The holy grail for most private client investors is to a select an investment manager who consistently and repeatably over time meets their investment objectives and out-performs their peers. And even though investment literature is peppered with warnings that past performance is no guide to future performance, the belief remains that decisions to appoint or remove a manager should be based, at least in part, on their performance track record.
At ARC, our manager selection methodology is primarily focused on investment style, prioritising seeking to understand how different managers approach investing and how that fits with client preferences and market conditions. We recognise that long term mutually beneficial relationships between clients and managers are built on shared values that can weather the inevitable variability of market conditions.
However, the question remains: are there discretionary managers that have persistently outperformed? And if there are, can a common explanation for such outperformance be identified?
Evidence For Discrete Period Performance Persistence
To examine the question, we have analysed the ARC Sterling Private Client Indices Universe, specifically the Steady Growth risk category which captures portfolios with a risk relative to world equities of between 60 – 80 per cent. If Fama’s observation is correct, and there is no persistence of out-performance, then we should expect no statistical relationship (ie correlation) between a manager’s positioning relative to peers from one period to the next.
To address Fama’s point concerning risk, our analysis is based on the Sharpe ratio, that is we compare risk-adjusted returns, thereby eliminating any return benefit from merely taking more risk in a bull market or less risk in a bear market.
The results are set out in the bar chart below which is most easily understood by going from right to left. The bar on the far right represents the top quartile managers in the ARC Sterling Steady Growth PCI Universe based on their Sharpe ratio for the three year period ended June 2021. These are the managers who are likely to be actively marketing and winning mandates and awards right now.
The other three bars represent three independent and separate periods going back in time in jumps of three years. Thus, the bar second from the right shows how those same managers performed for the three year period ended June 2018. The next bar to the left covers the period ended June 2015. The bar on the far left shows the results for the three year period ended June 2012.
If Fama is correct that risk-adjusted performance persistence is a myth, then apart from the bar on the far right the proportion of managers in each quartile should be equal; 25% each. That is clearly not the case. So what might be going on to explain what looks to be a pattern of decay?
Accepting that analysis such as this is always prone to start date dependence, the results do seem to indicate that there is some element of persistence of outperformance. Go back a decade or more and there is not much evidence of persistence. But if an investor had picked a top quartile manager based on their three year track record three years ago, they would have had a 50% chance of selecting a top quartile manager. By contrast, if an investor picked a bottom quartile manager based on their three year track record three years ago, probability of success was under 10%.
Pattern of Quartile Performance By Individual Managers
The analysis above provides some evidence that there is persistence of out-performance within the Sterling PCI Steady Growth universe of discretionary investment managers over the last decade but the results only show the point to point change in quartile position. For private clients, satisfaction with their manager will be determined by their journey as much as their end result. Thus, the proportion of time spent under or out performing peers is important. The reality is that, in the presence of considerable uncertainty, consistency is valued almost as much as quantum!
To visualise that journey, the chart below plots the quartile rank of the individual managers as a time series over the past six years, with the quartile position for each three year period based on the year Sharpe ratio. Going left to right, managers are ordered by their current three year Sharpe quartile. Thus, the manager on the far left of the bar chart is the one with the best Sharpe ratio over the last three years. The manager on far right of the bar chart is the one with the worst Sharpe ratio over the last three years.
Again, if there is no persistence of out performance then each manager would spend around a quarter of the time in each quartile.
It is clear that there is significant variability in the pattern of relative performance over time, indicating a diversity of styles and approaches. However, it is noteworthy that, whilst almost all currently top quartile managers have had periods of underperformance versus their peers, top quartile managers have tended to spend more than half the time in the top quartile. That general result is mirrored for bottom quartile managers.
Market Trends
The data above appears to show that there is evidence of persistent out-performance. So, is Fama wrong? Maybe there have been consistent themes over the last six years that provide an explanation for the apparent persistence in manager relative performance? The chart below investigates 14 asset class pairs.
These 14 asset class pairs represent some of the key asset allocation decisions facing discretionary investment managers: growth v value; domestic v international; duration v credit risk; etc. Relative risk-adjusted performance is plotted with the colour denoting the outperforming asset of the pair; and the strength of colour representing the performance differential .
The chart demonstrates that for much of the past six years there has been relatively little variation. That implies outperformance has been driven by strategic asset allocation decisions rather than tactical asset allocation decisions or stock selection.
Extending the analysis back to December 2003, the inception date for the ARC Private Client lndices, reveals a different picture with almost every asset class pair seeing one or more rotations and the regime change wrought by the Global Financial Crisis (“GFC”) clearly visible.
Do discretionary investment managers exist that have delivered persistent outperformance over the last decade or more? Yes! The explanation for that superior performance is most likely to involve strategic asset allocation choices rather than tactical trades and stock selection.
When discretionary manager philosophy and process aligns with the prevailing financial market regime, outperformance tends to be the result. However, regimes do not last forever and when change comes it can be very rapid. Such inflection points usually result in an upheaval in manager rankings.
It feels like we may be at such an inflection point. The post GFC themes of financial repression and digital economy may well be replaced by climate change; inflation; protectionism; or other factors. What is certain is that, whilst selecting an investment manager remains as much art as science, investors ignore understanding performance track records at their peril.
Sign up for more insights here
