I have done an awful lot of meetings with clients, rating agencies and other professional fund selectors for over 35 years, and one thing stands out to me. Everyone wants to talk to you about your idea generation process and particular ideas, yet no-one wants to talk about portfolio construction. As the years have gone on, I have found this increasingly surprising. The product I manage, and market is a portfolio of investment ideas, not specific ideas. My job title is Fund Manager, not stock picker. Obviously, we in the industry are partly to blame for this. When we put presentations together, we know that interesting stock stories are much more motivational than the dry and mathematically heavy topics of risk and portfolio construction.

There was a moment about twenty years ago when doing a joint roadshow with a number of other multi asset fund managers, when a canny client observed that the macro analysis and outlook we were all giving was essentially the same, yet our performances were so very different. The answer was of course how each manger scaled his positions, basically risk management and portfolio construction.

Over subsequent years in the industry, particularly those spent heading a large team of investors, it has become clear to me that the real key to long term success is not idea generation but portfolio construction. While finding the next great thing and getting in early can be a great strategy short term, it relies on a huge amount of luck, in reality. Yet most of the industry’s brain power is spent on this activity and very little on risk and construction.

It’s a simple fact of life that no matter how confident a fund manager might be, the real odds of making any individual stock call correctly are actually only slightly above 50%, perhaps 55% at best. Huge amounts of intellectual and emotional energy are spent in this area. Very little is spent considering portfolio construction, and in effect, how this skill is applied effectively to optimise returns to client’s needs.

Once you understand these, it makes sense to spread risk broadly across a portfolio to maximise the chances of skill coming through into investment returns. However, many managers concentrate their portfolios in a small number of high conviction positions. The reasoning being that they think they have a greater chance of being right in those positions than the rest of the portfolio. While it is logical, perhaps, to think that a manager might have some limited skill in predicting the future returns from a security, for him to be able to predict where he has the greatest chance of predicting these returns with any accuracy is a step too far for me.

A simple example can illustrate this. A fund manager has identified 10 stocks he likes. One of these he likes much more than the others. He puts 30% of his portfolio in this and the rest evenly across the remainder. Were his favourite stock to fall 10%, which realistically any stock can, the rest of his portfolio must rise by 4.2% to make up for this. Of course, if he thinks the rest of his portfolio is likely to rise by that amount, why does he not back those as much as his favourite. This strategy basically presumes a high level of skill than is evidenced by lived experience. It seems especially irrational when the outsized stock position arises as a consequence of price rises. Here the manager must be expecting higher future returns post the price rise than prior.

Our approach is to scale positions such that they make roughly equal contributions to risk. When positions rise, we trim them frequently back to start weight. In this way, we are not dependent on any individual stock going well for the fund to do well, we just need a small amount of skill across a wide range of positions. Basically, we don’t think we have a large amount of skill at predicting the future.

Position scaling is not the only factor in portfolio construction, but it is clearly one of the most important. While running relatively equal weighted portfolios may seem to lack conviction, it aligns much better with client needs than the aggressive conviction approach.

There have been huge amounts of discussion this year regarding the apparent dominance of a small number of stocks in the indices; the MAG7. It might be argued that this contradicts our approach, holding outsized weights in these stocks might have really helped this year. Managers who run active portfolios may be heard arguing that they could not outperform without an overweight position in these stocks. Unfortunately, this is a self-undermining argument. As an active manager, our role is to find those stocks that are doing well, these may include the MAG7, but also the other 3000 stocks in the benchmark. Even in a year such as this, a material number of stocks have beaten the benchmark.

Source: MSCI Word Equal Weighted Net Total Return USD Index and Bloomberg Finance L.P. 29.12.2024 – 05.07.2024
Past performance is not a guide to future returns.

It follows that you are either an active manager with a high probability of identifying successful investments and therefore, you will have picked successfully amongst the MAG7 and the other 3000 stocks, or you accept that you have much more limited skill than perhaps you previously believed. When active managers make the argument that the average stock has underperformed, they are undermining their very position as active managers.

Obviously, our equal weighted approach means that we have been underweight the MAG7, although we do hold a material position. In the current year we have had to work hard to find other areas of the market that have also done well. For us, these have been the US reshoring theme, in particular industrials, but also our nuclear energy exposure, and further afield in Asia.