FOOL'S EYE VIEW
Choosing An Investment Fund

By James Carlisle
January 25, 2002

Actively-managed investment funds are big business. No matter how much the Motley Fool talks about the benefits of a cheap index-tracking strategy, the reality is that most people's money gets directed to 'managed funds'. Part of the reason for this is that the managed funds have more money at their disposal for marketing, but it also comes down to the fact that people simply aren't happy with being average and average, before costs at least, is what a tracker is designed to achieve. So, if you want to buy a fund that's not average, how should you go about it?

The first thing to do is decide what you're looking for. If you've decided that you don't want to be average, you need to think about how exactly you want to differ from the average. For instance, you might decide that you want to have a bit more of your money invested in drugs companies. It doesn't take a genius to work out that this would lead you towards funds that had a greater than average amount of their money invested in drugs companies. Most likely, what you'd do is keep most of your money spread generally across the stock market and have a small bit, perhaps 10 per cent or so, in a pure 'pharmaceutical fund'.

Human nature being what it is, though, most people simply want a fund that has a chance of coming out ahead of the average. It's particularly fashionable at the moment after two years of poor stock market returns and many people theorising the returns for the next 20 years will be lower than the last 20. The difficulty with this is that, almost by definition, a fund that has a good chance of coming out ahead also has a good chance of coming out behind. In other words, to get the extra returns, you'll need to take on some extra risk. Obvious examples of this is include the technology funds that were sold heavily a couple of years ago. Sure they had a chance of coming out ahead but, as we've seen, they can also miss by a long way on the downside.

Sticking like a rash

What you're really looking for when comparing investment funds are factors that show persistence. It sounds a bit like some sort of rash and that's not a bad way of looking at it. You want to find things affecting a fund's performance that will stick to it like a rash. You can see all sorts of patterns in the historic performance of a fund, but unless it's something that's likely to persist into the future, then it's no good to you at all.

One of the most obvious things that people look for in a managed investment fund is its recent performance record, say up to five years or so. Is that likely to persist into the future? The answer, (according to research undertaken by the WM Company, amongst others) is a pretty resounding 'no'. So, even though it's tempting to pay a lot of attention to recent past performance, it's unlikely to tell you very much.

As a record of consistently good performance gets longer and longer, though, you might begin to conclude that it is beginning to tell you something. Not because of itself, but because it might begin to suggest that there's something else that's persisting in the performance. A good example of this is the legendary investor, Warren Buffett. After 30 or 40 years of consistently good performance, there are few people that would put it all down to luck. More likely there's something else that's persisting. Perhaps it's that he keeps costs very low at his investment vehicle, Berkshire Hathaway (NYSE: BRK.A). More likely than not, though, he's just very good at picking shares. But it takes a very long time to become confident about this and not long after he's become everyone's favourite fund manager, he must be close to retirement.

The National Lottery is an interesting example of what I'm talking about. In its first year of operation, the number 5 was drawn an impressive 10 times, compared to a shocking performance from the number 39, which appeared only once. Would you conclude from this that the number 5 is any more likely to appear than the number 39 in future? You certainly shouldn't! Some numbers have to be drawn more than the others. The most unbelievable thing of all would be if all the numbers were drawn exactly as much as expected (especially since they should each come out 6.3673469 times). But if, after 20 years, the number 5 was still coming out ten times as often as the number 39, then you'd be thinking it might be heavier or something.

Fund Rating Services

All over the Internet these days, you can see things called 'fund ratings' cropping up. It's not surprising, really. As I said at the beginning, this stuff is big business. They all come up with wacky new ways of comparing funds, but all you should care about is thinking which methods are likely to show persistence.

Take this one from ThisIsMoney. Each month, it splits the funds up according to whether they've performed in the top ten per cent, or the second ten percent, etcetera all the way down to the bottom ten per cent. Points are then awarded according to where each fund finishes up, with ten points going to the 'top' funds and 1 point going to the ones at the bottom. They then take the average of the point scores for the last 5 years and hey presto! It all sounds very clever and you end up with a nice list that shows abc fund at the top with an average score of 6.72 and xyz fund at the bottom with a score of 4.43, but persistence?! No way!! It's no better than ranking the appearances of lottery balls.

Other 'fund ratings' make an attempt to be more scientific. FT Fund Ratings, for example goes through a complex process based on 'arbitrage pricing theory'. Essentially it groups funds together according to whether their performances appear to be affected by the same things. It does actually sound like pretty good stuff. The trouble is that it's all so very complicated. I reckon I'm not bad at following this sort of thing, but I can hardly keep up with the basic explanation. Then they tell you that it takes a computer a week to work it all out and vooom, there it goes, over my head.

To make matters worse, the actual figures that the FT ends up with look pretty odd. Apparently, 98% of the FTSE All Share's industry exposure comes down to the financial sector and 'non-cyclical consumer services'. That might be how it's looked for the last three months, but it's hard to figure on much persistence. I might (very well) be missing something, but if you don't know where the numbers are coming from, then it's hard to have much faith in them.

Morningstar is another fund rating service that uses a highly technical methodology that falls down because, frankly, who knows what it means. They group together funds with 'genuinely similar investment policies' and look at risk-adjusted returns using the Stutzer index. This is then put into a pot, along with data on charges, given a stir and out pops a rating from one star to five stars. If you understand the Stutzer index, it's unlikely that you'd be interested in Morningstar's fund ratings.

Now I'm probably being a little unfair on some of these ratings services. As I've said, some of the stuff they're doing looks reasonable. The trouble is that we've got used to a financial services industry that's not afraid of ripping us off and which is more than happy to plug its wares based on irrelevant factors. In any case, some of the services ignore the low-cost investment trusts and exchange traded funds that you'd expect to perform the best. A little scepticism is fair enough!

In amongst the confusion, however, there is undoubtedly some interesting information. The FTFundRatings service provides information on charges (provided by the well-respected Fitzrovia) and all the evidence suggests that this is the most persitent factor of all in fund performance. The FT's service also includes investment trusts and the information on charges confirms how cheap some of the larger investment trusts are, with Alliance Trust, for example, having a ' total expense ratio' of just 0.24%. That's even less than your typical index tracker.

The information on how and where a fund's money is invested, if you can make sense of it, might enable you to target particular types of fund. Sometimes, they'll actually tell you what the fund's biggest investments are, although this information can quickly become out of date.

All in all, there's nothing wrong with using these services to sort through the world of investment funds. But you need to be very careful about what you're looking at and avoid taking too much on trust. When you've whittled down to a final few, there's really no substitute for getting hold of a fund's most recent annual report and looking yourself for the factors that you think might show persistently good performance.

More: The Fool's ISA Centre; Investment Trusts and Unit Trusts discussion board