Anthony Hilton: Timing is all if smart beta strategy is to win

On a primary level, there is nothing smart about smart beta because it simply formalises what some individuals have been doing for years. There have always been factors that appeal to some kinds of investor but not others. Some like to buy companies that have hit a rough patch and been left friendless; others swear by small companies and never look at the giants; another group likes big dividends, and yet another only follows firms with a powerful established franchise.

Smart beta today goes further than that. A lot of technical skill and computing power is required, and what the providers offer can be mathematically and operationally very difficult to put into practice — which they like because it means they can charge more.

The wisdom of crowds is that smart beta’s advocates may be on to something because people are queuing up to buy. On the other hand, the experience of lemmings suggests that there may not be a happy ending.

If anyone has the tools to decide between these extremes, it is the Credit Suisse researchers because over the years they have put together the world’s most comprehensive database of long-run stock-market returns, giving them the tools to test different smart beta strategies over unrivalled periods of time. The question was does smart beta — or factor investing, as they prefer to call it — really do what it says on the tin? Their report, published this week, contains the results.

The honest answer seems to be that sometimes it does and sometimes it doesn’t. That’s probably not what pension funds and private investors want to hear, but it does have a depressing familiarity. It is genuinely hard to give a definitive answer.

In the long run, they all seem to work but there can be periods of 20 years or so in which they don’t work at all. When you start measuring and when you stop can make a huge difference to how impressive or otherwise the results are.

The researchers found, for example, that size makes a difference because small firms beat the pants off large ones for 20 years. But then for the next 20 years, it was exactly the other way round and all the glory went to the big companies.

It is not always that much fun in the short term, either. The team has tested the five most commonly used factors in smart beta — value, momentum, size, low volatility and income — and plotted on a grid how they have done since 2008. There is no discernible winner over the period — one factor dominates one year only to be a dog the next. Low volatility had three years at the top and three years at the bottom.

Size also had three years at the top but only one at the bottom, as did momentum. Value had one year at the top and five at the bottom. Income never made it to the top but was never at the bottom. It spent all 10 years in the middle three of the five bands, although with a bias towards the lower.

The authors conclude that factors always make a difference but that you cannot guarantee it will be for the good.

There is a value effect whereby such stocks perform differently from growth ones. There is a high-yield effect whereby these perform differently from low-yielders. There is a size effect between large-caps and small-caps, and there is a momentum effect whereby stocks with relative strength generate returns that are different from stocks with relative weakness. These relationships affect portfolios. They should not be ignored, but that does not automatically mean they should be chased.

There are two other things to consider. The first is that several smart beta strategies require a lot of trading, and that obviously adds to costs. The second is that the factors tend to behave differently, or certainly in a more subdued fashion, when too many people get interested in them.

How they would behave if everyone rushed for the exit at once is an open question but could be alarming. Momentum investors, for example, have been seen to lose a mint when the market suddenly changes direction.

So the answer is that smart beta is certainly smart some of the time, but it does not have all the answers all of the time. The smart investor will still have to guess when might be a good time to try it — and then which slant to choose.