bonds: the rest always used broad-based funds such as VAB. There is quite a big difference between the two: VSB is a short-term bond fund (average term about 3 years) while VAB includes all maturities (average about 10 years). VSB should be less volatile, but should also have a lower expected return:
    http://canadiancouchpotato.com/2011/07/07/holding-your-bond-fund-for-the-duration/

  14. Franko January 27, 2015 at 8:00 pm #

    Hi Dan,

    Another great article, thank you!

    Can you give any examples of red flags or times where we SHOULD be worrying about tracking error issues beyond the above considerations? (say significantly underperforming the index over several years, etc)

    While I do trust Vanguard to keep their true tracking error minimal, I also like to be able to quantify the numbers for myself if possible. The reason I ask is that I hold a couple “gambling” ETFs in my couch potato portfolio that play in specialty markets such as emerging market small-caps, and I find that there are often substantial tracking errors with these funds, so I’d like to know how best to determine what is “normal” benchmark lag and what warrants additional inquiry.

    Frank

  15. Canadian Couch Potato January 27, 2015 at 8:14 pm #

    @Franko: I would suggest looking in the annual Management Report of Fund Performance document, which usually comes out in February or March. It will disclose the fund return and the benchmark return, and usually explain the reasons for the tracking error.

  16. John r January 28, 2015 at 3:02 am #

    Hello
    @pjb
    Vxus is bought in us dollars so I’m pretty sure there is exposure to us dollar.
    I was just thinking of going into a cdn hedged account to protect for now the 25% gain I received on the currency since I bought the us dollars a few years ago at par.
    How many times have you heard someone say I wish I bought a whole bunch of us dollars back when it was even with our dollar. That’s what I did and I was just hoping to take that gain for now and possibly buy back into us dollar down the road if our cdn dollar goes back to par. (I know it’s kind of gambling but that was kind of the idea back when I bought the us dollar at par. ) I guess I got lucky and would hate to see our dollar go pack to par without capturing the gain.

  17. pjb January 28, 2015 at 10:32 am #

    @JohnR

    Let’s say that VXUS holds 1 share of a European company, priced at €100. And let’s assume, for simplicity, that CDN$100 = US$100 = €100.

    Now, let’s say that US$ goes up by 10% while CDN$ and € don’t change relative to each other. Now, CDN$100 = US$90.91 = €100. So, while the underlying price of the share has not changed (it’s still €100), the price of the share drops by 10% in US$. If you’re an American, then it’s become 10% cheaper. However, for a Canadian, the 10% drop in price in US$ is cancelled out by the 10% increase in the CDN$:US$ exchange rate. Your CDN$100 still buys 1 share at €100, even though the intermediate currency (US$) has changed value.

    That’s the gist of Dan’s articles on currency exposure.

  18. Steve January 28, 2015 at 7:05 pm #

    @ John r :

    >> “Vxus is bought in us dollars so I’m pretty sure there is exposure to us dollar.”

    No, there is not USD exposure when a Canadian investor buys VXUS.

    The easiest way to think about this might be to fall back on primary school fractions:

    – Let the numerator (top) be the “home” currency of the asset buyer

    – Let the denominator (bottom) be the currency of the asset being purchased

    So when VXUS (in the US) buys world stock (outside US), the fraction looks like:

    USD / World Currencies

    And when you (in Canada) buy VXUS (in the US), the fraction looks like:

    CAD / USD

    Now to get your net currency exposure, multiply the 2 fractions in the way you learned in 3d grade or thereabouts:

    (CAD / USD) X (USD / World) = (CAD X USD) / (USD X World)

    The 2 “USD” in the numerator and denominator cancel out, leaving the reduced fraction =

    CAD / World Currencies

    This is your net currency exposure from buying VXUS as a Canadian – no USD.

    With respect, if this is a difficult concept, it might be best to just buy a good fund like VXUS and rebalance as needed per your asset allocation plan, while completely ignoring any and all currency movements.

