Very few active participants in financial markets would contest the critical role that strategic asset allocation plays in determining overall investment performance. The empirical evidence to support its role is indisputable, over both the long-term and periods as short as a few years.

But just how much difference can ‘good’ or ‘bad’ strategic asset allocation calls make? To illustrate the general concept, consider the following simplified model. The pie chart below illustrates a passive equal-weighted portfolio invested across seven major asset classes. Its investment performance, assuming annual rebalancing, is shown in the performance chart.

In comparison, the ‘good’ tilts portfolio performance line illustrates a historic simulation of how the strategy would have performed had the investor been smart (or lucky) enough to have allocated at the start of each year a half extra weight to the best performing asset class and halve the weight to the worst performing asset class for that year.
For example, in 1995, this would have been 21 per cent in the S&P 500 and seven per cent in US$ deposits. The ‘bad’ tilts portfolio on the other hand illustrates what the simulated strategy result would have been had the overweights been in the worst performing asset class and the underweights in the best performing.

While the performance analysis is purely hypothetical, relying as it does on repeatedly perfect foresights at the start of each year, the outcomes are nevertheless revealing as to the impact of even small asset allocation bets on the overall results. Over the 15 year period to end 2005, the good tilts portfolio would have compounded at 11.49% pa, the equal-weighted at 9.28% pa and the bad tilts at 7.09% pa. In other words, the difference between good and bad decisions for even limited strategic tilts in this model amounts to 4.4% a year – or 66% over the entire period. Strategic asset allocation seemingly does matter.

In the real world, of course, we don’t have the luxury of perfect foresight and the ability to over and underweight accordingly. There are nevertheless some very well established asset allocation techniques that all active investors should give strong consideration to in the management of their portfolios.

First, investors should avoid entirely the temptation to ‘bet the ranch’ on any particular asset class at a point in time – it seldom works and it is extremely costly when it doesn’t. There is considerable stability and value inherent in a globally diversified portfolio of less than perfectly correlated asset classes, and investors should try to maintain a discipline of restricting their asset allocation bets to the margin and within well-defined parameters – say double or half weights.

Second, mean reversion is one of the more reliable features of investment market behaviour. There is no simple rule of thumb, but asset classes that significantly underperform in one year often bounce back in subsequent years, while those that hugely outperform often fall back. Careful and judicious over and underweighting to capture this reversion can often yield meaningful gains, and simple rebalancing of portfolios helps too.

Third, it pays to be particularly cautious of sentiment, momentum and hype. There is a great risk to chasing themes and markets which are overplayed and extended. It is usually when ideas are most popular that they are near the end of their run.

Fourth, investors should try and structure portfolios such that for each asset class employed, a percentage of that asset class is allocated in a cost effective way so that changes to over and underweights can be implemented efficiently. For example, if the neutral allocation to global equities is, say, 30%, then 25% might usefully be considered as ‘core’ and invested accordingly, while 5% might be invested in products or funds which are easy and cost-effective to enter and exit so as to accommodate asset allocation decisions as and when they are made.

Finally, implementation is critical and talented fund managers can often help considerably to enhance portfolio results. But even great managers will not be able to offset the adverse effect of poor asset allocation decisions – ultimately it matters a great deal when and where you are – but they can lessen the impact. More compelling, of course, is that when asset allocation decisions are right, they can amplify results enormously.

Graham Wainer is group head of private clients & portfolio management at GAM

Top