Managers with the Midas touch
The successes of football clubs and investment funds are often attributed to their managers but attempts to analyse their influence over results is not as simple as it might first appear.
There has been much excitement about the meeting between Manchester United and Real Madrid in the UEFA Champions League, with all the usual hyperbole that accompanies this most over excitable of sports. One particularly giddy claim that caught my eye was that this was a contest between the world’s two best managers; not being able to just let this go and enjoy the match, it got me thinking about how that could possibly be quantified. Is this based on past performance, and if so how are achievements in different leagues compared? Are achievements at different clubs comparable? How much impact have their respective clubs had? How much can be attributed to chairmen, Russian billionaires or just random luck? While this talk is normally the preserve of the pub bores, it is similar to the problem we face with fund manager selection; many of the same biases and flawed logic can often influence our decisions and ultimately investment returns.
The problem is in trying to determine what impact the manager has had and what they should be credited with. There are many factors that go into determining the results of a football match just as there are with investment returns. The manager may influence some, but many will be outside their control. Typically, treatment of these external factors is extremely asymmetrical, with the manager taking the credit when results are good, with the significance of outside forces dismissed entirely, yet when results are bad the manager professes to have been powerless to do anything in the face of such overwhelming odds. Identifying such factors is vital to being able to identify a good manager.
The largest external factor is undoubtedly luck. There are many managers, in both football and investments who are revered as being at the top of their game, despite all their achievements being entirely random. To carry on with my football analogy; Jose Mourinho, the Real Madrid manager and self-styled special one, is considered by some to be an excellent manager, responsible for considerable success at every club he has worked at. There are some however, who argue the Mourinho is in fact just lucky. In 2004 he and his Porto side faced Manchester United in the UEFA Champions League, the game ended in a draw with Porto going through on aggregate and ultimately winning the competition. The game however saw a goal by Paul Scholes wrongly disallowed, which would have eliminated Porto from the competition. Had this random referring event not taken place, Mourinho would not have won the Champions League, would not have gained the attention of the notoriously fickle Roman Abramovitch and thus his career might have taken an entirely different course.
This applies equally to fund managers. Fund managers who consistently outperform their peers have plaudits heaped upon them and are often revered as stars and remunerated as such, regardless of what effect chance has had on their returns. Using the IMA UK All Companies sector as a sample we can see what distortion luck can have on results.
There are 197 funds with a track record of six years or longer. In any given year half of these funds will produce returns above the median and half will fail. If the fund managers picked their stocks through the tried and trusted dart board method and their returns were purely judged on random returns we should expect there to be three funds that have been above average in each of the last six years.
As it happens, there are none. No fund has been able to beat the median six years in a row. There are some who have beaten the sector index but this measure is full of distortions owing to extreme results. No fund manager has been able to replicate the results that could conceivably be achieved blindfolded.
But as I am trying to demonstrate the effects of randomness it would be wrong to accept this single result as representative of anything. If we look at just the last five years, based on luck alone I would expect there to be six funds that have beaten the median every year, but in fact there is just one.
Cazenove UK Opportunities is the only fund to achieve this feat.
The fund is run by Julie Dean and Steve Cordell who are both rated as outstanding fund managers and have been awarded FE Alpha Manager Status, but what can we draw from this analysis? Are they to be commended for not getting in the way of chance, or have they been able to overcome additional external circumstances that have felled their fellow managers?
Additionally a further 14 funds managed to beat the median five years out of six, although not concurrently. Probability suggests we should have expected in the region of 30.
Leaving aside luck, which surely plays a significant role in any success, there are other factors that ought to be considered. An important one is where the fund’s mandate ends and where the active decision making process begins. Within the UK All Companies sector, not all funds are following the same strategy; some are focusing on mid-caps and some are exclusively look at large-caps, while others are pursuing a blended portfolio of the two.
At various times these strategies will come out on top. Mid-caps typically do well when markets rise and large cap stocks are preferred when they fall. Naturally funds implementing these different strategies will rise up the rankings as markets cycle through these different conditions. Likewise we saw plenty of defensive-minded managers being praised for their steady hand and wise caution during the credit crunch and following sluggish period, who have since fallen from grace. But in reality how much credit can the fund manager take for a fund’s strategy? While many managers have significant freedom in where they invest and the approach they take and pursue an unconstrained style, many have these decisions forced upon them.
When deciding upon how much credit a fund manager should take for a fund’s performance, it is important to consider the scope of their decision making. Within the UK All Companies sector the Allianz UK Mid Cap fund is currently rated top quartile over three years; this however is down to the exceptional recent performance of mid-cap stocks, and the decision to invest in these is obviously outside the fund managers’ control. Compared to the FTSE 250 index of mid-cap companies however, the fund has actually underperformed over this same period. The correct frame of reference is essential in assessing a fund manager’s true ability.
While the mid cap example is glaringly obvious, there are often more subtle pressures applied. Many asset management groups have a house style which influences which stocks the manager buys. A good example is the Templeton arm of Franklin Templeton. Templeton have a deep ingrained belief in value investing. Fund managers are required to select stocks and create portfolios that possess value characteristics. As with all strategies, there are times when value investing comes into its own and Templeton funds will undoubtedly benefit. The decision to pursue such a strategy is not down to the fund manager however and instead portfolio managers at Templeton should be judged for their stock picks and research, which is the key duty they perform.
Additionally, many groups centralise their stock research. Recently I have met with four fund managers from the same group who have all told me identical stories about one of the stocks they own. In some instances managers are restricted to buying stocks from a central recommended list. In this instance how do we infer the impact of the fund manager when they only get to choose themes and stock analysis is outsourced? A talented analyst – or a terrible one – could completely eclipse the fund manager’s impact on the fund.
In reality many of these factors are beyond quantifying, but still useful lessons can be learned. All of these factors add a huge degree of randomness to fund behaviour. Even when statistically we expected to see a pattern and identify three funds that had beaten the median six years in a row, the influence of many competing factors ensured it did not happen. It is the multitude of competing factors that makes any pattern of repeatable behaviour especially interesting and suggests human intervention battling through random events.
This is the basis for the FE Alpha Manager accolades that were awarded at the beginning of February. Repeated behaviour through multiple market conditions is a good indicator of discretion over style and strategy and highlights an activist approach. Repeated behaviour over longer time periods goes further in demonstrating the manager as the common denominator. The rating tracks performance across funds and under different employers and repeated performance patterns are then less likely to be owing to conditions imposed from above.
While the adage that past performance is no guarantee of future performance has been so overdone as to become one of the biggest clichés in investing, it is still relevant. But by trying to strip out external effects and attempting to isolate repeatable performance patterns that can be attributed to the manager, past performance can at least be used to accurately describe contribution. Identifying managers who have made genuine positive contributions is still a significant advantage in manager selection. As with football however no amount of analysis will identify the lucky fund managers, and these are usually the most successful of all.