Since S&P Dow Jones issued its first SPIVA report 2002, this twice-yearly scorecard on active fund performance has consistently shown that, in the US, actively managed mutual funds routinely underperform their respective benchmarks in every asset class and almost every style. But, as Craig Lazzara, global head of index investment strategy for S&PDJI explains, the US figures are remarkably consistent with active fund performance all over the world.
When my documentary How to Win the Loser’s Game went online in November 2014, a typical response I had from the UK fund industry was that underperformance by actively managed funds is primarily a US problem. But SPIVA scorecards have repeatedly shown that this is a global phenomenon, haven’t they?
Yes, they have. It’s not just a US problem. We initiated a SPIVA report for the UK and Europe generally in 2014 and the results are consistent with what we see in the US. Generally speaking most managers underperform. There may be some style drift opportunities, for example, if a mid-cap manager drifts up or down the cap scale. But the results for Europe are very similar to those for the US. In fact, wherever we do SPIVA reports — Canada, Latin America, Australia, Asia — it’s always the same. The reason why, most of the time, the majority of active managers underperform, is that the aggregate portfolio of all the active managers is the same as the aggregate portfolio for the market. So the only way for one to be above average is for another to be below average. And yet investors must incur management fees, trading costs and so forth in the pursuit of above-average returns, and those costs are a deadweight loss that bring the average down.
Does it surprise you that the take-up of indexing investing is much lower in the UK and Europe generally than it is in the United States?
No, I don’t think so. I mean, the concept started in the US. The first index funds, for institutions, came in around 1970, and the first mutual fund in 1976. I guess the first indexer in the UK was Legal & General, and that goes back to maybe the mid-1980s. So there was a considerable gap, and that head start is a big part of the reason. The investment business is more heavily professionalised and less retail-oriented in the US than the UK, so that has also delayed the adoption of indexation, I think.
You’ve recently done research on dispersion. Briefly, what is dispersion? And what impact does it have on fund performance?
Dispersion basically means the degree of difference between the best performers and the worst performers. It’s the standard deviation of returns in an index. You can also think of it is as the gap between the winners and the losers. In a period of low dispersion, even if you posit that a manager is skilful and can pick the best stocks in a market, the reward for doing so will be less. You can think of dispersion as measuring market opportunity, or the measure of the value of skill. It doesn’t tell you anything about the existence of skill. I mean, a manager is either a good stock picker or is not a good stock picker. If he’s a good stock picker, he’s going to show that ability much more in a period of high dispersion than in a period of low dispersion. So it’s not surprising that in a period of low dispersion, which is what we are in now and have been in for the past several years, the percentage of managers who underperform is higher than usual. That’s simply because, even if you acknowledge that some of them are skilful, the value of that skill is less because the dispersion is less. 2014, for example, which was the worst year for active managers in the US that we’ve recorded also had the lowest dispersion that we’ve seen for the S&P 500.
There’s been plenty of discussion lately about the relative merits of cap-weighted and or equal-weighted indexes. In your opinion, which is best?
Well, as a former product manager for the S&P 500, I would never say that there is anything wrong with cap weighting, but the difference is this.. A cap-weighted index tells you the performance of the average investor weighted by the average amount of cap, or by the amount of capital that he controls. An equal-weighted index tells you the performance of the average stock. Now, what we find in the US and Europe is that, typically, an equal-weighted index outperforms its cap-weighted counterpart.
Why do you think that is?
Arguably, there are three reasons that contribute to that. One is that by definition, mathematically, the equal-weighted index would have a lower average capitalisation and typically, other things being equal, there’s a small-cap effect. So equal weight has that going for it — and that’s mathematically certain. The second thing equal-weight has going for it is that it is typically a more value-oriented portfolio, simply because more expensive stocks tend to have higher caps. So equal weight will tend to give you a value bias.. That’s not mathematically certain, but it’s not bad. The third thing that equal weight does is that it automatically rebalances. In the case of the US, say the S&P 500, the index rebalances every quarter. What that means is that every quarter, you sell your winners and buy your losers, and bring everything back to one 500th target weight. And typically that rebalancing is a source of value added as well. So those three things together kind of account for the equal-weight effect.
So, if you’re an active investor, how should you benchmark your portfolio? Equal weight or cap weight?
Equal weight is certainly something that you should pay attention to, and there are two reasons for that. One is that the equal-weight index tells you what a manager could do, simply by selecting stocks at random. So if an active manager underperforms at equal weight, what that basically says is that a bunch of monkeys could do as well.. In fact the monkeys could do better because their random selection would outperform. The second reason why equal weight is a better benchmark is that most active managers, if you look at the weights in their portfolios, are far closer to an equal-weighted portfolio than to a cap-weighted portfolio. What I mean by that is, if an active manager says that he is going to assemble a portfolio of 50 stocks, his default position is 2% each. Some are going to be more and some are going to be less, but he starts out with an assumption that he’ll have 2% each; in other words, he starts out with an assumption of equal weight. And in that sense it makes sense to compare him to an equal-weight benchmark as well.
If you enjoyed this article, and haven’t yet read Part 1 of the interview with Craig, you can find it here: SPIVA: What is is and what it tells us — an interview with Craig Lazzara (Part 1)