Money: Look back at your peril ...

Choosing investments by studying past performance? You might be better off taking a punt on the horses.
MARK FRYER and BRENT SHEATHER report.

Trying to choose a managed investment? Here's a way to make the task a little easier.

Before ploughing through the investment statement, tear out any glossy graphs showing how well the fund has done in previous years, tables of historic returns or any other references to past performance.

Throw the pages away, use them to light the barbecue - just don't read them.

Why? Because historic performance is a particularly useless piece of information to use when trying to choose a managed fund.

If that idea goes against the grain, it's because most of us can't resist the temptation to see patterns in the past. The fund produced 20 per cent a year for the past three years? Let's see, if we put in $5000 and they keep that up for the next five years we'll have ... um, $12,440.

Sure, there's probably a note saying "past performance is not necessarily a guide to future returns," but who cares about the small print?

Comforting as it may be to hope that past returns have a good chance of repeating, the evidence is against you.

The latest argument against paying attention to past performance comes in a report from the Financial Services Authority (FSA), which regulates the British financial services industry.

The FSA set out to answer a simple question - is a managed fund's past performance any use to investors trying to choose the best place to put their money?

The short answer? No.

In the jargon, what the FSA was looking for was "persistence" - any tendency for above-average investment performance, or below-average performance, for that matter - to be repeated.

As the authority says, if past performance does persist, you would be silly not to consider it when choosing a fund manager.

The trouble is, it doesn't persist.

The FSA studied the performance of British managed funds investing in shares, and also looked at evidence on such funds in the United States.

As well as asking whether investors should pay attention to past returns, the authority was trying to decide whether its own comparison tables of managed funds should include historic performance.

It started by reviewing other research in this area, and found only weak evidence that performance persisted.

What persistence there was tended to be short-term, and it was the poorer managers whose results were most likely to persist.

The FSA also did its own research. While some of its methods should be discussed only among consenting accountants, some are comprehensible to the rest of us.

For example, the study took all the unit trusts in the UK and divided them into five groups, based on their performance in the previous year - the top 20 per cent of funds, the next 20 per cent and so on.

If performance really did persist then the top group, when looked at a few years later, should still be better-than-average.

But that wasn't the case. Looked at a few years later, the group that had made up the top 20 per cent was producing returns which were about average.

The time it took for funds' performance to return to the average got shorter and shorter in more recent years.

After this, and much more sophisticated analysis, the FSA concluded that private investors are wasting their time poring over performance figures.

Some funds do out-perform their competitors, says the FSA. That's because if you take a big enough sample, sheer chance is likely to throw up some funds which do better than others for several years in a row.

But you can't use past performance to predict which funds those will be.

Nor does it mean fund managers are lazy or incompetent. On the contrary, the harder the competing investment managers try, the more difficult it becomes for any of them to consistently beat the pack.

On a smaller scale than the FSA study, investment consultants Aon Consulting this year asked much the same question about New Zealand fund managers.

What if you had begun investing in 1978, asked Aon, and every year you had given your money to the manager who produced the best results the previous year?

If you had started with $1 million, by 1999 that strategy would have seen you with investments worth $17.4 million.

Not bad - except that your investments would have been worth $18 million if you had been content with the average return earned by all the managers Aon studied.

In fact, chasing the best performance would have yielded much less than $17.4 million, because Aon's study ignored the cost of switching managers every year.

Aon's conclusion: "Past performance alone is not a guide to future performance."

A Massey University study of local managers in 1996 also found no evidence that using past performance to choose funds would result in better returns in future.

All very fascinating, but why does it matter for those of us without post-graduate degrees in statistics?

One reason it matters is that fund managers often emphasise past returns when trying to sell their products. But, as the research shows, you would be wise to ignore those historic figures when deciding where to put your hard-earned money.

It also matters because technology, especially the internet, makes it easy to switch from one fund to another, in an attempt to follow the "hot" funds.

In the US, the average time investors leave their money in mutual funds - their equivalent of unit trusts - has fallen from 10 years in 1969 to 2 1/2 years last year.

But the FSA approach shows fund-chasing is no guarantee of beating the market.

The best approach is the old boring one - find the best fund, or funds, for your money, then leave it alone for as long as possible.

Mark Fryer is the Herald's personal finance editor. Brent Sheather is a Whakatane sharebroker.

A full copy of the FSA report can be found at: www.fsa.gov.uk/pubs/occpapers/op09.pdf