Money: To pick one unit trust from 1,600 you need to do your homework

Once you get started on a subject, it sometimes takes more effort than you thought to get off it. So it is with the issue of unit trust performance. This week, I propose to add some more thoughts on the subject, picking up where I left off last Saturday and adding some new advice from another source.

I mentioned last week some statistical research by an academic at Sheffield University, John Cuthbert, which attempts to cast light on which fund management houses have the best overall track records in the last few years.

He highlighted 11 firms which had demonstrated consistently value-adding performance over three-year periods with the majority of their funds. I should emphasise that these covered general UK equity funds only, and not specialist sector funds (eg property, technology and so on) or international equity funds.

The figures on international equity funds are just as interesting as the UK ones. Mr Cuthbert found that only 11 unit trust providers in this sector have consistently beaten the relevant market index over the last three years with more than three out of five of their funds. (To narrow the field, he only included those firms which had at least four international equity funds in their portfolio.)

The firms that did the best on this measure were Prudential, Norwich Union, Fidelity, St James Capital, Scottish Widows, Cazenove, Provident Mutual, NatWest, Scottish Equitable, Mercury Asset Management and Edinburgh Fund Managers. Two things are immediately striking.

One is that there is hardly any overlap at all between this list of consistent houses and those for the UK equity sector. Not one of the top 10 names in the UK list appeared in the international one. This is partly because some of the best UK fund managers do not have enough overseas funds to qualify for the screening - an honourable mention to Lazards here, which has five equity funds (UK and international), all of which have consistently beaten the market over the last three years. Britannia and Jupiter both boast a 50 per cent strike rate with their international funds and 100 per cent records with their UK funds.

But in many of the cases, the performance simply does not seem to travel well. Several of the consistent international performers, including Prudential, Norwich Union and Edinburgh Fund Managers, have no funds at all which consistently outperform in the UK market.

The second striking thing is that many of the consistent performers with international funds are insurance companies, organisations that are traditionally - for both good and bad reasons - not associated with outstanding investment performance. And who would have thought that one would find a clearing bank in such a list?

Mr Cuthbert's conclusion reinforces his general argument that there is something of a revolution going on in the way that the management of unit trusts is organised. He singles out Cazenove and Lazard for their success in developing a discipline in their asset allocation process which enables them to share ideas across the various teams managing funds in different regions.

He reiterates his point that the kind of changes taking place in the investment management process in many cases makes past performance an even more unreliable guide to the future than it is normally. But it does not invalidate the general conclusion that some firms can achieve consistent outperformance as a result of good organisation and skill.

They do remain a minority, however. When it comes to finding individual funds where the outperformance can be confidently attributed to skill rather than luck, what is striking is how few fund managers can pass this particular test. Mr Cuthbert uses probability theory in order to rate the consistent outperformers.

His conclusion is that there are only six funds in the whole UK equity sector where the results can be attributed to the skill of the fund manager with more or less absolute confidence. These are Johnson Fry Slater Growth, Jupiter Income, Jupiter UK Growth, NPI UK Special Opportunities, Perpetual UK Growth and Schroder UK Enterprise. Another five funds can be said with 80 per cent confidence to pass the "skill, not luck" test.

It is no surprise that these funds include several where the fund manager is willing , or even encouraged, to take big bets against the market consensus. You would not expect funds managed by a conservative institutional fund manager, or someone like the Pru, to feature in this kind of list. They are rare birds.

For those who are really interested in how to go about picking a unit trust from the 1,600 or so that are available, and are prepared to read nearly 300 pages on the subject, I can recommend a new book, Picking The Right Unit Trust, by Douglas McWilliams (FT Pitman Publishing, pounds 21.99). This is well laid out, with plenty of easy-to-follow illustrations, and sensible advice from professionals in the field.

Having gone through all the reasons why reliance on the performance figures you read in the advertisements is of only limited value, his main conclusion is that ordinary investors should aim for funds that have consistent track records of above-average performance (not necessarily the very best), coupled with low volatility. If they cannot do the calculations themselves, best to pay a good IFA, with access to ratings services such as Micropal or Fund Research, for advice.

This is broadly sensible, although in my view Mr McWilliams' technique for picking such funds errs rather on the mechanistic side. (He does have a good chapter on the merits of index tracking funds, however).

As Bill Mott of Credit Suisse, who runs one of the top performing funds at one of the best unit trust houses, unhelpfully points out in the book: "There are no rules. People who try to get investment advice expect it to be like accountancy. But the investment world is constantly changing." And so it is on the fund management side too, as Mr Cuthbert has usefully reminded us.