  19. JohnR January 28, 2015 at 8:08 pm #

    @pjb

    No I still dont think so.

    If you buy vti or vxus I believe you are essentially lets say an american, if you have converted your money to usd and are trading on a us stock exchange. You dont become a canadian again until you sell (vti,vxus) and you convert it back to canadian dollars.
    example 1.- Lets say you buy vti and down the road you just sell the it in american dollars and open up an american account, you could live down in Florida and everything would be at par.
    On the other hand you could sell the etf(vti,vxus) , keep it in american dollars and lets say 1 week later the american dollar got stronger and lets say 1 us dollar costs 1.50 canadian. This would be a good time to be a canadian again and convert back to canadian currency. ( lets say you had $100,000 usd it would now be worth 50% more and you would have $150,000.00 to spend in canada.

    Now hedging is different. Lets say you want to buy into the us stock market and want to hedge.(I think xsp ca follows the us market but is hedged) You are essentially staying canadian the whole time(keeping your money in cdn dollars) and your canadian investment is just mirrorring the us market. Someone else(not sure who) is doing the hedge risk,and taking any currency conversion out of the picture)

    Dont think vxus is any different. Once you exchange your money into us dollars and trade on an american stock market the canadian exchange between canada and say europe doesnt matter. It only matters between the us and europe. Now if you sold vxus and converted back to canadian dollars thats still only affects the us and canadian dollar conversion.

    The other way to buy into us market is to buy a etf on toronto stock exchange that is not hedged. This way you dont have to convert your money into us dollars but the currency still comes into play. If us dollar gets stonger its a bonus.

    Ok Thanks. Let me know what you think. This is what I took out of the article and what I have learned.

  20. Canadian Couch Potato January 28, 2015 at 8:43 pm #

    @JohnR: Steve is correct. If you measure your returns in Canadian dollars, the USD/CAD exchange rate will have no effect on the performance of an international equity ETF like VXUS. This post may help illustrate the concept more clearly:
    http://canadiancouchpotato.com/2011/04/07/a-case-study-in-currencies/

  21. pjb January 28, 2015 at 9:18 pm #

    @JohnR

    “Let me know what you think.”

    I think @Steve is absolutely correct. His explanation explicitly added US$ in the numerator and denominator terms that I left out, which made the math even more clear.

    I think Dan is absolutely correct in his articles.

    I think your analogies are incorrect, and that you’re confusing yourself with them.

    I think that if you can find any financial writer anywhere who thinks that it’s a good idea to dump VXUS/VTI now and shift the money back to Canada, you should go for it.

    But in the meantime, and with all due respect, I strongly echo @Steve’s response advice “…just buy a good fund like VXUS and rebalance as needed per your asset allocation plan, while completely ignoring any and all currency movements.”

  22. John r January 29, 2015 at 8:15 pm #

    Ok why don’t you all admit you were wrong and we can get on with it. (Just kidding)
    As far as vxus goes I think I got it now. To put it simple as with pjb’s analogy, if the us dollar gets stronger compared to say the European dollar while in vxus you don’t do as well because their currency got weaker and converting back to us you get less us dollars . Then if you go back to Canada you make up the difference.
    (Please don’t say I’m still totally off)
    Their is a good saying that if you go to a party and 3 people tell you your drunk, your probably drunk.

  23. Canadian Couch Potato January 29, 2015 at 8:47 pm #

    @John R: No worries, hope you didn’t feel people were piling on. This is concept is completely counterintuitive, and no one gets it the first time. There are many people in the investment industry who still don’t get it.

  24. pjb January 29, 2015 at 11:43 pm #

    @John r

    “As far as vxus goes I think I got it now. ”

    Yep – you’re right on the money.

  25. oldie January 30, 2015 at 1:06 am #

    @pjb: with all due respect, when you say

    “I think that if you can find any financial writer anywhere who thinks that it’s a good idea to dump VXUS/VTI now and shift the money back to Canada, you should go for it.”

