[Note: This post was originally published on January 6, 2010. I’ve updated it with the data for the past two years on the performance of the iShares MSCI EAFE CAD-Hedged Index Fund (TSX: XIN) relative to MSCI EAFE local currency returns. Bottom line: Though the performance lag of the past two years was slight, the annualized lag for the past six years is still significant due to the large performance drag observed in 2009.]

I’ve looked at the tracking error of S&P 500 currency-neutral funds in past years but the tracking errors in the iShares CDN MSCI EAFE 100% Hedged to CAD Dollars Index (TSX: XIN) remained a mystery because I didn’t have the annual return data for the MSCI EAFE Index* in local currency. XIN holds the iShares MSCI EAFE Index fund (NYSE Arca: EFA) and hedges the foreign currency exposure that EFA’s holdings are denominated in, so that the returns of stocks will not be impacted by changes in the exchange rates between Canadian Dollars and Yen, Pound, Euros and other currencies. (As an aside note that even though EFA trades in the US, Canadian investors holding EFA are not affected by fluctuations in the exchange rate between the CAD and USD but are exposed to the fluctuations between the CAD and a basket of currencies such as Yen, Pound, Euros etc.).

Fortunately, MSCI Barra reports the returns of MSCI EAFE and other MSCI indices in local currencies on their website. Armed with that data, we can look at how well XIN tracks the MSCI EAFE in local currency terms. The following table shows the annual total returns of MSCI EAFE Index in its local currencies (column 2) with XIN (column 3). The results are consistent with our earlier analysis of the tracking error of XSP. XIN shows an annualized tracking error of 1.30%, which is lower than the tracking error shown by XSP but still wide enough to suggest that currency hedging is highly likely to be unprofitable.

 Year   Local Currency   XIN   Difference 
2011 -12.15% -12.71% 0.56%
2010 4.82% 4.59% 0.23%
2009 24.72% 18.11% 6.61%
2008 -40.27% -40.58% 0.31%
2007 3.54% 1.96% 1.58%
2006 16.46% 16.75% -0.29%

 
When a Canadian investor holds a foreign investment directly, they take on the risk that currency fluctuations will affect their returns. Sometimes, the fluctuations will be in the investor’s favour. Other times, as Canadian investors directly holding US securities can readily attest to, fluctuations will hurt returns. Canadian investors in the iShares MSCI EAFE Index Fund (EFA) would have experienced a significant boost from the currency effect. In local currency terms, the MSCI EAFE Index lost 17.3% over the 2006 to 2011 period. Since investors in XIN trailed the index by an annualized 1.30%, XIN’s loss over the same six year period is 23.73%. However, a Canadian investor holding EFA directly would have a loss of 13.91% over the same time period.

The verdict on currency-hedging then (based on an admittedly short history of just 6 years) is clear: Long-term investors are highly unlikely to profit from hedging their currency exposure because currency effects have to overcome significantly large tracking errors simply to break even. When currency effects are negative (as it was the case of the CAD/USD and US markets over 2006 to 2011), currency-hedging still did not show a profit due to tracking error. With positive currency effects (as was the case with CAD/basket and EAFE index over 2006 to 2011), currency-hedged investors are trailing even more because investors did not get the currency boost and paid for their hedging efforts through tracking error.

* – MSCI EAFE Index tracks stock markets in Europe, Australasia and Far East and holds securities that trade in countries such as Japan, the UK and Germany.

This article has 15 comments

  1. Agree with your conclusions… it’s not worth the effort for long-term investors to hedge their currency exposure.

  2. It seems that comparing tracking error of index ETFs to their MERs is a good way to see if there are any hidden drags on returns.

  3. This has been a great series CC. Thanks for doing the hard work.

  4. CC, were you a superb detective in your former life? !!!

    Excellent!

  5. I just set up my couch potato 1 month ago (XIN, XBB, XSP, XIC) I did this over the ETF as I would be making once yearly contributions inside my TFSA – with this in mind. What would be a better alternate long term to XIN?

  6. EDIT. – doing three things at once ! probably three to many !

    I just set up my couch potato 1 month ago (XIN, XBB, XSP, XIC) I did this AS AN ETF as I would be making MORE than once yearly contributions – inside my TFSA – with this in mind. What would be a better alternate long term to XIN?

  7. Why pick a 4 year period? Running the same numbers (using TD e-Series funds, though) over a 3 year period shows mixed results: US Currency Neutral does better than the non-hedged version whereas the International Currency Neutral does worse than the non-hedged version.

    However, this discussion is irrelevant. I don’t buy into the assumption that currency fluctuations even out in the long term. Therefore, currency neutral funds are useful as they hedge out this long-term risk. I agree, however, that the hedging is likely more expensive than it has to be.

    Slightly off topic: I’m not sure where I recall people complaining about emerging market funds that hedge the C$->US$ risk. While at first glance this seems to be a silly thing to do, keep in mind that most LARGE emerging market companies (the ones the emerging market funds hold) do a lot of business in the US in US$. Therefore, there still exists a significant US$ risk despite the company reporting in local currencies.

  8. Another way to look at the math based on the numbers above, if you invested 1 CAD in both EFA and XIN, after the 4 year period, you would have 0.943 CAD if you invested in EFA and 0.832 CAD if you invested in XIN. It looks like XIN removes almost all of the effect of exchange rate variation over 1 year and some of it over 4 years. Comparing it to XSP VS IVV, XSP appears to remove almost no effects of exchange rate variation over 4 years (passes it on to the investor). It’s probably best to buy IVV or EFA if you are planning to hold it longer than 1 or 2 years.

  9. Pingback: The Costs of Currency Hedging | Canadian Capitalist

  10. Pingback: Why currency-hedged funds have large tracking errors | MoneySense

  11. Pingback: Currency-Hedged Funds Underperformed in 2010 | MoneySense

  12. I am setting up my coach potato portfolio but I need some advice regarding the international index. What is the difference between the Altamira international index fund (which invests in securities and derivatives based on international indexes) and the RBC international index fund currency neutral (which actually tracks the MSCI EAFE index but with currency hedging). Is the Altamira international index fund actually tracking the MSCI EAFE index ?..what would be the best alternative for the long term investor ?

  13. The author’s numbers are suspect. He claims that Cdn investors benefited from holding US investments in the period, but the Loonie increased in value by 13%. http://stockcharts.com/freecharts/perf.html?$CDW Any US holdings would have LOST 12% (100/113-1) just because of the exchange rate.

    The objective of the hedged i-Shares product is to hedge the US/Loonie exchange, not hedge the local-currency/Loonie, It tries to replicate the returns of the US product. The only relevant comparison is between the XIN and the EFA. Any difference is the ‘cost of hedging US/Loonie currencies. You compare that cost, to the cost you would personally incur to do the same hedge yourself.

    The benefits of hedging depend on the movements of the currencies. You cannot conclude from past FX changes that future FX changes will be in the same direction.

    There is no practical way for Canadians to replicate the EAFE results measured in local-currencies.

  14. @Chris: The MSCI EAFE Hedged to CAD Index does *not* hedge the USD/CAD exposure because its fluctuations are immaterial to a Canadian investor. It hedges away the exposure of the currencies the components are denominated in.

    http://www.msci.com/eqb/methodology/meth_docs/MSCI_Hedged_FX_Hedge_Global_Currency_Methodology_May11.pdf

  15. Yes, I double checked and you are correct. Sorry. This question was much discussed at the time of the 2005 switch to hedging, and at that time it was my memory that all the documentation indicated the US/Loonie hedge. But the MSCI people’s returns in local currency do match the ETF’s returns more closely.