Equities: the ups and downs of benchmarks

As I begin this column, the FTSE 100 index stands at 5,215. This is important, apparently. Not because it is much higher than seemed possible six weeks ago, but because it is a benchmark against which all performance can be measured. Your wealth manager will quote the 379-point fall in the FTSE this August to explain that your portfolio has merely declined “in line with the benchmark”.

All of which is very reassuring but for one inconvenient fact, summarised last month by Charlotte Thorne, partner at Capital Generation Partners, the family office firm: “Most of the benchmarks that are used to represent the market – the FTSE 100, for example – are actually hobbled. They have an inbuilt distortion that means they can easily be beaten.” And what is this distortion? “Most of the indices used to represent a wider market are capitalisation weighted.”

In other words, when a company’s share price – and therefore its market cap – rises, it forces an index to give more weight to that company’s shares when calculating the overall change, and it forces portfolios benchmarked to the index to hold more of those shares. When a share price falls, this happens in reverse. In fact, it has just happened to the FTSE – it has dropped 27 points.

This is fine for a passive index-tracker fund, but for a portfolio manager “there is a buy-high-and-sell-low dynamic embedded in the benchmark”, warns Thorne, who is not the first to worry about this effect. Wegelin Asset Management, the investment arm of Wegelin & Co, the Swiss private bank, suggests that cap weighting has distorted asset allocation for years.

It is even worse with bonds, as cap weighting means benchmarks are biased towards countries with the biggest debts. As Lombard Odier, the private bank, points out: “Investors following the traditional market-cap bond benchmark approach saw their allocation to Greek government bonds grow from being negligible in 1999 to over 5 per cent at the beginning of 2010.”

So why is cap weighting still so widely used? It has to do with modern portfolio theory, the efficient-market hypothesis and the University of Chicago.

In 1952, Professor Harry Markowitz’s theory posited that the “market portfolio”, weighted by market cap, must yield the highest utility, on the assumption that all assets are at all times correctly priced by the market – which, conveniently, is what Professor Eugene Fama then hypothesised in 1970. However, this also assumes that investors do not make mistakes.

Rob Arnott, founder and chairman of Research Affiliates and pioneer of non-cap-weighted indices, points out that theory is one thing, but reality another. If the market is infallible, he argues, the 10 companies with the largest cap weighting should become the top 10 companies in terms of “economic scale”. In the US, though, cap-weighted indices have suggested 34 different companies would become the 10 economically largest. This, he says, is the flaw in cap weighting: it creates too many “fallen angels” – companies highly valued by the market that fall from grace.

With the UK’s angels losing altitude – the FTSE is down another 25 points – it is time to consider the alternatives.

Equally weighting the shares in any benchmark would instantly remove distortions but require such frequent rebalancing that it would add transaction costs. Gross domestic product weighting a global portfolio may seem preferable, but overlooks the lack of correlation between domestic economies and domestic stock markets.

Fundamental weighting – whereby shares are weighted according to their dividends, book value and cash flows – has more supporters, including Arnott. He argues it would have avoided fallen angels generally, trimmed media and technology stocks in 2000-02, reduced value-stock holdings mid-decade, and predicted changes to cap-weighted indices two-thirds of the time. Crucially, it produces a tougher benchmark. Wegelin calculates that fundamental weighting has historically outperformed cap weighting by 2 per cent a year.

But for some managers, it relies too much on company accounts. Yves Bonzon, chief investment officer at Pictet, says: “The construction of an index on the basis of a set of quantitative factors is ... a venture into the waters of active management.” He argues that, because managers are paid to pick stocks from public markets, cap-weighted indices measuring these markets in aggregate “will remain the foundation of investing”.

For how much longer, though, if that foundation has just fallen 75 points in 75 minutes?

“Some portfolio managers hold stocks that are big in indices to ‘control risk’, but the industry is moving away from it,” says Cesar Perez, managing director at JPMorgan Private Bank.

Thorne of Capital Generation Partners suggests clients should accelerate this move. “They should look for active managers who are agnostic on the benchmark,” she says. “The benchmark is not difficult to beat.”