    I think I know what you mean. But as Dan says

    “There are many people in the investment industry who still don’t get it.”

    and indeed I think he’s right, not only on this question, but also on other issues that you’d think they ought to understand better. So your hope that the collective wisdom of the financial industry’s advisers and writers will always back up your opinion is not one you should count on, and it’s not because you’re wrong. There’s a lot of stupidity out there, and what’s frightening is that these stupid (sorry, misinformed) people are sometimes making a living giving financial advice and writing financial articles in newspapers.

  26. Steve January 30, 2015 at 3:34 am #

    @ pjb & oldie:

    >> “I think that if you can find any financial writer anywhere who thinks that it’s a good idea to dump VXUS/VTI now and shift the money back to Canada, you should go for it.”

    Well, this is actually a very interesting question …

    As a general point, investors choose their asset allocation as a function of their unique financial goals, available resources, and constraints – and as these variables change over time, naturally the asset allocation may change as well.

    The particular case of portfolio currency exposure is just a subset of that general point. Imagine:

    Someone moving overseas for an extended work assignment where pay and expenses are in local currency; someone receiving a bequest of valuable property located in a foreign country; someone changing retirement residency plans from one country to another …

    In all these cases, a review and possible change of portfolio currency exposure might be warranted, and would probably cause no controversy. But these examples are all at the “micro” level – are there ever any “macro” reasons for a buy-and-hold indexer to review and possibly change their currency exposure?

    Imagine we were back in “northern peso” territory, with CAD trading at USD 0.62 – if someone was starting out investing at that point, would that situation justify a different portfolio currency allocation than would be the case with CAD trading at par with USD? After all, the magnitude and directionality of the currency risks in those 2 cases are very different.

    Because if those 2 cases would justify a different portfolio currency allocation for someone starting out … then why wouldn’t they also justify an already established investor changing their portfolio currency allocation midstream?

    In other words, if CAD again fell to 0.62 US from par – should investors respect ideological purity (and usual good sense) by staying long USD in their portfolio, or would they be justified in committing the couch potato heresy of changing their portfolio currency allocation in response to market conditions by reducing USD exposure (remember, in effect reducing USD exposure in that case amounts to selling USD high and buying CAD low, a Good Thing™)?

    Whether this argument applies to CAD at 0.79 USD is your call.

    Just a thought.

  27. John r January 30, 2015 at 7:01 am #

    @steve

    Regarding Us dollar exposure. That was really my original question. (I have a lot more allocation in vti than in vxus.) Vxus just seemed to take over the discussion.
    I bought the us dollar a few years ago at par thinking par was a good time to buy the currency. I then thought why not invest it in vti hopefully getting a double whammy with hopefully the us markets doing well and the Canadian dollar falling. (I know this is kind of gambling but in the midst of doing this is when I discovered couch potato)
    That was really my original question is should I sell my vti take the double whammy gain and invest back into the us market with a hedged etf (say xsp.ca) to the cdn dollar. This way I at least lock in the currency gain for now.
    By the way I think I’m just going to stick to the couch potato way and rebalance taking into account the curency gain.

  28. oldie January 30, 2015 at 12:27 pm #

    @John r:

    “By the way I think I’m just going to stick to the couch potato way and rebalance taking into account the curency gain.”

    If “by taking into account the currency gain” you merely mean a straightforward recalculation of your current asset allocations in Canadian Dollar values and rebalancing according to whatever asset class has strayed too far from your original specification for comfort, that is the standard “Couch Potato” protocol, which according to my current understanding, does not allow for exceptions or “hedging” behaviour at times of emotionally disturbing currency gyrations such as we are experiencing now.

  29. Richard January 30, 2015 at 1:09 pm #

    @John r: I understand when you say you would hate to see the dollar go back to par and miss that gain. But the truth is that as investors we miss 90% of those opportunities because no one knows the future. Sticking to a solid plan is a great way to make sure we don’t miss the other 10% too.

