Secrets of Success: Make sure it really is active management

One theme that came out strongly from my recent analysis of Anthony Bolton's career at Fidelity is the importance, if you are proposing to pay for an actively managed fund, of having an investment approach that creates the potential for you to outperform the market by a significant degree. That might sound obvious, but you would be amazed how many funds sold to the public for hefty active management fees turn out not to fall into that category at all.

There are, as has been argued in this space many times, perfectly valid reasons to own both index-tracking funds and actively managed funds: it is not simply an either/or choice. Many investors will sensibly choose to use cheap and efficient index funds to obtain the majority of their exposure to the main markets and asset classes, and then top this up with one or more carefully selected actively managed funds in specific market or sector niches.

This makes sense because there are times when mainstream equity index funds will lag behind a fair number of actively managed funds because of style and other effects. The period since the end of the bull market in 2000 has been a good example, with smaller companies and a value approach outperforming the market by a significant margin. As it can be quite difficult to capture these effects through suitable index funds, a proven actively managed fund with the relevant approach can make sense (though the fast-growing range of sector, style and country exchange-traded funds on offer can also often do the same job just as well for lower fees, and will become increasingly attractive as an alternative).

What you must avoid is the great mass of dysfunctional funds that charge standard active management fees for doing something little different from the market. To illustrate how big a problem this can be, I recently did a simple spreadsheet analysis of the 400 or so mainstream UK equity unit trusts/OEICs in the Fund Expert database.

The metric I used was a simple statistic known to its friends as r-squared. This measures the extent to which the performance of a fund can be "explained" by the movement in the stock market over comparable periods. A fund whose performance can be entirely explained by the movement of the market (as should be the case for the best index funds) will have an r-squared of 1.0. All other funds will have an r-squared of less than 1.0, the precise number being dependent on how divergent the investment approach of the fund is from the market.

Bolton's Special Situations fund has consistently displayed a low r-squared relative to its competitors. Over its 25-year history, the fund's average score on this measure has been around 0.7. This is consistent with Bolton's belief that the only way to outperform the market over time is to do something different from the market, which is reflected in his contrarian value-oriented investment approach.

How does this figure compare with other top-performing funds? To check that, I first screened out the 10 per cent of equity funds with the highest Sharpe ratios over the past three years. The Sharpe ratio is another statistical measure, used in the fund business to measure risk-adjusted returns, by relating the returns of a fund to its volatility. The next step was to look at the range and average of these funds on a variety of measures, including the r-squared figure. The index chosen to represent the market was the FTSE All-Share index. The results are shown in the table.

To qualify for a top 10 per cent ranking, you needed to have a Sharpe ratio of more than 0.3 per cent, quite a high figure for this kind of analysis, reflecting the fact that the past three years have been a relatively good period for active management. (There was a point during the depths of the bear market when less than 10 funds out of 4,000 had a positive Sharpe ratio).

The great majority of funds today (65 per cent) still have Sharpe ratios of less than zero, underlining the well-established fact that most funds do not offer value for money.

But the average r-squared for the top-performing funds was as low as 0.6, with the range running from 0.46 to 0.87. These funds were doing something very different from the market as a whole, though in many cases this was simply because they were operating in the small and midcap sectors of the market, which were themselves less correlated with the All-Share index than normal.

But at least these funds had an opportunity to deliver something that might add value to those who invested in them. About half the 400 UK equity funds, by contrast, had a three-year r-squared of more than 0.9, making it virtually impossible for them to outperform the market by a sufficient margin to produce a positive risk-adjusted return or justify their active management fees. The moral is not that a low r-squared is necessarily a good predictor of future outperformance. Like all backward-looking measures, it only tells you what worked in the past. Doing something different from the market can work both ways: a poor fund manager who deviates from the market by a large margin may well end up losing, for the simple reason that beating the market is difficult.

But if you are going to pick a fund that has a chance of justifying active management fees, it will certainly need to show a low r-squared over the term you choose to hold it.

Funds that have demonstrated that in the past and have an investment philosophy that is consistent and stable should then come into consideration. But the scores of funds that have r-squareds of more than say 0.9 (about half the total) can and should be struck out of consideration, unless they are index funds.

jd@intelligent-investor.co.uk