The index is not enough
When markets were soaring and active managers were struggling, indexation was flavour of the month. Now, with markets in the doldrums, the product getting all the attention is enhanced indexation. As ever, the two drivers for the change in fashion are our old friends risk and return. Risk, not only because underperforming the index has unpleasant connotations following the Merrill Lynch/Unilever case, but also because underperforming the index means there is a greater chance of the fund failing to meet its liabilities. And return because there is now a growing awareness that slavishly tracking an index can be damaging to your wealth, and because in a market that’s struggling to rise, every little counts.
Enhanced indexation scores on both these dimensions. By picking the low hanging fruit and eschewing big bets and style biases, enhanced funds are able to deliver relatively consistent, if modest, outperformance. And by diversifying across many positions and multiple sources of return, an enhanced fund contains comparatively little active risk versus the benchmark. The result is a higher ratio of active return to active risk (information ratio) and hence a lower probability of underperformance. And as an added bonus, the active return streams from an enhanced portfolio are likely to show little correlation with those from other active managers, and consequently the overall information ratio for a multi-strategy fund can benefit from the inclusion of an enhanced indexation mandate.
In this article we develop these themes in more detail and examine the virtues of enhanced indexation both as a management strategy in itself and as part of a wider portfolio framework.
For some two decades now, indexation has been the management style attracting converts. In the US, indexation has captured some 40% of the institutional market, whilst in the UK its share is over 30%. Although indexation continues to grow in other markets, in the US and the UK its rate of penetration is slowing. In part this is because a substantial fraction of the investment community is unwilling to give up the thrills and spills of active management, but there is also a growing recognition that there is a natural limit to indexation’s market share.
Growing beyond this limit leaves active managers, including those running hedge funds, with increased opportunity to exploit those hugging the index. For example, slavishly following index changes and requiring liquidity at the moment a stock enters an index can result in index managers overpaying for new entry stocks by some 3–5%. If index turnover from new entrants is 2.5% per annum, the result is a performance hit of 7.5–12.5 basis points each year. This can be a significant fraction of the management fee for an index mandate.
In contrast, enhanced index managers are not bound to mimic index changes exactly and can trade the changes intelligently in the day or so immediately surrounding the effective date. Indeed, some go further and seek to trade changes to indices other than that for the benchmark. For example, an enhanced manager given MSCI Europe as a benchmark can trade changes not just for MSCI Europe, but also for other European indices that form subsets of the benchmark, such as the CAC-40, Dax, and FTSE. To minimise the risk of such index change strategies, the enhanced manager is most likely to trade a selection of index entrants versus leavers in an industry-neutral basket.
Trading an industry-neutral basket of index changes is but one example of a generic set of strategies in the enhanced manager’s toolkit. Another strategy is to take overweight positions in takeover candidates with compensating underweights in the bidder, the rationale being that acquirers generally overpay. Similarly, abnormal pricing differentials in different share lines of the same company can be exploited by taking over and underweights that will pay off should the price differential return to more normal levels. Given sufficient opportunity, enhanced managers can hope to win more times than they lose from these abnormalities and so generate modest outperformance. Unfortunately, market conditions may not be so kind as to throw up sufficient numbers of such opportunities. This has been especially the case in the post-bubble era and managers relying on these arbitrage-type strategies have found consistency difficult to deliver.
For this reason, it makes sense for enhanced managers to use a forecast-based strategy at the heart of their process: stocks that are forecast to outperform the market in the short to medium term are overweighted, while those with relatively poor performance forecasts are underweighted or excluded altogether. Again, it is important to implement the forecasts in a framework of strict risk control and this means ensuring, among other things, industry, country and currency neutrality.
Traditional active managers use their stock picking skills to create a relatively narrow portfolio compared to the benchmark. The result is an unbalanced portfolio with an asymmetry between the large overweights in the stocks selected and the underweights in the stocks excluded. (If shorting is not allowed, the size of the underweights is necessarily limited to the weights of the stocks in the benchmark.) The contribution to the potential value added from the underweight stocks is consequently constrained. Just how much this matters is shown in figure 1. This demonstrates how the return/risk trade-off is increasingly hampered by the inability to short as active risk increases.
In contrast, enhanced managers will hold far more stocks and effectively run a sampled index fund in which the sampling is based on the manager’s forecasts of future performance as well as on the contribution of individual stock positions to the fund’s active risk. The result is that the fund’s performance depends on a larger number of positions, both over and underweight, compared to a traditional active manager. This in turn increases the consistency of the strategy. Figure 2 shows the probability of a strategy underperforming its benchmark for various values of the information ratio. Because of their greater breadth, enhanced strategies operating at the left-hand side of figure 1 can expect information ratios close to one. Active strategies, on the other hand, would be top quartile if they had an information ratio of greater than 0.25.
Enhanced strategies can play a valuable role in improving the information ratio of a fund using a multi-manager strategy. A fund using a core-satellite structure with a passive core and an active satellite may be better off using enhanced management instead. This can be seen from the efficient frontier in figure 1. If the active manager sits on the same efficient frontier as the enhanced manager, figure 1 shows that combining the active manager with an index fund in a 50:50 ratio delivers less active return than employing an enhanced manager operating at the same active risk as the active/passive combination.
However, this argument only truly applies if the active manager is exploiting the same information and is pursuing a similar strategy to the enhanced manager. If the active manager is following a quite distinct strategy and using different information, then the active and enhanced managers will not lie on the same efficient frontier and the correlation of their active returns will be low. In this case, combining the active and enhanced strategies can result in an improved information ratio at the overall fund level.
Figure 3 shows an example of this. In this example we have assumed that we have identified index and enhanced managers who can run portfolios against the relevant benchmark. We have further assumed that we have identified specialist active managers who can run portfolios against parts of the benchmark. For example, if the benchmark is a US index, we might look for value and growth managers, or for large and small cap managers; if the benchmark is MSCI Europe we might divide it between Euroland and non-Euroland specialists. So as not to give the enhanced manager a head start, we shall ignore the evidence from figure 1 and assume that the enhanced manager and the two active specialists have the same information ratio, a conservative 0.2 in this example.
We further assume that the three managers are using essentially different investment processes and that the correlation between their active returns is essentially zero. By definition the correlation of the active returns with an index manager will also be zero. Figure 3 shows how the optimal mix of indexed, enhanced and active strategies varies as the fund’s propensity to take on risk increases. The central message is that all types of management style have a role in institutional portfolios provided they have positive and significant forecast information ratios and that they are following distinct, and hence uncorrelated, management styles.
Enhanced indexation has a lot going for it and it deserves the airtime it is currently receiving from managers, sponsors and consultants alike. By not trying to do too much with too little, enhanced management can add value over an index by taking on multiple small, risk-controlled and diversified positions. The result is a strategy with an ex-ante information ratio that by definition is higher than for a strategy using the same information but taking on more active risk.
Furthermore, the lesson of diversification applies not only to the positions held within a portfolio, but also to the overall fund. Enhanced portfolios exploit strategies not readily accessible to traditional active managers and so are largely uncorrelated with them. Combining enhanced management with an indexed core can therefore provide significant improvements to a fund’s overall information ratio.
Christopher Woods is chief investment officer at State Street Global Advisors UK in London