Enhanced indexation has grown significantly since the mid-1990s. There are good reasons behind this. Enhanced indexation scores well in direct comparison to pure indexation – it shares the attributes of the latter, having lower fees, lower transaction costs and a lower risk of underperformance, while offering the scope for outperformance versus an index. The only downside when considering an enhanced approach instead of a pure passive product is that it introduces the risk of underperformance, albeit this is a reduced one compared with the underperformance risk of a traditional active product. The size of this risk, and hence the size of the downside potential is dependent on the unique characteristics of the chosen approach. However, when considered at a portfolio level, this apparent increase in risk taken may be less significant.

If one assumes that the risk of equities is 15 per cent relative to a pension fund’s liabilities, and the tracking error of the portfolio is 2 per cent then the actual risk is 15.1 per cent. This is because risks do not add up linearly. The dominant risk decision for a pension fund is deciding to invest in equities in the first place. Once that decision has been taken, the options available can be characterised as: passive, enhanced index, full active management or a combination of the three.

Defining enhanced indexation

At this point, we should clarify what we mean by enhanced indexation. The growing popularity of such an approach has led an increasing number of investment managers to offer products under this designation. However, some products, while based on quantitative techniques, have active risk levels of around 2 per cent or even higher, and would historically have been classified as ‘low risk’ traditional active products. We believe this is slightly misleading and that such products should be defined as ‘active quant’.

More recently, the term has been stretched to include products that are simply scaled-back versions of traditional active products. We favour a broad definition, but one that excludes these type of approaches. In our view, enhanced indexation is an approach that aims to outperform the index modestly, while also retaining the overall characteristics of the index.

An alternative to pure indexation

There are two reasons for adopting an element of enhanced indexation in an equity allocation:

1. As an alternative to pure indexation;

2. To improve the efficiency of a manager structure that combines active and passive elements.

The appeal of indexation rests on a number of key features: a very low risk of underperforming the index, low fees, low turnover and hence low transaction costs. But what should a pension fund consider when comparing pure indexation strategies to enhanced index strategies?

When evaluating enhanced index products, we believe clients should reflect on four factors:

• tracking error;
• the achieved information ratio (IR);
• the techniques employed to gain excess performance over an index;
• the overall cost to the client.

In broad terms, when comparing an enhanced indexation to a pure indexation product the lower the tracking error the better, as long as the client feels that the incremental return targeted is worthwhile. In essence, a low tracking error seeks to reduce the risk of significant loss relative to the index. The flip side of this is that a low tracking error will reduce upside return potential, as the portfolio needs to closely mirror the index in order to maintain its low tracking status and therefore opportunities to outperform are limited.

The importance of tracking error is therefore determined by a client’s desired level of return.

The significance of information ratio

The information ratio (IR) is defined as the rate at which risk, relative to a benchmark, is expected to be converted into return in excess of that benchmark. Ex-ante risk relative to a benchmark is usually described as ‘tracking error’, and outperformance relative to a benchmark as ‘alpha’. Thus, in a formula, the information ratio could be described as follows:

IR = expected alpha/tracking error

For any given skill level, a product with a large number of smaller bets will produce a higher information ratio than one with a smaller number of larger bets. This is because for maximum efficiency, a manager wants to have equal scope to over- or underweight stocks relative to the index. The ability to fully underweight depends on the ideal size of that underweight.

If the stock is only 0.1 per cent of the index but the manager’s normal size of position is to be +/-0.5 per cent, for example, then a loss of efficiency occurs and consequently the information ratio is brought down. An enhanced process might only take positions of +/-0.1 per cent to control risk and therefore achieves a full underweight in many more stocks. Because an enhanced index fund aims to outperform the index, IRs are highly relevant and in fact a key differentiator between provider products.

The higher the expected risk conversion ratio, the more attractive enhanced indexation looks. Indeed, a number of enhanced index products have been able to deliver significantly higher information ratios than traditional active products, and often over long periods.

