Tracking Errors on Canadian ETFs

April 26, 2011

An index fund’s tracking error is the difference between the return of the fund and the return of its benchmark index. It’s an important measure of how well your index funds are performing, and it’s worth keeping an eye on. The table below shows the tracking error of Canadian equity ETFs and index funds in 2010.

As with many things in investing, raw numbers don’t tell the whole story. Here’s some guidance to help you understand the figures:

  • I’ve grouped the funds using some broad categories, but they don’t all track the same indexes, so it may be misleading to compare their absolute returns. The point here is to look at how well each fund tracked its own benchmark, not necessarily to compare funds that use different strategies within the same asset class.
  • When a fund has an unusually large tracking error, there may be a specific reason, such as change to the index during the year. If you’re alarmed by the performance of one of your ETFs or index funds, look in the Management Report of Fund Performance (available from the SEDAR website). You may find an explanation under the heading “Recent Developments.”
  • Funds report their returns in different ways. iShares, Horizons and BMO report ETF returns according to changes in net asset value (NAV), while Claymore reports them according to both NAV and market price. In order to make the comparisons fair, the returns below are all based on NAV.
  • Some funds report their returns to two decimal places, while others report only one. The figures below have been rounded off to one decimal place.

To further help put these numbers in context, I have included several footnotes below the table.

Fund Index Tracking
Broad Market Ticker return return error
iShares S&P/TSX Capped Composite XIC 17.3% 17.6% -0.4%
TD Canadian Index – e-Series TDB900 17.2% 17.6% -0.4%
RBC Canadian Index RBF556 16.8% 17.6% -0.8%
Large Cap
iShares S&P/TSX 60 XIU 13.6% 13.8% -0.2%
Altamira Canadian Index NBC814 13.1% 13.8% -0.7%
BMO Dow Jones Canada Titans 60 ZCN 13.6% 13.9% -0.2%
Claymore Canadian Fundamental CRQ 13.7% 14.6% -0.9%
Horizons S&P/TSX 60 HXT 9.3% 9.3% 0.0% (1)
Small Cap
iShares S&P/TSX SmallCap XCS 34.4% 35.1% -0.7%
Dividend
iShares DJ Canada Select Dividend XDV 12.8% 13.3% -0.5%
Claymore S&P/TSX Cdn Dividend CDZ 15.2% 17.8% -2.6% (2)
Claymore S&P/TSX Cdn Pref Share CPD 6.6% 7.7% -1.1% (3)
Real Estate
iShares S&P/TSX Capped REIT XRE 21.9% 22.6% -0.7%
BMO Equal Weight REITs ZRE 17.9% 18.8% -0.8% (4)

Notes:

1. The Horizons S&P/TSX 60 ETF (HXT) was launched on September 14, 2010, so this tracking error covers only three and a half months. However, the structure of HXT, which uses a swap to deliver the total return of the S&P/TSX 60 without paying distributions, should guarantee  that the fund’s tracking error will not exceed its 0.08% MER.

2. CDZ outperformed XDV in 2010, but with a much larger tracking error. This is a good example of why you should look beyond absolute performance numbers when comparing ETFs. The index tracked by CDZ, which is based on dividend growth, may be harder to track than XDV’s more straightforward benchmark.

3. CPD’s return based on market price was 7.1% — some 50 basis points higher than its return based on net asset value. CDZ also returned almost half a point more on market price than on NAV. Both ETFs seem to have traded at a premium in 2010, probably because of the huge popularity of dividend funds.

4. BMO’s real estate ETF launched on May 19, 2010, so these returns should not be compared with those of its iShares counterpart.

 

1. The Horizons S&P/TSX 60 ETF (HXT) was launched on September 14, 2010, so this tracking error only covers three-and-a-half months. The structure of HXT, which uses a tax-friendly swap to deliver the total return of the S&P/TSX 60 without paying distributions, should guarantee that the fund’s tracking error will never exceed its 0.08% MER.

 

2. CDZ outperformed XDV in 2010, but with a much larger tracking error. This is a good example of why should look beyond absolute performance numbers when comparing ETFs. The index tracked by CDZ, which is based on dividend growth, appears to be harder to track than XDV’s more straightforward benchmark.

 

3. CPD’s return based on market price was 7.1% — some 50 basis points higher than its return based on net asset value. CDZ also returned almost half a point more on market price than on NAV. Both ETFs seem to have traded at a premium in 2010, probably because of the huge popularity of dividend funds.

 

4. BMO’s real estate ETF launched on May 19, 2010, so these returns should not be compared with those of its iShares counterpart.

{ 3 comments… read them below or add one }

Patrick April 26, 2011 at 7:14 am

This is great stuff Dan. I’m amazed at how much I continually learn from this blog on the seemingly simple topic of index investing.

Preet April 26, 2011 at 9:29 am

Good post Dan.

Question: I believe there is an academic argument for tracking error being an absolute number, but does it make sense to indicate all the above tracking errors with a negative sign to indicate the fund returns tracking error resulted in underperformance?

While no funds outperformed their benchmarks, it is conceivable in future editions that an index fund outperforms the benchmark. Positive tracking error can certainly exist, especially with smaller funds and funds that re-sample the index constituents.

Canadian Couch Potato April 26, 2011 at 9:41 am

@Preet: Good point — I agree it’s more correct to indicate tracking error with a negative number and changed the table above accordingly. As you say, there are a few instances where ETFs have outperformed their benchmarks: Vanguard’s VWO was one of them last year.

Leave a Comment

Previous post:

Next post: