FACEOFF: The best type of indexing

The benefits of indexing may be obvious to some, but there’s debate about which kind will deliver good performance

By Dean DiSpalatro | November 1, 2011 | Last updated on November 1, 2011
6 min read

  • Joel Greenblatt,

    CO-FOUNDER & CO-CHIEF INVESTMENT OFFICER, GOTHAM ASSET MANAGEMENT, LLC

  • Oliver McMahon,

    HEAD OF PRODUCT DEVELOPMENT, ISHARES CANADA

  • Yves Rebetez,

    MANAGING DIRECTOR AND EDITOR, ETFINSIGHT

  • Som Seif,

    PRESIDENT & CEO, CLAYMORE INVESTMENTS

MARKET-CAP WEIGHTING

McMahon: For most investors looking to get exposure to the Canadian marketplace, it makes sense to have greater exposure to a large firm like Suncor, for example, than to smaller firms like Perpetual Energy and SouthGobi Resources.

Rebalancing needs to be done much less frequently. As the price of a stock appreciates, its price in the index and in the portfolio will go up by the same amount because they’re held in exactly the same weight.

This is the traditional method of indexing funds.

With an equal-weighted portfolio, the portfolio needs to rebalance back to equal weights on a regular basis and these extra transactions can become expensive for the fund and reduce your return considerably.

One of the [tech] funds we have isn’t purely market-cap weighted; [it caps] individual names at 25%. Research In Motion (RIM) is by far the largest technology company in Canada, and if we had an uncapped index, the fund would be made up of 50% RIM stock, so we cap this one stock at a weight of about 25%. Other than times like that, we’re in favour of market-cap weighting across the board.

Seif: The main drawback of market-cap weighted indexes is the structural problem of tying price and weight together. If a company’s price starts to appreciate aggressively versus the rest of the market, you’ll see its weight enhance in the index. So when a company is overvalued (or undervalued) in the market, it’s overweight (or underweight) in the index.

The second problem is market-cap indexes tilt towards higher-growth companies. This is favourable when markets are moving up, but it causes significant challenges when there are bubbles, or when a stock is overvalued. We saw this with Nortel, Cisco, RIM, and gold stocks.

If you believe in reversion to the mean—that over time, overvalued and undervalued stocks will come back to their fair values —then [with market-cap weighting] you’re putting more money into companies that are going to revert down, and less money into companies whose valuations are going to revert up. It’s like buying high and selling low.

EQUAL-WEIGHTED

You have to rebalance periodically to get the portfolio back to equal weighting because stock prices change. That means more trading, and greater expense.

McMahon: There are situations where equal-weighted indexing performs well. An example its proponents like to cite is Nortel, which was up to 30% of the Canadian marketplace at one point. When the price of Nortel fell, investors lost a lot of performance. Of course the opposite can also occur—think of tech company Apple. If you had traded back to equal weights each time the stock went up, you would have underperformed other U.S. indexes by a considerable amount as the stock kept moving higher.

With an equal-weighted product, you’re holding a lower weight in large-cap companies and more small caps, which generally means a more volatile portfolio. If you want exposure to small caps, you should buy a market-cap-weighted small-cap fund.

Seif: Equal weighting tries to delink price and weight—and over time, it has added value.

But there are problems. It’s crude: you’re weighting the smallest company the same as the biggest, which doesn’t make sense. It’s tax-inefficient because you’re constantly rebalancing. You’re also rebalancing smaller names. I’m going to have much bigger spreads and costs when buying equal amounts of the 60th company on the S&P/TSX 60 than I will with Royal Bank.

The returns of equal-weighted indexes outperform the benchmarks over time, but they do it with much greater volatility and risk because there’s a bias towards smaller companies.

Equal weighting makes more sense when you’re dealing with individual sectors like energy than with broad segments of the market.

Rebetez: Equal weighting is useful where the underlying assets have similar drivers and structural attributes. It could be applied to the Big 5 banks and the REIT space in Canada, or the U.S. tech space.

FUNDAMENTAL

This weights companies based on economic size using measures such as sales, book equity value, cash flow and dividends paid. Weights are not affected by market price, so pricing errors are random.

Seif: Fundamental indexes were created with a view to capturing the benefits of indexing while avoiding weighting companies based on market price. Fundamental indexing uses metrics that capture the economic footprint of the company—its fundamentals.

You rebalance once a year, so if a company’s price has appreciated faster than its fundamentals, you rebalance away from it; whereas, if a company’s price has underperformed its fundamentals, you put more money towards it.

The outperformance of fundamental indexing versus traditional benchmark indexing has been outstanding. This performance stems from the delinking of price and weight. Our RAFI (Research Affiliates Fundamental Index) Canadian Fundamental Index product has beaten the benchmark by one point annualized, net of fees, since its inception about five years ago. It has also beaten over 90% of all the active mutual funds in Canada. What RAFI doesn’t do is beat the markets every single day, week, month or year. Over any three-year cycle it outperforms the benchmark 80% of the time.

Rebetez: Fundamental indexing should not only give you better returns than market-cap indexing, but also do so under risk-adjusted parameters that will help you sleep better at night. So it’s worth the little extra you pay for it. But the time horizon is absolutely critical. You should be prepared to stick with this methodology for at least 24 months, though a three-to-five year period is better.

The market will go through phases of momentum, and sticking to the methodology’s fundamentals may make you feel you’re missing out on the wave everyone else is riding—until the wheels fall off. At that point you’ll see it pays to have checks and balances against the madness of the crowds.

Greenblatt: As tested over the last 40+ years, fundamentally weighted indexes can add back the approximately 2% lost each year due to the inefficiencies of market-cap weighting, with the last 20 years adding back even more.

VALUE WEIGHTING

This form heavily weights underpriced, out-of-favour companies, with the expectation that prices will appreciate.

McMahon: Numerous studies indicate the average fund manager doesn’t add value above the fees charged. So if you can get an index that just looks at some of the cheapest stocks in the marketplace, you’re getting the benefits of value investing without the high fees of an active fund manager.

If you’re closer to 50 years old, you may want to make your portfolio less volatile, and invest more in value companies than growth companies. If you’re 25, you can look to exploit growth stocks, riding out the volatility over time. Growth stocks will often perform well over the long term, but can be quite unstable along the way.

If you feel we’re going into another recession, but want to maintain your equity market exposure, you probably don’t want the volatility of a bunch of technology companies, so you probably shouldn’t hold a growth-index product. You’d want a value product. If you think concerns about the years of slow economic growth are overblown, then perhaps you’d want to own a growth portfolio and sell your value portfolio.

Greenblatt: We use two simple factors to create an index that combines ‘good’ and ‘cheap.’ Identifying cheap businesses involves asking how much cash flow we’re getting for the price we’re paying. The greater the cash flow relative to the price paid, the cheaper it is. The ‘good’ criterion relates to return on tangible capital. A business that costs $400,000 to establish and spins out $200,000 in profits per year will have a 50% return on capital.

We rank companies based on how high their return on capital is, and then separately based on how cheap they are. We then combine these two rankings [to] pick companies.

Value-weighted indexes seek not only to avoid the losses due to the inefficiencies of market-cap weighting, but also to add performance by buying more of stocks when they are available at bargain prices.

Value-weighted indexes are continually rebalanced to weight most heavily stocks priced at the largest discount to value. Over time, these indexes can significantly outperform active managers and other types of indexes.

Dean Dispalatro is senior editor of Advisor Group.

Dean DiSpalatro

  • Joel Greenblatt,

    CO-FOUNDER & CO-CHIEF INVESTMENT OFFICER, GOTHAM ASSET MANAGEMENT, LLC

  • Oliver McMahon,

    HEAD OF PRODUCT DEVELOPMENT, ISHARES CANADA

  • Yves Rebetez,

    MANAGING DIRECTOR AND EDITOR, ETFINSIGHT

  • Som Seif,

    PRESIDENT & CEO, CLAYMORE INVESTMENTS

MARKET-CAP WEIGHTING

McMahon: For most investors looking to get exposure to the Canadian marketplace, it makes sense to have greater exposure to a large firm like Suncor, for example, than to smaller firms like Perpetual Energy and SouthGobi Resources.

Rebalancing needs to be done much less frequently. As the price of a stock appreciates, its price in the index and in the portfolio will go up by the same amount because they’re held in exactly the same weight.

This is the traditional method of indexing funds.

With an equal-weighted portfolio, the portfolio needs to rebalance back to equal weights on a regular basis and these extra transactions can become expensive for the fund and reduce your return considerably.

One of the [tech] funds we have isn’t purely market-cap weighted; [it caps] individual names at 25%. Research In Motion (RIM) is by far the largest technology company in Canada, and if we had an uncapped index, the fund would be made up of 50% RIM stock, so we cap this one stock at a weight of about 25%. Other than times like that, we’re in favour of market-cap weighting across the board.

Seif: The main drawback of market-cap weighted indexes is the structural problem of tying price and weight together. If a company’s price starts to appreciate aggressively versus the rest of the market, you’ll see its weight enhance in the index. So when a company is overvalued (or undervalued) in the market, it’s overweight (or underweight) in the index.

The second problem is market-cap indexes tilt towards higher-growth companies. This is favourable when markets are moving up, but it causes significant challenges when there are bubbles, or when a stock is overvalued. We saw this with Nortel, Cisco, RIM, and gold stocks.

If you believe in reversion to the mean—that over time, overvalued and undervalued stocks will come back to their fair values —then [with market-cap weighting] you’re putting more money into companies that are going to revert down, and less money into companies whose valuations are going to revert up. It’s like buying high and selling low.

EQUAL-WEIGHTED

You have to rebalance periodically to get the portfolio back to equal weighting because stock prices change. That means more trading, and greater expense.

