Stay the course
Many questions have been asked in the last few months about how valuable long-term asset allocation models have been in dealing with the extreme market downturn of the last 18 to 24 months.
Stochastic models incorporating Modern Portfolio Theory techniques as used by a number of leading actuarial firms have been challenged as not being appropriate for developing strategies for retail investors.
The majority of IFAs are now, however, using these models through one route or another be it directly licensing them or accessing them through platforms or providers. According to NMG’s IFA Census in April last year, over 60 per cent of IFAs now use risk profiling and asset allocation tools in the UK.
This figure rises to 70 per cent where the IFA specialises in investment or retirement planning. Financial and investment planning tools continue to grow in their popularity and they are increasingly central to many firms’ investment advice processes.
Stochastic asset models provide advisers and their investors with a forecast range of likely investment returns over time and an assessment of how likely those returns are to happen. MPT expects that extreme events, both good and bad, will happen but that they will happen much less frequently than ‘average’ returns.
If a model has incorporated extreme events successfully then that should provide advisers with comfort that it is robust and that the long-term investment plans prepared using it are solid.
Many commentators are focusing on the hugely negative nature of the recent market turmoil as if it is a one-off. This is not the case. It has been rare, but not unusual. History suggests that such events occur once every 50 years or so but with intervening periods of over 100 years and as short as 20 years. The table lists six stock market collapses that were clearly global in terms of the investing participants and which have occurred in the last 300 years.
Other local events may have had repercussions elsewhere in the world, but each of the six had a clear impact globally where there were investment markets. While the limited number of data points prevents us easily putting an expected frequency on a likely global collapse, in principle the level of returns we have seen in the credit crunch collapse should be shown by a model to happen rarely.
Stochastic modelling companies determine assumptions for modelling future investment strategies through a combination of historical data analysis, implied current market yield expectations, current equity risk premiums and assumptions as to rational investor decisions. Historic data gives us a sense of the degree to which returns from certain asset classes move around their long-term averages, their volatility. We then create model portfolios that meet the various needs of investors based on their attitude to risk.
In one provider's standard asset model, it provides 10 risk-constrained asset allocations. The model portfolios start with 100 per cent cash and for each subsequent portfolio the cash is replaced with a percentage of bonds, property and equities of different types. The most risky portfolio has all equities, most of which are in emerging markets. Hence the level of volatility in each portfolio gradually increases with a commensurate expectation of an increase in expected returns. The increasing volatility is reflected by an increasing range of confidence limits around the expected return, that is, larger potential up-side potential but also larger down-side risk.