    If I really have a strong feeling about something I might shift my regular assets by a few percent or put less than 1% into something I don’t normally buy (if there’s a chance to multiply that).

    If you really think the exchange rates will reverse course, betting a large portion of your portfolio on that is very risky. It’s up to you and you could be right — but what if it causes you to miss out on another 20% gain because you were hedged? Would you be kicking yourself for making that move?

    Even if you do know what’s going to happen, the only way to profit from it is to guess the start and end dates of that change accurately. That kind of market timing is even harder.

    If you don’t actually know the future the next best thing is to stay diversified. With about 2/3 of my portfolio outside of Canada and unhedged I’ve seen the value going up pretty quickly in the last few weeks (since I adjust the value of each ETF for the exchange rate if it’s traded in the US). It might go back down a bit by the end of the year but over the long term I trust that I’ll do ok.

    Either way normal rebalancing is pushing me to put a bit more into the Canadian market now. And since a lot of the things we buy are imported from other countries (or at least priced on a global market), having your portfolio be 100% hedged to Canadian dollars actually increases the long-term risk that your expenses will rise while your portfolio stays flat or loses ground.

    For now let’s just enjoy our rising portfolios and low gas prices 🙂

  30. Steve January 30, 2015 at 3:26 pm #

    @oldie:

    Aw c’mon – we all know and follow “standard couch potato protocol” (at least I do, have done exclusively for 15 years now!), but simply calling something correct doesn’t make it so.

    Portfolio asset allocation – including foreign currency exposures – is first and foremost about managing risk.

    For example, as other comments have noted, as Canadians we import much of what we consume, so some foreign currency exposure in our portfolios serves to mitigate the risk of rising prices. Many other examples of currency exposure as mitigation of risk could be provided.

    I gave an example of an extreme change in portfolio risk profile caused by a massive currency swing – one that really happened within recent memory – and for the sake of discussion I argued the case for deviating from “standard couch potato protocol” in the interest of better managing very different new portfolio risks.

    So make your case – an asset allocation made with CAD/ USD at par is now facing CAD trading at 0.62 USD … How is portfolio risk management better served by “staying the course” than by accounting for the new risks?

    Believe me, I want to be convinced 😉

  31. Richard January 31, 2015 at 5:32 pm #

    @Steve even in extremes, guessing the turning point is difficult.

    For example it’s hard to argue that bonds have prospects for good returns in the short or medium term at this point. There are just too many factors working against them and this is a clear example of an extreme where a lot of investors would be comfortable betting against them just as they would want to bet on the loonie rising from historic lows (which it’s not at right now).

    And yet the boring old Canadian bond index has returned 10% in the last 12 months with an 8% gain this month alone. Investors who otherwise should hold bonds but decided to time the market just lost out.

    No matter what I think I will never bet a large portion of my portfolio on any specific view because the risks of getting it wrong are too big. If you’re talking about moving 5 – 10% of your portfolio around that’s less risky but I wouldn’t do it unless I’m ok with the outcome if I get it wrong because that is very likely.

    To answer your question with another question — what makes the certainty that the CAD will rise from a low point different from the certainty that “house prices will always go up”? The more I learn the more I see that I don’t know that much. Diversifying is the safest guess.

  32. Steve January 31, 2015 at 6:44 pm #

    @Richard:

    Thank you for your reply.

    >> “even in extremes, guessing the turning point is difficult.”

    To be clear – I’m not talking about forecasting anything.

    In my example, the magnitude and directionality of the currency risk faced by a portfolio with sizeable exposure to USD will be very different if CAD is trading at par, or trading at USD 0.62 – that is not forecasting the “future”, it is probabilities in the “present”.

    >> “No matter what I think I will never bet a large portion of my portfolio on any specific view because the risks of getting it wrong are too big.”

