Greg B Davies

On one side of the suitability dancehall stand investors. On the other, investments. In the middle, responsible for finding everyone a suitable dancing partner, sits an enormous elephant.

The elephant wants to help. But it doesn’t know how. This makes everyone very embarrassed. They struggle to even look at the beastly blockade. Matching investors to investments, they conclude, is best not thought about too hard.

Lost in translation

On the surface, suitability is simple. Profile an investor, map that profile to a portfolio and check in periodically. Yet few ‘solutions’ offer any credible means of accomplishing this. At Oxford Risk we have developed a methodology for making this link which we believe to be unique in both its reliability and its accuracy.

It’s possible to be perfectly happy with both your risk-profiling and investment processes, but how do you know you can trust the map? Was it created using the latest cartography techniques, or is it drawn in crayon with ‘here be dragons’ scribbled in one corner? Linking investor and investment risk requires putting the two into the same language, but whose responsibility is the translation?

In the world of multi-asset funds, the ‘multi’ can apply to the interpretation of the fund’s name as much as its contents; one firm’s ‘balanced’ is another one’s ‘conservative’. Maybe your profiling tool suggests a growth/defensive split, but isn’t clear about what that means. Maybe it uses boundaries so broad as to be almost meaningless. Or maybe it goes to the other extreme and prescribes asset allocations with suspicious and spurious precision. Spurious precision helps precisely no one.

In short: are the two sides talking to each other or past each other?

The impossibility of ‘perfect’ doesn’t preclude a ‘best’

Part of the problem is that there cannot be a perfect empirical way of mapping investor risk to investment risk. You cannot link the long-term risk people prefer with the short-term choices they make. Substantial evidence tells us that actual decisions are a terrible reflection of long-term preferences. But there can be a best way, based on what we’ve learned so far from investment theory, decision theory, and behavioural science. A way, not of shying away from these complexities, but of navigating them.

This graph provides a simplified overview of Oxford Risk’s methodology. The ‘optimal’ portfolios (in practice, ‘sufficiently optimal’ portfolios with which an investor is sufficiently comfortable) are where the combinations of long-term risk and reward that an investor is equally happy to accept meet the universe of available investment options.

Portfolio risk is a measure of the dispersion of possible long-term portfolio outcomes, and the best long-term returns for each level of risk are shown as the ‘available frontier’. The frontier is of course only a prediction, and relies on an estimate of Sharpe ratios to determine its slope. It is absolutely vital that this available frontier is not an overconfident hope of how a favourite investment will perform, but rather a sober assessment of the long-term risk and return of the investment universe as a whole.

To translate an investor’s risk profile into the same language, we use ‘indifference curves’ such as those labelled as high and low risk tolerance on the graph. The suitable portfolio will be one from the range of risk-return trade-offs that an investor is indifferent between.

A risk profile is a measure of how an individual should trade off risk and return, as applied to their total investible wealth (it cannot be simply applied to any portfolio). It is a blend of their risk tolerance and their risk capacity: their emotional and financial constraints that set the limits of acceptability amid the investment universe.

Anchoring the complete range between zero risk tolerance at one end, and a Spock-like emotional acceptance of risk (if Spock had an infinite time horizon) at the other, enables us to accurately draw the indifference curve for each risk profile in the middle, and then calculate acceptable standard deviation ranges for each real-life profile. The footnote to explain the precise mechanism would, sadly, be longer than this entire article.

For example, a moderate-risk investor should expect long-term outcomes with an annualised standard deviation of between 8% and 12.5%. A simple 60-40 equity-bond portfolio is expected to be around 10.5%, bang in the middle of the middle risk profile, where experience and industry practice suggest it should be.

Making it personal

The elephant’s been confronted and trained. Everyone’s matched up. But solving the matchmaking problem is only the start. Even the most well-matched dance partners can encounter obstacles. For that, we need a different sort of map – one tailored to each investor’s emotional-comfort needs.

These can – and should – be accounted for separately, albeit as part of the same overall suite of suitability assessments, using behavioural scales to establish investors’ financial personality. Comfort-seeking behaviours are not constraints you want to feed into the calculation of the ‘right’ long-term portfolio, but you do want to consider them when helping investors stick with this portfolio.

Leave a Comment