One of the most important things in asset allocation is to have a policy for its ongoing management once a long-term strategic benchmark has been determined. That policy might be to adopt a passive approach with some type of rebalancing regime, or it might be to take a more active stance with tactical asset allocation. What is not appropriate is to simply drift along with the flow, allowing the markets to do your asset allocation for you.
What has been established by our work on rebalancing, and that of others, is that over most time periods a disciplined rebalancing methodology will produce better returns at lower risk. Even where one asset is for a while performing so far ahead of others that going with the flow would win in raw return terms, we have found no period where it is superior in risk-adjusted terms. This means that, passive or active, there will be a requirement to make adjustments to asset allocation positions with reasonable frequency.
Moving assets is expensive. Transaction costs are truly like icebergs and are usually bigger than you expect. Suffering mainstream transaction costs on physical assets is particularly unattractive. This is where derivatives come in, and especially stock and bond index futures.
The cost advantages of using futures to act as a proxy for physical assets to effect exposure changes are so pronounced that on those grounds it would be very hard to justify moving physicals. The ratio of total costs for physicals versus futures is for many markets in the 10 to 20 times range (see table for comparisons for a number of markets). This means that quarterly rebalancing of a typical pension fund portfolio can be achieved for as little as two basis points in transaction costs annually.
Cost is not the only consideration, however. In many cases the liquidity in the futures markets is superior to that of the underlying physical market. It is often easier to move large volume quickly in futures and with far less market impact. This is particularly important for large funds where active TAA is being employed, even if control ranges are relatively modest. Simultaneous next-day settlement for all futures contracts is a useful advantage compared with still disparate settlement cycles for physicals. There are also important tax advantages for futures as well as the avoidance of disruption to underlying portfolio management.
So what are the drawbacks? There are legitimate questions that need to be asked about risk control concerning the systems and processes used by the manager. It is important to distinguish, however, between normal investment risk of the type routinely undertaken by all funds and other sorts of control risks which are actually associated with the use of derivatives. It is essential to have strong controls when using futures and the installation of real-time pre-trade compliance systems represents a significant step forward in the risk control process.
What is not useful is where control type risk is confused with investment risk. This can happen because the mark-to-market feature of exchange-traded futures means that profits and losses are crystallised daily on a P&L or cash basis, rather than show up three months later on a balance sheet basis in a valuation, where a physicals portfolio has seen identical market movements.
Currency is often managed actively alongside TAA programmes because the returns are largely uncorrelated with asset returns. It is usually better to transact currency in the forward markets, which are superbly liquid, rather than in exchange-traded currency futures, which are not well developed. As forwards do not mark to market, but only crystallise profit or loss when closed, they do not have to be margined. This means that if currency is added to TAA, often no additional funding is required and settlements can be effected from the existing margin account for settling asset-related futures contracts.
There is one other caveat regarding the use of futures for the implementation of asset allocation decisions. There is an effective minimum size of fund below which there are likely to be some difficulties in realising all the advantages. Each futures contract has a size or value designated by the relevant exchange. In some cases these are quite high, so that the value involved in making a relatively modest exposure change to some markets would fall below a single contract size unless a fund was sufficiently large. This threshold is in the region of £50m fund size.
Overall, the merits of using derivatives to implement asset allocation programmes, whether active or passive, are sufficiently strong as to merit the institutional and organisational frictions be faced in setting up such a programme for the first time.
William Goodsall is chief executive of First Quadrant in London