Adding value to model portfolios

By sticking to model portfolios, investors might be missing out on major opportunities for growth.

More and more often, the starting point for asset allocation advice for a client is a model portfolio selected from a range prepared in advance by a product provider or specialist investment advisory firm.

That is perfectly understandable, and I fully appreciate the reasons why advisers take this route. It is said 90 per cent of returns come from asset allocation, while only 10 per cent from stock selection, but this is a retrospective statistic. Asset allocation is hellishly difficult, so relying on the brain power and number-crunching capability of a specialist is sensible.

Moreover, in an ever more regulated world, to be able to justify a recommendation on the basis of the analysis of a leading investment house is a low-risk option, and it sure beats starting from scratch with each client.

When using such tools, it is sensible to understand the assumptions they are based upon, and the implications. The crucial parameters are:

• The asset allocation choices;
• The anticipated returns for each asset class;
• The anticipated volatility of each asset class, and;
• The correlations of each asset class with every other asset class.

Past and future returns

These are inevitably derived from the analysis of past returns. As an actuary, I cannot dispute the future can be predicted from past information. This is how actuaries derived the mortality statistics that provided the foundations of the world’s life assurance industry. However, you always needs to be aware of any changes that are likely to make the future differ significantly from the past. I will give you two examples: one flippant, the other less so (I’ll let you decide which is which).

Suppose you were trying to predict the future results in the English Premier League. If you looked at past results, going back 20 years or more, you would conclude that Manchester City would struggle to get into the top half of the table and Liverpool would be challenging for the top spot. Sadly, (and I am a Red) that is not likely to be the case for the foreseeable future: in fact, quite the opposite.

Or, suppose you were trying to predict the likelihood of default on a portfolio of randomly selected US mortgages. If you looked at historic default rates, you might conclude that significant losses on the whole basket were highly unlikely. However, if the mortgages were advanced during the 21st century, then the opposite was the case. Some quite brainy people (at least on paper) made the fundamental mistake of failing to identify differences between the past and the future. This made the past experience seriously misleading, something that has cost the rest of us dearly, (but by the time their incompetence was realised, they had taken the money and run).*

Asset allocation choices

Today, there are typically between ten and 15 of these, and the number keeps on increasing, as new areas of investment are explored. The obvious ones such as the UK, US and European equities and UK gilts have been joined by emerging markets, index-linked bonds and Asia-Pacific. Then, sometimes, large-cap and small-cap equities are split apart.

But there are difficulties here. Can the major equity markets today be considered the same as they were 20 years ago? Look at the constituents of the London market today, and then look back 20 years. Some stalwarts of that time such as ICI have disappeared, and others, such as Beecham, have changed out of all recognition. And some asset classes are still not well-defined. What exactly is an emerging market these days? It certainly is not the same as an emerging market 20 or even ten years ago.

When the parameters have been decided upon, a computer simulation uses standard statistical techniques to determine optimal portfolios for various levels of risk, (taken as volatility). Well, almost: human intervention always overrides the computer to make sure the resultant asset allocation is acceptably close to preconceived notions. For example, that a balanced investor (however defined) should have 50 per cent of their portfolio in UK equities.

Taken altogether, the above process produces a ‘safe’ result. But, despite the process being called ‘optimisation’, does it produce an optimal result? To my mind, no.

Ignoring the continuous secular changes and developments that are going on around us all the time by using an exclusively backward-looking process guarantees that major opportunities for growth will be missed, and investors will be locked into declining economies and sectors. The biggest opportunities for the future will, in my view, be found much more often in the emerging economies, which are not weighed down by decades of accumulated debt, both government and personal.

A company to whom I am an adviser, INSYNERGY Investment Management, has recently launched two funds investing in just such economies: the Absolute China fund, managed by Michael Lai of GAM; and the Absolute India fund. Both economies have the prospect of high growth for years to come, as infrastructure spending expands and the emerging middle classes try to bring their standards of living up to western levels.

But surely, you will say, it is not possible to invest in such funds and yet preserve the security that comes from following the model portfolio route? I disagree. No asset allocation model is so inflexible as to prevent a significant element of discretion, and not to use it in this way is to miss a major opportunity.

* The second example is the flippant one. As Bill Shankly famously said: “Football isn’t a matter of life and death; it’s more important than that.”

Jim Roberts is head of the investment committee at INSYNERGY Investment Management.