Don't look back

By MARK FRYER

Of all the cliches in the investment business, one of the most timeworn is the one about past returns being no guide to what may happen in the future.

But, no matter how often those words are repeated, past returns continue to exert a powerful hold over many investors. Something about human nature makes us want to believe that, if something has done well, it's likely to go on doing well.

Many investment managers are only too happy to cater to that viewpoint, promoting themselves in advertisements that emphasise how well they've done in the past.

But a new Australian study shows once again that backwards is not a good place to look when choosing investments.

The study, issued at the end of last month, was done by the Australian Securities and Investments Commission (ASIC), which regulates the investment business across the Tasman.

The commission looked at how the financial services business uses past performance figures to market things such as managed funds, superannuation schemes and financial advice.

And it has has suggested new rules to tighten up the way in which those figures are used.

Its proposals apply only to Australia, but the commission's report has lessons for investors anywhere, and shows why past performance figures should be taken with a very large grain of salt.

But first, why is it such a bad idea to judge investments on their track record? After all, that's exactly what you'd do if you were trying to pick a winning racehorse or a rugby team to bet on.

To answer that question, the commission asked a group of academics to look at the many studies which have tried to measure the importance of past performance.

After assessing about 100 studies done over the past 20 years in various countries, the committee concluded that relying on past performance as a way of choosing investments is a bad idea for the simple reason that it doesn't work.

What those studies were looking for was evidence of "persistence", or any tendency for investments that have outperformed comparable investments to go on outperforming.

If good performance does persist, then it would make sense to choose investments by looking at their track record, and putting your money into those that have beaten their peers.

But, in general, the studies found that good past performance seems to be, at best, a weak and unreliable way of predicting good performance over the medium to long term.

About half the studies found there was absolutely no relationship between good past performance and how well investments did in future.

Others uncovered some relationship, but typically only in the short term rather than the longer term, which is presumably what most investors are interested in.

And some studies suggested that, while good performance did not tend to persist, poorly performing funds were slightly more likely to go on doing badly.

To cite just one of the studies the experts examined, US researchers looked at almost 300 sharemarket funds which advertised in two investment magazines. On average, in the year before the advertisement, they had outperformed comparable funds by 6 percentage points. But in the year after the advertisement, the funds that advertised did worse than comparable funds by about 0.8 percentage points.

Other studies used different approaches but the overall message was much the same: good performance in the past doesn't mean good performance in the future.

This may be news to many investors, but it's hardly a revelation. A similar study by the Financial Services Authority in Britain two years ago found much the same thing.

The authority is now looking at controls which would prevent British advertisers making any link between past and future performance, and force them to give more prominence to warnings about the past being no guide to the future.

But whatever the academics say, investors like to look back. The Australian report cites one survey from last year, which asked investors to choose the most important thing to consider when selecting an investment.

More than half of those surveyed - 54 per cent - said past performance was the thing that mattered most. Surveys in Britain have come up with similar results.

The commission says professional advisers also place a heavy emphasis on past performance, not only to help them select funds but also to help justify those choices to their clients.

Our fondness for looking back should not come as a surprise. After all, as the ASIC report observes: "It goes against human experience in almost every other aspect of life where past performance is a strong guide to future performance ...

"In most other areas of human experience, past patterns tend to persist. In selecting investments, it is not surprising that consumers apply (or can be led to apply) familiar concepts such as backing recent winners."

But the report says the idea that performance does not persist is not as illogical as it may seem. For one thing, investment management is a competitive business, and if one manager is doing better than the pack, the others will try to copy its methods and/or headhunt its staff.

As well, a large part of that past performance may have been good luck rather than good management.

And the methods which work best in one set of market conditions will not work best at other times. The recent past provides a perfect example; managers who poured money into high-tech shares did spectacularly well before the dotcom crash of 2000, but anyone who has persisted with that strategy will have been sorely disappointed.

One of the dangers of concentrating on past performance, says the ASIC report, is that it encourages investors to chase returns by buying funds which have done well in the recent past.

That strategy, it says, is likely to have the opposite result to the one intended, because it means investors are "buying high and selling low".

Too much emphasis on past returns can also give investors an unrealistic view of what they are likely to earn, says the commission.

And while investors are busy trying to work out which investments have performed best in the past, they are less likely to be paying attention to other questions, such as the risks involved and the fees they will have to pay.

The commission has suggested some changes to tighten up the rules on the way past performance figures can be used in advertising. Again, those rules are specific to Australia, but the changes provide some lessons for would-be investors anywhere:

* If you insist on looking back, at least make it a long look. The ASIC says that where an advertisement shows performance figures, for however long a period, it should also include the return for the past five years, or the annual return for as long as the investment has been running, if that is less than five years.

* Be wary of "hypothetical" or "simulated" past performance figures. Some new investments try to entice investors by showing how much the investment would have earned if it had been operating for a certain period. Sometimes the figures are based on the returns from a similar fund, or on a certain market index, or on computer models of how a particular strategy would have worked in the past.

There are many ways in which those sort of figures can be misleading, says the report, and ways of manipulating them with the benefit of hindsight.

* Be cautious about performance figures that aren't up to date, or don't clearly show the period in question.

* Check whether the performance figures take into account any fees that investors would have paid. As much as possible, says the report, performance figures should be shown after deducting fees, and where that is not possible, would-be investors should be told there are other fees which have not been deducted.

* Ask about the risks as well as the returns. The commission says advertisers should show the risks an investment involves, perhaps by comparing its risk level with other investments, or by including a graph showing how volatile past returns have been.

* Be careful about figures that compare returns from competing investments. Those sorts of comparisons can be misleading, says the report, if they simply compare performance and don't also take into account how risky the competing investments are.

* To contact Personal Finance Editor Mark Fryer write to: Weekend Herald, PO Box 32, Auckland.

Email: mark_fryer@nzherald.co.nz

Ph: (09) 373-6400, ext 8833

Fax: (09) 373-6423.