Better Business: Model behaviour

David Ingram, a partner at threesixty, examines the evolving role of model portfolios.

There is very little dissent from the view that correct asset allocation is key to maximising the potential returns from an investment portfolio.

Research (Nobel Prize winner Harry Markowitz ‘Portfolio Selection') suggests that asset allocation can account for around 90% of the return from investments.

The theory goes that combining asset classes which are non-correlated reduces risk and it certainly seems to work. The risk of a portfolio consisting of 100% gilts can, for example, be shown to be higher than that of a portfolio of 90% gilts and 10% UK equity.

So an asset allocation model for a "balanced" investor looking to invest over the long term might be:

Money Market 5%
UK Sterling Corporate bond 20%
UK All Companies 45%
UK Smaller Companies 10%
Global Growth 5%
Property 15%

There are a number of different methods of obtaining these asset allocation models, specifically the various risk profiling built around psychometric questioning although it is important to check the reliability of these.

Populate the portfolio

Having identified the appropriate asset classes there is, obviously, a need to identify suitable assets to use to populate the portfolio. At present, this is generally a need which is met by the use of some sort of fund - OEIC, life or pension fund. Other types of assets, such as ETFs or direct investment in property are being used as well but this is, so far, a minority approach.

There is then the question of what happens when, as inevitably happens, the different asset classes grow (or fall) in value at different rates and the balance of the portfolio changes.

Selecting the appropriate funds is a potentially onerous task involving considerable time and effort in researching fund performance, charges and the actual asset allocation of the fund over a two or three year period to assess consistency.

Increasingly, there is a trend for IFAs to use model portfolios with currently around 54% following this route.

Most asset allocation models are fairly static. They are based on a strategic view of how markets are expected to perform over a medium to long investment term. This leads to some clear problems from time to time when particular markets are undergoing downturns. Property is, for example, a part of many asset allocation models but few advisers will have felt comfortable recommending property investment as part of a cautious portfolio during late 2007 and 2008. This led to clear temptations to substitute one asset class with one of a similar risk rating (corporate band instead of property) despite the damage this might do to the asset correlation.

Alternatively, clients were advised to defer making part of their investments until conditions improved - thus leaving the adviser to undertake market timing planning.
The last 18 months will also have caused some concern to advisers wanting to rebalance portfolios with a consequent disposal of any "good news" funds and reinvestment in areas where performance has been poorer - to keep to the appropriate risk profile. Such rebalancing is, of course, an important part of reviewing a client's investments; it ensures that a client's portfolio does not become so heavily skewed that the risk profile no longer meets the client's risk tolerance. A generally accepted rule of thumb is that portfolios should be rebalanced if any ‘class' becomes unbalanced by more than 5%.

There is, however, a potential danger here. Asset allocation models are very static because those who design them generally take a very long term view so that following them in some market conditions can lead to almost immediate investment losses.

As a result of this, a new range of model portfolios is gaining traction amongst advising firms. These portfolios are based on the assumption that the traditional asset allocations are probably where the portfolio will be based on average over the medium to long term but that those allocations should be varied to meet short term opportunities or avoid short term problems. They adopt a tactical approach and, in effect, bring a limited element of market timing into the process.

These portfolios are able to take account of current risks and adjust the risk-rated asset allocations accordingly. There are clear benefits to such a system:

• Less client concerns over performance of individual asset classes
• Less need for adviser to undertake market timing/asset substitution exercises
• Reduced risk of investing in ‘unsuitable' asset classes

Returning to the ‘Balanced' investor as outline, the appropriate tactical asset allocation at this time (October 2009) could be along the lines of:

UK Equity Income 20%
Money Market 15%
Europe Excluding UK 7.5%
UK All Companies 15%
Global Bonds 5%
Asia Pacific 10%
Sterling Corporate Bond 10%
UK Index Linked Gilts 5%
North American All Companies 4%
North American Smaller Companies 3.5%
Global Emerging Markets 3%
Japan 2%

Extra admin

Of course, this tactical form of model portfolio is not without its own downsides of which the most important is, clearly, the extra administration and therefore cost which the almost inevitably increased rebalancing will generate.

All model portfolios suffer from the problem of getting client consent to make the recommended switches. As such, they work best with those clients who receive regular reviews.

There are, of course, alternatives which can remove/reduce the administrative problems. It would be possible, for example, to create a series of risk rated OEICs around the model portfolios or to obtain the necessary variation of permissions to become a discretionary manager although, of course, this would cause some problems itself with increased capital adequacy and qualification barriers.

There is no perfect way of dealing with a client's investment needs, especially with the need to consider alternative asset types such as structured products, unregulated collectives, and so on in order to retain ‘Whole of Market' stakes in the post RDR world.

The advent of tactical model portfolios is, however, a very valuable addition to an IFA's armoury and will prove a useful way of meeting the needs of segments of their client base.