The following originally appeared in www.moneymanagement.com.au
Reprinted with permission.
Throw Out the Fund Form Guide
By Leeanne Bland
www.moneymanagement.com.au
August 2005
We are all well trained and responsible financial services professionals, so of course we all chant "past performance is no guarantee of future returns" when the occasion arises. But there's no denying that investors like to know the past performance of a fund.
It's a behavioural phenomenon that hasn't been lost on Professor Tim Brailsford, head of the UQ Business School at the University of Queensland, and a member of the PortfolioConstruction Journal editorial board. We asked him to explain, from a behavioural finance perspective, why it's human nature to believe that fund managers have 'form'.
What do you mean by form?
When it comes to sports, we all like to have a go at picking winners.
A crucial ingredient in the debate is form - how a player, team or horse has performed in the past becomes an essential element in our rationalisation about what will win in the future. We all believe that form matters.
Do fund managers have form?
For some time now, there has been detailed examination of the entrails of performance figures in the funds management industry. There is a large volume of published studies that reach quite similar conclusions despite the variety of markets and time periods that have been studied.
The early studies compared fund performance against market benchmarks and overwhelmingly concluded that managers did not beat the market on average.
Recent research is a little murkier in terms of overall conclusions, as the evidence has become muddied by different models and new performance measures.
However, it is still fair to say that the majority of studies have shown that managers, on average, do not consistently beat the market. In addition, there is generally little consistency in performance - that is, the relative position of a fund compared to peers in one period has little correlation with its future relative ranking.
The exception is at the extremes - there is some evidence that funds in the top decile over a one-year period reappear in the top decile in the next one-year period, and there is stronger evidence that bottom decile funds persist.
So the evidence tells us that there is generally not much to be gained from looking at the past performance of most funds - that is, form is largely irrelevant in picking future fund performance.
So why do we behave as if form matters?
If we take a look at how funds promote their products, there is invariably some mention of past performance - if not an explicit set of figures presented in an appealing manner showing upward trajectory. The reason is that ultimately the investment decision is driven by an individual mind.
It is now well accepted that individuals suffer from loss aversion - that is, they tend to downweight unrealised losses in the (often false) hope that those losses will somehow turn around. Consequently, they hang on to losing stocks for too long.
For instance, evidence of this is demonstrated through studies of initial public offerings (IPOs). Research that has examined trading activity surrounding the early listing days of IPOs shows that when the stag profit is greatest, investors trade a greater percentage of their stock compared to listings where the IPO initially trades at a discount to subscription.
Loss aversion interacts with what is known as cognitive dissonance - the mental conflict that individuals experience when placed in the situation where evidence contradicts their recollection of events that relate to their own past decisions. People have a tendency to consider their past decisions as better than they actually were.
In relation to the investment markets, the recollection of our own past performance is invariably better than reality.
This has been proven in experiments where both novice investors and professional managers had to recall past performance. Both groups overestimate their actual returns, although the professionals are generally more accurate.
Then add hubris to the mix - that is, people have a tendency to overestimate how much they really know. We tend to overvalue anecdotal information.
We also suffer from an illusion of knowledge, whereby the attainment of information is associated with greater knowledge, even if we don't know how to interpret that information properly. And we suffer from an illusion of control, whereby we are poor at distinguishing outcomes caused by chance from those caused by skill.
Attribution theory tells us that we tend to attribute good outcomes to our skill and decision-making, and poor outcomes to chance, irrespective of reality.
We also suffer from myopia, or a fixation on the short-term - to use an analogy, we focus too much on a batsman's last few innings and forget their career average. In financial markets, this is often translated into overreaction.
An influential paper by DeBondt and Thaler showed that extreme stock returns tended to undergo reversals in cycles of three to five years. In brief, the worst 10 per cent of performers tended to reverse their performance and become winners in the following three years.
Subsequent studies that sought to improve the methodology and repeat the experiments on different markets (including Australia) generally confirmed the anomaly.
Rational explanations have focused on market reaction to news, and particularly earnings news - that is, the market overreacts to events, including a shock to prices, and it takes a series of earnings figures to restore confidence in the stock, causing prices to drift back to a value more consistent with fundamentals. This could be seen in the wake of September 11, and the resulting price paths of insurance stocks.
So what should the industry be doing in light of this?
If investment was easy, then many of us would be looking for employment elsewhere. The fact is that financial markets are extremely difficult beasts to tame. They possess an unpredictable nature; are subject to a vast array of influences; and are underpinned by a massive information set.
For many investors, managed funds are the door to financial markets. Predicting the performance of managed funds is an extremely difficult task, and the evidence tells us that the majority of fund managers do not carry their form forward.
In this environment, the way in which the industry deals with past performance figures is important. Already, the Australian Securities and Investments Commission has issued guidelines on the topic. Faced with a competitive industry, fund managers have natural incentives to chase clients and their funds.
However, we need to appreciate the way in which investors process information and how they make decisions.
There are strong moral arguments to even further clean up the way in which past performance is reported and used within the industry. Indeed, an argument could even be made in light of the evidence that past performance should not be used at all when promoting managed fund products.
However, such an outcome is unlikely to eventuate as, paradoxically, investors will demand this information.
Nevertheless, the industry needs to continually remind itself that the art of investment is not a precise science, and that investors have a right to know just how imprecise the reality is.
Leeanne Bland is editor of PortfolioConstruction Forum.
4 August 2005