Is one index really better than another?

Couch Potato investors try to capture the returns of entire asset classes rather selecting individual securities. And we do that with passively managed funds designed to track indexes that represent each of those asset classes. But which index, exactly?

Since Vanguard announced last fall that it would be ending its relationship with the index provider MSCI, the question is worth considering. Vanguard’s Canadian equity ETF, for example, was originally benchmarked to the MSCI Canada Index but now tracks the FTSE Canada Index. Can investors expect the fund to behave differently now?

The short answer is yes, but the full story is more subtle: if the past decade is any guide, the performance of the two indexes will vary from year to year. But the differences are likely to be the result of random noise, and neither of these benchmarks can be said to have any meaningful advantage over the other.

There’s more than one way to measure the market

Investors often refer to “the index” the same way they refer to “the dictionary.” But in both cases, the definite article isn’t really appropriate: there isn’t a single authority. Just as Oxford, Merriam-Webster and American Heritage all define words slightly differently, index providers use various methodologies to define an asset class. And even if they look only at cap-weighted indexes, the final products can vary a lot.

Here’s how MSCI and FTSE describe their Canadian equity benchmarks:

“The MSCI Canada Index is designed to measure the performance of the large and mid cap segments of the Canada market. With 95 constituents, the index covers approximately 85% of the free float-adjusted market capitalization in Canada.”

“The FTSE Canada Index [currently with 78 holdings] is a market-capitalization weighted index representing the performance of Canadian large- and mid-cap stocks. The market-cap-weighted FTSE Canada Index represents about 75% of the Canadian equity market.”

Although the above indexes mention mid-cap stocks as well as large caps, the more popular S&P/TSX 60 is also comparable, since it covers almost the same slice of the market as the FTSE benchmark:

“Designed to represent leading companies in leading industries, the S&P/TSX 60 covers approximately 73% of Canada’s equity market capitalization. Its 60 stocks make it ideal for large cap coverage and a cost-efficient way to achieve Canadian equity exposure.”

Different paths to the same place

You might think all three indexes would show virtually identical performance—and you’d be right if you look at their 10-year track record. But here are the year-by-year returns since 2003. The “Variance” column indicates the percentage-point difference between the best and worst performer in each year.

If you invested for the full 10 years, it would have made no meaningful difference which index you decided to track. But there would have been some significant variance along the way: how about 2.3% basis points underperformance of the S&P/TSX 60 compared with the FTSE benchmark in 2003, or the 2.4% outperformance of the MSCI index over its FTSE counterpart in 2010?

Now imagine you run an actively managed Canadian equity fund and you earn 13.9% in 2010. Did you beat the index? If your fund is benchmarked to the MSCI index you lagged by 100 basis points. But you outperformed the other two. Someone tell the marketing department there’s an idea here.

The takeaway message for investors is simple: short-term performance differences between similar indexes are meaningless. Sometimes the reason is one or two bog movers happened to get more weight in one index than another. It might be as simple as the dates the indexes are rebalanced. Remember this when you’re comparing ETFs tracking the same asset class, whether it’s VCE and XIU in Canadian equities or comparable bond funds such as the BMO Aggregate Bond (ZAG) and the iShares DEX Universe Bond (XBB). If the benchmark indexes are very similar, ignore any differences in returns over one or two years and go with the one that charges the lowest fee.