Why the benchmark shouldn't matter
Professional fund managers may often be heard to say that they are “not benchmark cognisant”. This group of high conviction investors typically charges a premium fee to undertake active management of an investment portfolio. To avoid being labelled as “index closet huggers” or frauds, active fund managers should construct portfolios that differ materially from the composition of a market index.
Active investment management involves a careful analysis of the economic environment and scenario planning as to the likely future outcomes (“top down” in investment jargon), as well as assiduous fundamental analysis of the individual securities in which investments can be made (“bottom up” in investment parlance). The composition of the market index is simply not a factor in this process. In fact, the only purpose served by the benchmark is to measure the benefits of this active approach over investing passively in the market index.
The discussion as to the merits of active and passive management rages on. The debate is predicated in the main on cynicism regarding the persistence of the skill of an active manager. However, there are two issues that cannot be overlooked. The first is the existence of active managers who have a demonstrable long-term record of success in beating a market index. The second is the inherent risks of passively investing in a market index.
There are a handful of investment houses in South Africa that have performance track records dating back multiple decades that prove that active management can work in share markets that are not fully efficient, like the exchange operated by the JSE Limited. In this context, efficiency speaks to the degree to which share prices reflect all available and relevant information. In theory, less efficient markets will reward high conviction investors that conduct thorough research with superior returns. The track records of the successful fund managers are typically both very public and verifiable. The track records of the less successful fund managers are often carefully worked out of the public domain.
The second important issue dealing with the construction of the benchmark is the main subject of this article and will be dealt with in more detail. The benchmark for most general equity funds in South Africa is the FTSE/JSE All Share Index (“ALSI”). Given that this might be construed as the default option for a passive investor, careful consideration should be given to the precise composition of that index and its merits. As will be demonstrated below, the index labours under two quite profound investment risks, namely “concentration” and “the size/momentum effect”.
Although there are variations on the theme, a market index is most often market value-weighted and the ALSI is no different. Simply stated, the market value of the exchange is the sum of the market values of the component shares (that is, the number of shares multiplied by the price per share for each listed company).It follows that very large companies which trade at high prices will enjoy considerable significance in a market value-weighted index.
The dominance of an index by large market capitalisation companies leads to considerable concentration (alternatively, lack of diversification) inherent in the index portfolio or benchmark. As an illustration, at the end of June 2014, the ALSI’s top five constituents (out of the total index composition of 165 shares) comprised more than 41% of the total index capitalisation. Four of the top five shares, amounting to 35% of the index, are rand hedge counters. This means that they are materially exposed to movements in the rate of exchange of the South African rand. Ten years ago, the top five shares of the index amounted to a similar 40% of the index. However, all five companies were rand hedge stocks and three were resource counters. While the rand has experienced a weakening bias over the past few decades, there have been extended periods of rand strength which were negative for rand hedge counters.
As has been described already, the weight of a constituent in the index is partly a factor of the price of the share. It follows that if a share price has increased relatively more than others, all things being equal, that share will enjoy a greater weight in the index. Conversely, if a share price has fallen relatively more than the others, then that share’s weight in the index will be reduced. Hence, the basis for one’s exposure to a particular share in a passive portfolio is not its inherent value (a forward-looking concept), but rather the momentum of the price (a backward-looking metric).
A passive investor is therefore increasing his or her exposure to a particular counter merely by virtue of its price movement. This does not provide the investor with the opportunity to ascertain whether that price is in excess of value (if it is overvalued, it would rationally require a reduction in the holding, rather than an increase).In contrast, where a share price weakens relative to others, the holding’s weight is reduced by default, and the investor has no opportunity to increase a holding (if the fundamentals indicate this to be appropriate) of a counter whose value is in excess of its price.
Market value-weighted indices therefore encourage behaviour akin to buying high and selling low. Even the least astute of investors will understand this to be irrational investment behaviour. It would be remiss not to note that there are new age or "smarter" indices that attempt to address the concentration and factor biases of market weighted indices. While these smarter indices improve these risks to some extent, the rules are mechanical and arguably arbitrary.
In contrast to this size/momentum effect, active fund managers typically build small portfolios of high conviction ideas that diversify risk suitably. This, after all, is what they are paid to do. If a high conviction investor attributes a particularly high weight to a constituent of the portfolio, it is certainly not going to be merely because the company is large or because the share price has moved up. Rather, a concentrated position in an active portfolio will be grounded in the firm conviction that the company is fundamentally well position for earnings growth. The weighting of that position in the portfolio will serve to benefit the portfolio from that future growth, while diversifying other investment risks and managing the risk that the fund manager’s assessment is incorrect. The composition of the main benchmark index therefore does not matter to the high-conviction active fund manager.
