There is evidence that tilting portfolios towards certain factors at play in South African financial markets can harness good returns at low costs, but you need to be smart about how you choose a so-called “smart beta” investment strategy.
This is the view of Shaun Levitan, the chief operating officer and co-founder of investment management company Colourfield, who presented the company’s research into what is known as factor-based investing to the recent annual conference of the Actuarial Society of South Africa.
Factor-based investing is often referred to as smart beta investing. “Smart beta” is used broadly to describe any investment that is neither actively managed nor passively tracks an index made up of shares according to their market capitalisation, such as the FTSE/JSE All Share Index (Alsi). (Market capitalisation is the share price multiplied by the number of shares in issue.)
Levitan says about 95 percent of South African investments are actively managed, but more actively managed funds underperform indices (weighted according to the market weighting of shares) than outperform them.
The recent Standard & Poor’s Indices versus Active Funds scorecard shows that more than three-quarters of actively managed South African equity funds underperformed certain benchmark indices over one, three and five years to the end of June this year (See “The Spiva scorecard”, below).
If you fear choosing an active fund manager who is unable to outperform a market-capitalisation index after fees, factor-based investing in smart beta passive funds is an alternative worth considering, Levitan says.
Research over long periods has shown that you will be better compensated for taking the risk of being more exposed to shares with certain characteristics than being invested in a market-weighted index such as the Alsi.
The characteristics that have been identified in research conducted around the world include the size of a company (in other words, large-cap or small-cap shares), value and profitability.
Colourfield has researched the factors that deliver returns in South African financial markets over the past 20 years. Levitan says this research shows that:
• A portfolio that is tilted to small-cap shares has outperformed the market by 0.9 percentage points a year;
• Tilting a portfolio towards shares with value characteristics rather than growth characteristics has returned three percentage points a year more; and
• Being exposed to shares of companies with high profitability has delivered one percentage point a year more.
Levitan says low volatility is often identified as a factor that should deliver good returns and is being offered by some smart beta funds in South Africa.
However, he says, international market data shows that, from 1970 to 2015, shares exhibiting low volatility also had a value bias.
Similarly, funds that track the Research Affiliates Fundamental Index (Rafi) have had a very high value bias over time. This is because the Rafi weights shares according to a company’s book value, sales, cash flows and dividends, Levitan says.
However, the Rafi ignores share price in its construction, which is counter-intuitive, as the return achieved is dependent on the price paid, he says.
Levitan cautions against an investment strategy using momentum as a factor, because the “premium” you expect to get from momentum shares does not last long. This is because the market allocates higher and higher prices to a momentum share and a fund manager needs to buy it at the lower price and sell at a higher price to earn a good return. You need to invest and then sell quickly to earn this premium, Levitan says. This can create a high turnover of shares, which results in high trading costs, he says.
Levitan says it is possible to use factor investing to create portfolios with better certainty of returns and lower volatility than those offered by active fund managers, but you or your adviser need to take care when you choose a factor-based or smart beta investment that it is well configured.
He says you or your adviser also need to pay attention to how the factors are integrated, taking into account any inter-relationships between them. It is no good putting together, for example, three different smart beta funds only to find you have created the same effect as owning the market by tracking a market-capitalisation index.
Finally, Levitan says, the investment provider needs to take into account the trading costs.
A factor tilt, such as one towards small-cap shares, may appear amazing on paper, but when a fund manager comes to buying and selling these stocks, they may be difficult to trade because they are illiquid – in other words, there are so few of them in circulation. This is particularly true of the 50 smallest stocks on the JSE.
In order to buy these shares in the quantity it needs, a manager may destroy, rather than create, value, Levitan says.
Investments that track market-capitalisation indices have to trade shares quickly, and active fund managers also have to act fairly quickly on their research. This can increase trading costs. But these costs can be contained in a portfolio that takes more time to invest in shares with the right characteristics, Levitan says.
Levitan says trading costs are also an issue in funds that equally weighted shares, such as those that offer equally-weighted exposure of 2.5 percent to each of the shares in the FTSE/JSE Top 40 Index.
These strategies appear to work because essentially they tilt the portfolio away from the very large shares by down-weighting them. But Levitan says that keeping the portfolio allocations at 2.5 percent per share requires a lot of rebalancing, and this makes the trading costs high.
And you shouldn’t expect a smart beta investment that tracks shares with a certain characteristic to deliver good returns every year – you need to be invested for a meaningful period – at least three years, he says.
At times, a factor may not deliver for a very long time – beginning in 1981, the small-cap sector in the US failed to deliver good returns for seven years.
Factor-based investing means that, in order for an active fund manager to earn what is known as alpha, the manager needs not only to outperform an index, but an index adjusted for the tilts that are represented in that manager’s style.
Levitan says renowned US economists Eugene Fama and Kevin French found that less than three percent of active managers in the US deliver better than factor- adjusted indices.
THE SPIVA SCORECARD
The Standard & Poors (S&P) Indices Versus Active (Spiva) scorecard measures the performance of actively managed South African equity and fixed-income funds denominated in rands against certain benchmark indices over one-, three-, and five-year investment horizons.
The Spiva scorecard to the end of June this year shows that:
• 86.3 percent of actively managed funds underperformed the S&P South Africa Domestic Shareholder Weighted Index (SADSW) over five years. South African equity funds returned 12.4 percent a year, on average, while the S&P SADSW returned 15.8 percent a year; and
• 85.0 percent of actively managed funds underperformed the S&P SADSW over three years. South African equity funds returned an average of 11.9 percent a year, while the S&P SADSW returned 15.1 percent.
The Spiva also tracks the performance of global equity funds available in rands against the S&P Global 1200, a composite of seven major indices, including the S&P 500 (US), S&P Europe 350 (Europe), S&P/TOPIX 150 (Japan), S&P/TSX 60 (Canada), S&P/ASX All Australian 50 (Australia), S&P Asia 50 (Asia Ex-Japan), and S&P Latin America 40 (Latin America).
In their analysis to the end of June this year, Spiva researchers found that:
• 96.4 percent of actively managed funds underperformed the S&P Global 1200 over five years. Global equity funds returned, on average, 19.6 percent a year; the S&P Global 1200 returned 24.8 percent; and
• 96.6 percent of funds underperformed the S&P Global 1200 over three years. Global equity funds returned 15.9 percent a year, on average; the S&P Global 1200 returned 22.3 percent a year.