    If CAD is trading at USD 0.62, wouldn’t sticking to an asset (currency) allocation made under very different conditions (CAD at par with USD) effectively be betting on a specific view? In addition, if the main point of asset allocation is to manage risk – and I think it is – wouldn’t that imply accepting greater market risk than necessary? That sounds like a failure of risk management.

    But you’re getting very close to my view, why I “buy-and-hold” despite the questions I’m asking here, when you said “the risks of getting it wrong” …

    >> “Diversifying is the safest guess.”

    Adopting a “couch potato” strategy implies agreement with that view – but it still begs the question(s) of when – or if – to change one’s diversification – and more importantly, why (not)?

  33. Eric February 12, 2015 at 9:46 pm #

    This is probably a dumb question, but I noticed that my VXUS returns are dramatically different than my holdings of 1/3 VEE and 2/3 VDU over the past year – is this b/c of the rise of the U.S dollar relative to all currencies?

    I hold all 3, b/c I started my couch potato portfolio in 2012, so only VXUS was available then. I still hold VXUS, but now have been purchasing VEE and VDU over the past year for my international exposure.

    Thanks in advance

    Eric

  34. Canadian Couch Potato February 13, 2015 at 8:36 am #

    @Eric: It’s not a dumb question at all. The reason is that VXUS reports its returns in US dollars. In Canadian dollar terms it would have performed similarly to a blend of VDU and VEE.
    http://canadiancouchpotato.com/2014/12/05/decoding-international-equity-etf-returns/
    http://canadiancouchpotato.com/2013/01/07/calculating-foreign-returns-in-canadian-dollars/

  35. Roman July 17, 2015 at 10:30 am #

    As VDU invests in the US fund VEA, does this mean that VDU has US dollar currency exposure? If so, as the fund holds ex North America stocks and if I don’t want US dollar exposure, wouldn’t it be better to invest in an ETF similar to VDU that does not have US dollar exposure?
    Basically, why should I hold an ex North America ETF that is tied to the US dollar?

  36. Canadian Couch Potato July 17, 2015 at 10:46 am #

    @Roman: A great question, and a common source of confusion. VDU has no US dollar exposure:
    http://canadiancouchpotato.com/2014/01/16/currency-exposure-in-international-equity-etfs/

  37. Roman July 17, 2015 at 11:20 am #

    @CCP thank you for pointing me to the right post!

  38. Shaun February 1, 2016 at 11:48 pm #

    @CCP
    Thx again for the quick reply. Sorry my last post here: http://canadiancouchpotato.com/2014/03/06/why-currency-hedging-doesnt-work-in-canada/ , was intended for this page. I had both open simultaneously.

    I checked ZEM to see if it also lagged similar to XEM to exclude the “wrap” effect. ZEM started in October 2009. It has lagged by 1.23% annualized since inception. This equates to a very similar 0.77% annualized drag greater than MER despite owning the underlying securities directly.

    Regardless of the precise reasons, the fact that these drags in the order of 1.3% total for all the ETF providers (vanguard is also similar), coupled with the fact that Canadian Equities have exhibited a high correlation with emerging markets, has me questioning the merits of including emerging markets in a Canadian portfolio. It seems to me the far lower total cost associated owning Canadian Equities would very likely outweigh the diversification benefit of including emerging markets. It seems likely to me that these emerging market tracking errors would proportionally drag the all world ETFs like VXC as well.

    To clarify, your comment “The index returns we use for the international and emerging markets equities are already reported net of foreign withholding taxes (at least since 2000) so this accounted for already.”

    Does that mean the MSCI indices you are using to back calculate your model portfolios are different that the MSCI benchmarks that are displayed on the Blackrock and BMO websites? They do not specify if the benchmark they compare to is net or gross of withholding taxes.

    Cheers

  39. Canadian Couch Potato February 2, 2016 at 8:12 am #

    @Shaun: I am not sure which benchmarks the ETF providers are using: it might say in the prospectus.

Leave a Reply