Is past performance really no guide to the future?

Past performance is no guide to future returns we are warned by the finance industry at every opportunity. But is that always the case?

Financial returns are the net product of two factors; the return of the asset class (known as beta) and additional returns (known as alpha) from selecting only a few, or avoiding some or even most, securities within that asset class. Alpha is not always positive of course.

There is now a large body of work that includes the perennial Barclays Equity Gilt Study, the Credit Suisse Annual Returns Yearbook and books such as “Triumph of the Optimists” by Dimson & Marsh, that analyses data stretching back over a century that have provided the basis of modern investing practise for many years. While returns for individual years can vary a lot these studies provide solid evidence for the sorts of returns specific asset classes, such as UK equities, can generate over decades. Most people in the finance industry now recognise that the past performance of asset classes does give an indication of their likely future returns, at least in relative terms. It is widely accepted that, over long periods of time, equities will outperform bonds which will in turn outperform cash.

To maximise these returns of the asset class, the beta, it is understood, based on simple maths and historic data, that costs should be kept as low as possible, that the investment should be made for as long as possible and that income should be reinvested.

Where life gets more complicated is analysis of alpha; the additional (hopefully) return over and above that of the asset class generated by selecting a subset of securities from within that asset class.

Funds that actively seek to outperform the asset class they are benchmarked to by doing this run two risks. First that the stocks they select do not do as well as the rest and second that the ones they avoid do not perform better than the asset class a whole. If the manager gets it right the returns will be better than the asset class, although at the cost of higher risk. Conversely, if he gets wrong the increased risk will reveal itself as a lower return than the asset class.

The tricky bit is quantifying how these securities are selected and whether the process is systematic and repeatable, in other words based on a set of rules that can be followed by others. Conventional wisdom is that this part of fund management is the “secret sauce” and cannot be written down as a step by step process. Because it cannot be documented it is assumed it is not repeatable and this therefore is what lies behind the phrase that the past is no guide to the future.

Perhaps therefore it is more accurate to say that past performance is no guide to future returns in the pursuit of alpha. However, there is a strong argument to say that the same logic does not apply to rules-based funds that seek to deliver beta over long periods of time.

Download the article here