For most investors who are choosing unit trusts, the first, and often only, thing they consider is past performance. This is despite the well-worn disclaimer on every fund fact sheet that past performance is no indication of what might happen in the future, and decades of research that proves it isn’t.
Yet it is not unreasonable. People need to base their decisions on something, and historical returns are an obvious choice.
“Unfortunately, the reality is that performance is the easiest thing to measure,” says Anil Jugmohan, senior investment analyst Nedgroup Investments. “That’s why everyone does it.”
Humans are also predisposed to extrapolate what has happened in the recent past into the future. This is a natural cognitive bias because the world is generally confusing, and we can only ever see a small bit of it. We have to fill in the rest. Particularly since we cannot know what the future holds, we tend to believe that it will be similar to the past.
Investors and financial advisors therefore continue to use past performance, despite its flaws, not because they are stupid. They do it because it offers a way to make sense of a difficult decision. The only way to remedy this, therefore, is to gather more information.
Fill the void
As a starting point, it’s important to recognise the shortcomings of looking only at history. To illustrate this, Nedgroup Investments compared the three-year returns of unit trusts in the South African multi-asset high equity category over two consecutive three-year periods: from January 1, 2013 to December 31, 2015, and from January 1, 2016 to December 31, 2018.
The results were remarkable, and are illustrated in the chart below. Each marker on the chart shows a particular fund with its performance over the first three years plotted on the horizontal axis, and the second three years on the vertical axis.
The two blocks shaded in teal blue show that the funds that performed particularly well in one period delivered distinctly weak returns in the other. This stands out because it is at the extremes, but even among the funds in the middle green block, persistency is hard to find.
“If you run a correlation analysis on those funds, what you find is that it is almost close to zero in terms of their performance in one period and the subsequent period,” Jugmohan points out. “So, even ignoring the extreme outcomes, this shows very little performance consistency.”
What makes this analysis particularly interesting is that Nhlanhla Nene was fired as minister of finance at almost the exact mid-point – December 2015. This marked a distinct change in markets as bond yields blew out and the rand collapsed. It’s notable, therefore, that the managers who benefited most from the market conditions before this happened, were in the worst position afterwards, and vice versa.
This raises serious questions about how much skill was involved in that outperformance. At either extreme, were managers simply positioned for a single market outcome, which made them look clever when it worked in their favour, and much less so when it went the other way?
Through the cycle
It’s true that this was an unusual period, but when looking back at the two that preceded it, the picture does not change much.
Comparing fund returns for the three years from 2012 to 2015 with their performance between 2009 and 2012 there is a slightly higher correlation, although it is still low. These were however two consecutive periods when markets were buoyant and there was little change in the overall environment.
If one goes a further three years back, however, just about any sign of performance consistency once again disappears. Comparing returns from 2006 to 2009 with those between 2009 and 2012 delivers almost zero correlation.
The first period in that analysis includes the financial crisis and the 2008 market crash, so this is once again looking at two distinctly different market environments. It shows that doing well in one period is no guarantee that a manager will do well in another.
Do managers back up their talk?
So if consistency is so absent, does this mean that past performance tells investors nothing? Not entirely. Only looking at the numbers themselves may be inadequate, but there are lessons to be learnt from them.
“There is definitely value in past performance, particularly if you look at how a fund is positioned and how it performs before, during and after a market crash,” Jugmohan argues. “It helps you to understand whether a manager is doing what they actually say they are going to do, according to their stated philosophy and process.”
This is, again, about gathering more information.
“The man in the street can’t set up a meeting with the investment team at Allan Gray or Coronation,” Jugmohan says. “But they can at least read through the quarterly commentaries on fund fact sheets, where managers go into a lot of detail about how they position their funds. Most of the time you will find a fair amount of honesty there. Go back and read through what they have said over the years. It can take a long time, but you also have to prioritise this and say this is my future that I am investing for.”
Without that information, using past performance alone as a guide carries little value. For it to be worthwhile, investors need to appreciate how that performance was generated, and therefore how repeatable it is likely to be.
“On a superficial level, using past performance is dangerous,” Jugmohan says. “But if you are willing to do the hard work and dig deeper, then there is a lot to gain out of it.”