Techniques for adding value

There are a number of techniques used in enhanced indexation approaches that are not focused on overweighting one stock and underweighting another; but instead, they concentrate on alternative ways of getting exposure to the same stocks, but in a much cheaper way. These are known as ‘substitution techniques’. They create minimal additional tracking error as the additional exposure is not different to the index. If successful, they will increase the information ratio. When we examine enhanced indexation in the context of the overall structure, it is an advantage if the alpha is not highly correlated with that of the active management component – this is more likely to be the case if the techniques employed are different to those used for traditional active management.

Also, products that use multiple strategies that are effectively independent of each other, add value by providing an internal diversification benefit, further enhancing the information ratio and creating a multiple-manager structure.

Cost to client

Fee levels will vary but, generally enhanced index products offer lower fee scales than active management. Fees are therefore more comparable to passive products. Some providers offer performance fees, offering a base fee similar to that payable for pure passive management; any additional fee to the manager is dependent upon performance being in excess of the index and proportional to the amount of outperformance achieved.

Transaction fees are another significant area for clients to consider. Changes of index constituents as a result of corporate activity etc. will result in transactions in a pure indexation portfolio, as fund managers need to buy and sell stocks in order to mirror the index. Consequently, transaction costs will be a debit to performance, though on a much smaller scale compared to an active portfolio. This principle also carries over to enhanced index products. Since the positions taken are much smaller and closer to the index, the scale of turnover is usually lower than for traditional active management. Compared to pure indexation, the transaction costs associated with enhanced indexation are not really an issue, particularly if one adopts a performance-related fee structure.

Risk of underperforming the index

While we have established some of the similarities and differences between the modus operandi of pure and enhanced indexation, we have not examined the question of downside risk. Given that tracking the benchmark is the basic premise of pure indexation, one must assume that any client deciding between pure and enhanced strategies will want to err on the side of caution in this area.

Figure 1 quantifies the probability of underperforming the index over one year by 1 or 2 per cent, according to the risk level of the enhanced index product.

Initially, we assume an information ratio of one for all risk levels. In practice, as risk levels increase above a threshold, we would expect it to deteriorate. The final column shows how a 2 per cent tracking error product can only sustain an information ratio of 0.75, due to higher risk levels. The risk of underperformance is very small for the lower risk 'purer' enhanced index product, but as the inherent risk increases, the risk of significant loss mushrooms. This is because the risk/return relationship is not linear. The potential for significant underperformance is very much a factor of the risk level of the chosen product: the lower the risk level, the lower the risk of underperformance. For low risk level products, the risk of even 1 per cent underperformance over a single year is very small. Thus enhanced indexation via a low-risk product seems the more attractive option when augmenting or replacing passive portfolios.

A simple portfolio model

A detailed analysis of different portfolio structures lies outside the scope of this article. However, based on some straightforward assumptions, we can generalise about the impact of substituting pure indexation with a tranche of enhanced indexation in a typical active passive structure. This brings expected return benefits; increases in the overall net-of-fee-information ratio; reduction in the probability of underperformance at typical thresholds and a greater potential for gains according to the size of the allocation to enhanced index. Clearly this is a very desirable outcome, but how so? The biggest factor is that enhanced indexation introduces another way for the portfolio to win. If the active component of a fund’s assets outperforms, the enhanced index component is generally unlikely to do significant damage to the overall result. However, in those cases where the active element underperforms, the enhanced index element can often mitigate the situation by making a positive contribution. Something pure indexation will never achieve. This benefit of diversification of alpha within a limited risk framework is unique to enhanced indexation. Enhanced indexation shares the attributes of low fees, low transaction costs and a low risk of underperforming the benchmark with pure indexation. Yet it also offers scope for outperformance, making enhanced indexation an attractive alternative to passive management.

In co-operation with: BlackRock

Contact:
Anthony Biddulph,
BlackRock,
33 King William Street,
London, EC4R 9AS

Tel: + 44 207 743 4656
Email: anthony.biddulph@blackrock.com