From The Globe and Mail ē Gordon Pape ē Tuesday, November 25, 2003
The Great Indexing Debate, which was relatively muted during the bear market, is back again in full force. As management expense ratios (MERs) for actively managed funds continue to escalate, more people are looking to low-cost indexing alternatives. The growing popularity of index-based exchange-traded funds (ETFs) such as the iUnits, offered in Canada by Barclays Global Investors, has added fuel to the fire.
Advocates of indexing point out that they can save as much as three percentage points annually by going the route of what is known as "passive investing." For example, the iUnits that track the S&P/TSX 60 Index, which comprises the 60 largest companies on the TSX, has an MER of just 0.17 per cent. Most actively managed Canadian equity funds have an MER somewhere north of 2.5 per cent and many are over 3 per cent.
The MER covers such costs as management fees, brokerage charges, marketing expenses, etc. In most cases, it is deducted from a fundís gross return to arrive at a net figure, which is the gain or loss the individual investor actually sees. So the lower the MER, the better it is for your personal bottom line.
Actually, brokerage fees are not included in fund MERs. That's yet another reason why actively-managed funds, especially "hyper"-actively managed ones, have a greater hurdle to cross than just the MER in competing with index funds.
So why doesnít everyone invest exclusively in index funds and tell the active managers to take a hike? Because if you choose wisely, active managers can outperform the indexes over time.
Easier said than done. Few people, if any, know how to "choose wisely." Let's consider Mr Pape and his colleagues who publish annual mutual fund books. If these experts know how to "choose wisely," then why is it that each one chooses a different set of funds for their "top-10" fund list? And why do they have to change the composition of their lists each year? If these experts can't agree on which funds are the best, then how can the typical mutual fund investor be expected to "choose wisely"?
Not convinced? Consider these numbers, which were generating by using the filter system on Globefund. Out of 425 Canadian equity funds reporting a three-year average annual compound rate of return to Sept. 30, no fewer than 341 (80 per cent) beat the S&P/TSX Total Return Index, which recorded an annual loss of just over 9 per cent during that period. Not one was a broadly based index fund!
Of course, if you carefully pick an index and carefully pick a time period, you can torture your data to "prove" almost anything. Mr Pape omits to mention that this particular period saw shares of Nortel Networks, then representing some 35% of the TSX, drop in price from $124.50 to under a loonie. Active fund managers aren't allowed to have more than 10% of their fund in one stock, so even the most ardent fans were unable to indulge in their infatuation with NT.
However, Globefund isnít perfect in this regard. The problem is that the benchmarks used on the site donít always coincide with the investment objectives of the fund. Sometimes, you need to look more closely.
Dan Hallett, in Benchmarking Problems, carefully summarizes several problems. Keep in mind too, that the Foreign Content problem gave actively managed "Canadian" funds that invested as much as 20% (now 30%) in foreign equities, a large boost in performance due (until recently) merely to the decline of the loonie against other major currencies.
For example, the TD Canadian Index Fund is set up to track the S&P/TSX Composite Index. However, Globefund uses the blue-chip 60 Index for its benchmark, which did not fare as well as the broader index over the past three years. Using the correct benchmark, the regular units of this fund almost matched the three-year return of the Index, although not quite. Over longer terms, however, it has lagged behind the benchmark, although it outperformed the category average over five and 10 years.
The newer ďeĒ units, which have a much lower MER (0.31 per cent versus 0.85 per cent) have fared better. They did slightly better than the benchmark over three years and almost matched it in the latest 12-month period. But, and this is important to note, 331 actively managed funds beat it over three years despite their much higher MERs.
Over longer terms, the majority of actively managed Canadian equity funds failed to beat the benchmark index. For the decade to Sept. 30, a total of 45 out of 114 reporting funds (about 40 per cent) did better than the S&P/TSX Total Return Index. But not one was an index fund!
Why the astonishment? Since index funds have MERs too, it's only reasonable to expect that their performance will trail the benchmark by roughly the MER. What would be truly astonishing would be an index fund that consistently beat its benchmark despite the MER.
