Whether browsing the financial press, reviewing proposals from discretionary investment managers or comparing fund factsheets, it seems that all managers are able to make the claim that their investment performance is better than the competition. Indeed, users of the ARC Private Client Indices (‘PCI’) and clients of ARC have frequently asked “How can all discretionary managers be top quartile?” or in the weaker form “Why have I never met a fourth quartile manager?”.
Can such oxymorons be true? Can all managers be “simply the best”? In this article we consider some of the differing factors involved in making performance comparisons and whether these factors can explain why, anecdotally, all managers seem to have outperformed.
To examine this question, we have analysed the performance of the discretionary investment manager universe in the ARC PCI Sterling Steady Growth risk category. The Steady Growth risk category accounts for slightly over 50% of client portfolios in the PCI Sterling universe of more than 200,000 private client portfolios. We consider start date dependency; end date dependency; internal dispersion of returns; and risk equivalence.
Timing Matters
Investment managers have considerable latitude regarding the time periods over which performance is presented. In the chart below, we have taken four common time periods, 1, 3, 5 and 10 years and for each of the 72 Data Contributors to the ARC Sterling Steady Growth PCI universe we have highlighted in teal those that have delivered first quartile performance for that period. The dark blue bars indicate other quartile results and the grey bars denote where data is not available.
The results for each period show that as expected, for each of the individual discrete periods, only 25% of managers are in the first quartile. However, by combining the results across the four chosen time frames, 29 of the 72 Data Contributors (or 40%) can claim to have delivered top quartile performance over at least one period.
Let us now extend the analysis merely by looking at calendar years rather than ending mid-year. The chart below sets out the results when these two potential end dates are considered.
The results again show that for any single period, only 25% of managers are top quartile.
However, when considering whether a manager is top quartile over at least one of the 8 discrete periods, it turns out that 41 of the 72 Data Contributors (or 57%) of managers can claim to have delivered top quartile performance.
Thus, simply by varying the time period used for presenting performance allows 57% of Data Contributors to accurately claim that they have delivered top quartile performance. It is worthwhile noting that when market conditions are more volatile and persistence of performance is harder to achieve, rotation in peer group results may be materially higher than the relatively benign period examined above.
Portfolio Dispersion
The analysis so far has been based on the average portfolio performance for each Data Contributor as calculated by ARC within the Steady Growth risk category. ARC calculates that average by taking the mean return of all portfolios run by a given Data Contributor within the chosen risk category. However, for most Data Contributors, the average result spans a range of client outcomes reflecting different investment mandates, approaches to portfolio construction and holdings differences.
The chart below plots for each Data Contributor the range of actual client outcomes over the three years to June 2023 with lines indicating the 10th to 90th percentile range and bars indicating the 25th to 75th percentile range. The horizontal teal line indicates the return required to have achieved a first quartile outcome. The managers are ranked left to right by their “average” portfolio return.
The chart shows that for 32 of the 69 Data Contributors with three years history (46%) at least 25% of their client portfolios experienced top quartile returns.
Further, some 47 Data Contributors (68%) can rightfully claim that at least 10% of their client portfolios have achieved top quartile returns over the last 3 years.
To the extent that the top performing portfolios for a particular Data Contributor represent a particular investment solution offered to clients, it is possible to see how by selecting a particular investment solution to showcase performance rather than the average outcome, a majority of discretionary managers can present results that show them to be top quartile performers.
Risk Equivalence
It might reasonably be assumed that when a manager compares the performance of an investment solution to a peer group that there is risk equivalence. However, that is by no means certain. By construction the ARC PCI risk categories reflect risk bands – in the case of the Steady Growth PCI it is 0.6 to 0.8 times the volatility of world equities. If an investment solution offered by a discretionary manager has a volatility relative to world equities of 0.8 times, it might be reasonably compared versus either the Steady Growth PCI or the Equity Risk PCI.
Over the last three years, the Sterling Steady Growth PCI has recorded a return of 10.5% whilst the Sterling Equity Risk PCI has increased by 14.8%. If we assume a linear risk-return trade off between these two PCI risk categories, then the “average” portfolio with a volatility relative to world equities of 0.8 times would be expected to have achieved a return of 12.7%. Versus the Sterling Steady Growth PCI universe that would be almost a top quartile return. Compared with the Sterling Equity Risk PCI universe that would approximate a bottom quartile return!
Conclusions
In this article, we have considered some of the ways in which the oxymoron “All managers are top quartile” can be considered true! At any point in time, well over half of any given peer group are able to demonstrate top quartile performance. For private clients and their advisers, understanding exactly what performance is being presented is crucial.
Pitfalls we have highlighted cover start and end date dependency; the dispersion of returns delivered by a given manager; and risk equivalence. Good questions to ask whenever presented with performance data include:
Asking these questions will help to place performance more accurately into context. However, as we have shown, the majority of managers can rightfully claim to be better than the rest as they are top quartile performers! With apologies to Mark Twain, there are lies, damned lies and then there are performance statistics!
Sign up for more insights here