McMahon: There are situations where equal-weighted indexing performs well. An example its proponents like to cite is Nortel, which was up to 30% of the Canadian marketplace at one point. When the price of Nortel fell, investors lost a lot of performance. Of course the opposite can also occur—think of tech company Apple. If you had traded back to equal weights each time the stock went up, you would have underperformed other U.S. indexes by a considerable amount as the stock kept moving higher.

With an equal-weighted product, you’re holding a lower weight in large-cap companies and more small caps, which generally means a more volatile portfolio. If you want exposure to small caps, you should buy a market-cap-weighted small-cap fund.

Seif: Equal weighting tries to delink price and weight—and over time, it has added value.

But there are problems. It’s crude: you’re weighting the smallest company the same as the biggest, which doesn’t make sense. It’s tax-inefficient because you’re constantly rebalancing. You’re also rebalancing smaller names. I’m going to have much bigger spreads and costs when buying equal amounts of the 60th company on the S&P/TSX 60 than I will with Royal Bank.

The returns of equal-weighted indexes outperform the benchmarks over time, but they do it with much greater volatility and risk because there’s a bias towards smaller companies.

Equal weighting makes more sense when you’re dealing with individual sectors like energy than with broad segments of the market.

Rebetez: Equal weighting is useful where the underlying assets have similar drivers and structural attributes. It could be applied to the Big 5 banks and the REIT space in Canada, or the U.S. tech space.

FUNDAMENTAL

This weights companies based on economic size using measures such as sales, book equity value, cash flow and dividends paid. Weights are not affected by market price, so pricing errors are random.

Seif: Fundamental indexes were created with a view to capturing the benefits of indexing while avoiding weighting companies based on market price. Fundamental indexing uses metrics that capture the economic footprint of the company—its fundamentals.

You rebalance once a year, so if a company’s price has appreciated faster than its fundamentals, you rebalance away from it; whereas, if a company’s price has underperformed its fundamentals, you put more money towards it.

The outperformance of fundamental indexing versus traditional benchmark indexing has been outstanding. This performance stems from the delinking of price and weight. Our RAFI (Research Affiliates Fundamental Index) Canadian Fundamental Index product has beaten the benchmark by one point annualized, net of fees, since its inception about five years ago. It has also beaten over 90% of all the active mutual funds in Canada. What RAFI doesn’t do is beat the markets every single day, week, month or year. Over any three-year cycle it outperforms the benchmark 80% of the time.

Rebetez: Fundamental indexing should not only give you better returns than market-cap indexing, but also do so under risk-adjusted parameters that will help you sleep better at night. So it’s worth the little extra you pay for it. But the time horizon is absolutely critical. You should be prepared to stick with this methodology for at least 24 months, though a three-to-five year period is better.

The market will go through phases of momentum, and sticking to the methodology’s fundamentals may make you feel you’re missing out on the wave everyone else is riding—until the wheels fall off. At that point you’ll see it pays to have checks and balances against the madness of the crowds.

Greenblatt: As tested over the last 40+ years, fundamentally weighted indexes can add back the approximately 2% lost each year due to the inefficiencies of market-cap weighting, with the last 20 years adding back even more.

VALUE WEIGHTING

This form heavily weights underpriced, out-of-favour companies, with the expectation that prices will appreciate.

McMahon: Numerous studies indicate the average fund manager doesn’t add value above the fees charged. So if you can get an index that just looks at some of the cheapest stocks in the marketplace, you’re getting the benefits of value investing without the high fees of an active fund manager.

If you’re closer to 50 years old, you may want to make your portfolio less volatile, and invest more in value companies than growth companies. If you’re 25, you can look to exploit growth stocks, riding out the volatility over time. Growth stocks will often perform well over the long term, but can be quite unstable along the way.

If you feel we’re going into another recession, but want to maintain your equity market exposure, you probably don’t want the volatility of a bunch of technology companies, so you probably shouldn’t hold a growth-index product. You’d want a value product. If you think concerns about the years of slow economic growth are overblown, then perhaps you’d want to own a growth portfolio and sell your value portfolio.

Greenblatt: We use two simple factors to create an index that combines ‘good’ and ‘cheap.’ Identifying cheap businesses involves asking how much cash flow we’re getting for the price we’re paying. The greater the cash flow relative to the price paid, the cheaper it is. The ‘good’ criterion relates to return on tangible capital. A business that costs $400,000 to establish and spins out $200,000 in profits per year will have a 50% return on capital.

We rank companies based on how high their return on capital is, and then separately based on how cheap they are. We then combine these two rankings [to] pick companies.

Value-weighted indexes seek not only to avoid the losses due to the inefficiencies of market-cap weighting, but also to add performance by buying more of stocks when they are available at bargain prices.

Value-weighted indexes are continually rebalanced to weight most heavily stocks priced at the largest discount to value. Over time, these indexes can significantly outperform active managers and other types of indexes.

Dean Dispalatro is senior editor of Advisor Group.