The problem with index funds is that managers have nowhere to hide in falling markets. They must stay fully invested, with their weightings matching those of the target index. Active managers can raise their cash positions or move to lower-risk securities. As a result, index funds tend to underperform when stock markets decline. They are at their best when the broad indexes are on the rise. But even then they have trouble beating the benchmarks.
Numerous studies have shown that fund managers have their lowest cash positions at market tops, including when the tech stock bubble burst in 2000, and that actively managed funds fail to outperform index funds even during market declines. But regardless, since equities have more years of positive returns than negative, cash in an actively-managed fund retards its performance on the way up more often than it buffers it on the way down.
So what about exchange-traded funds? How do they stack up? Unfortunately, we donít have much history to work with because only one Canadian ETF has a track record of three years or more. It is the iUnits S&P/TSX 60 Index Fund, and it has done relatively well. The units have beaten the benchmark in all time frames from six months out, going back to their launch in September 1999. That makes them the best bet if you want a fund that will faithfully track the 60 Index.
Looking at the iUnits with shorter histories, they tend to be pretty good at tracking their benchmarks, but the benchmarks themselves may be flawed. For example, iUnits S&P/TSX Capped Gold Fund showed a one-year gain of 27.2 per cent to Oct. 31. Thatís consistent with the S&P/TSX Capped Gold Index itself.
However, Barrick Gold Corp. is the largest single component of that particular index, and it has been a drag on returns because of poor performance. The Globe Precious Metals Peer Index, which is not so heavily weighted to Barrick, gained 61.4 per cent for the year to Oct. 31. The average actively managed gold and precious metals fund was ahead 63.4 per cent during that time. So in this case, the iUnits underperformed despite a low MER of 0.55 per cent.
So the best example that Mr Pape can produce is a 2% difference on a 61% gain?
So is it worth buying index funds or not? Yes, if you donít want to be constantly juggling your portfolio and are willing to stick with them when they slide in bear markets, as they inevitably will. But you need to follow a few basic rules.
But why would the typical investor want to "juggle"? What evidence is there that they can expect to outperform by doing so? There's lots of evidence that they can't.
1) Go for the lowest MER you can find. Presumably, you are not going to trade in and out of these funds. If youíre in for the long haul, a low MER is essential. For mutual funds, look closely at TDís ďeĒ units and at the Altamira Precision funds. Otherwise, go for an ETF if there is one available that tracks the index you want.
2) Diversify, just as you would if you were creating a portfolio of actively managed funds. There are now bond index funds available, as well as iUnits that emulate the returns of five and 10-year Government of Canada bonds. There are also iUnits that track income trusts and REIT indexes. For geographic diversification, you will find plenty of U.S. and international index funds from which to choose. However, if it is international ETFs that you want, youíll have to look to the U.S. exchanges.
3) Check the track record. Look to see how closely the fund has tracked its benchmark index Ė and make sure that the index being used for comparison is indeed the one the fund is designed to emulate.
4) For sector index funds, which includes many of the iUnits series, look closely at the makeup of the underlying index. You may find that it is heavily weighted towards only a few stocks. More than 46 per cent of the iUnits Gold basket of shares is invested in just two stocks: Barrick and Placer Dome Inc. Over half of iUnits Capped Energy is in three securities: EnCana Corp., Petro-Canada, and Suncor Energy Inc. I seriously question whether such weightings represent true indexing.
So do I. And I also question why most investors should want or need to hold sector funds in the first place.
5) If you want to indulge in some tactical asset allocation, consider reducing the weighting of your equity index positions during times of market declines and increasing bond weightings correspondingly.
The time to reduce equity holdings is not "during" but before market declines. There's no evidence that anyone can consistently call market declines and advances before they occur.
Gordon Pape is one of Canada's most respected financial authors and the nation's leading expert on mutual funds. This article first appeared on GlobeinvestorGOLD.com.
Bylo Selhi is a Toronto-based fund industry gadfly and proprietor of a website for independent Canadian mutual fund investors. Bylo rebuts is an occasional series of articles that debunk misguided investment publications.