There is a widespread belief that the majority of active funds underperform their benchmarks, meaning that investors are better-off in low-cost index trackers.
However, our data shows that those portfolios which make meaningful bets against their index in the IMA UK All Companies sector have a strong tendency to outperform over the last decade.
We took all those IMA UK All Companies funds benchmarked to the FTSE All Share and examined their record over 10 years.
The r squared measure tells us how much of the fund’s performance is explained by the performance of the index.
Those with an r-squared score of less than 0.75 to the index were excluded as not having the appropriate benchmark.
We then took all those with a tracking error of at least 5 per cent over that time to exclude index trackers and those that make minimal bets against their index.
There were actually a couple of index funds with tracking errors even higher than this, which would be a concern to those looking for a passive fund.
The results show that the composite of the active funds significantly outperform over 10 years, returning 172.91 per cent against the 165.35 per cent of the FTSE All Share, dividends reinvested.
Performance figures are net of charges, meaning that the outperformance from the actual stock-picking would be higher.
Performance of active funds versus index over 10yrs
Source: FE Analytics
The average All Share index tracker has made just 146.3 per cent over that time, 19.65 percentage points lower than the benchmark.
The picture is more complicated over five years. The composite of all funds with a tracking error of at least 5 per cent has slightly underperformed the All Share over that time.
The FTSE All Share has made 36.5 per cent over that time as the average fund with a 5 per cent or higher tracking error over that time has made 34.5 per cent.
Performance of active funds versus index over 5yrs
Source: FE Analytics
However, a majority of such funds have added value, with 24 recording a positive information ratio against 21 that have a negative information ratio – 53.3 per cent.
The information ratio gives us the value added per extra unit of risk that the manager has taken on versus the index.
And the active funds have outperformed the index trackers, which made only 31.9 per cent over that time.
Performance of index and trackers over 5yrs
Source: FE Analytics
Interestingly the period of highest outperformance came in the bull market from March 2003 to July 2007, when the average actively-managed fund made 116.52 per cent against the FTSE All Share’s 104.67 per cent.
Performance of active funds versus index 2003-2007
Source: FE Analytics
There is no clear pattern of active funds doing better in up markets, however, with the average active fund marginally underperforming in the bull market since 2009.
Our figures highlight the dangers of looking at sector averages without digging into the details.
There are a number of reasons why managers would be tempted to stick relatively close to the benchmark.
By making fewer bets against the benchmark there is less likelihood of large underperformance, which could ultimately lead to a manager losing his job.
Rob Gleeson, head of research at FE, says that he isn’t surprised at the outperformance of genuinely active funds.
He points out that few people focus on the arbitrary nature of the stock market indices.
Most major indices are market-cap weighted, meaning that they are constructed by weighting each stock by the value of shares in issue. Indices are then rebalanced regularly, typically once or twice a year.
Research from Cass Business School found that a market-cap weighted index was one of the worst ways to construct an index in terms of the returns provided. In fact, they found that a randomly selected group of stocks beat the major indices over long time periods.
They programmed a computer to randomly pick and weight the 1,000 stocks they were studying 10 million times over each of the 43 years of the study, and almost every portfolio outperformed the market-cap weighted index.
The researchers also showed that a traditional market-cap weighted index of US stocks would have performed worse over the past 40 years than one based on each of 13 other metrics including sales, dividends and annual cash flow.
In a series of articles in coming weeks FE Trustnet will be looking in more detail at how actively-managed funds fare in different IMA sectors.