Tuesday, 17 February 2015

The inefficiency of market capitalisation weighted indices

The (sensible) advice to people saving for their pension has long been to avoid active fund managers with their high fees and put their money into cheaper exchange traded funds (ETFs) which track major indices.

This is sensible because active fund managers often try to track the index but generally do not add enough value to justify their fees. However a great body of research  over the past 25 years has shown that this is still not optimal. For a number of reasons, probably to do with systematically over-weighting the most popular stocks, and possibly also to do with the sheer volume of trading based on the composition of indices, this has been shown to be inefficient.

My white paper on the subject comes out of research for Dacharan's market neutral equity fund. It shows again how a market capitalisation weighted basket of stocks is, in the long term, considerably worse than any other weighting tested. I suggest that a constrained inverse beta weighting is a good choice of weighting as it is reasonably balanced but also weights more the lower volatility and less correlated stocks.

In fact, the market capitalisation weighted indices are so bad that for a large part of our portfolio, we don't even look for bad stocks. We just sell futures in the index, confident that they will under-perform our portfolio on average. The other advantage of this is much lower costs.

So how can you improve your pension portfolio to avoid this systematic loss, which I would calculate at approximately 1.5% per year. I would look to invest in low cost ETFs which track alternative weightings. For example 'minimum volatility' weighted indices or dividend weighted indices.

Whilst it could be argued that the popularity of these in recent years counts against them, the market is still tiny compared to the market capitalisation weighted indices and I think that in the long term, on a risk-return basis these will outperform..



  1. Ari,

    There is no need to do all this work.

    1. Neil Woodford has a target of beating the FTSE all-share tracker by 10% pa with his CF Woodford UK Equity Income fund.

    2. Simon Ward, chief economist at Henderson, has a regular post on Mindfulmoney.com telling readers the reliability of the "monetarist rule". When narrow money, M!, increases output increases six months later. It is a golden rule and never fails.

    1. David, thank you for this.

      1. With due respect to Neil Woodford (who I'll take your word is an excellent manager), if he beats the index by around 10% a year for the next 5 years I will post a video on YouTube of myself eating an article of clothing of your choice. It is just an unreasonably high level of consistent overachievement.

      2. Personally I don't believe in rules that do not have many data points to support them. But it does potentially have some sense to it and if it is working for him and you then good luck and I wish you every success.


Sorry, I have had to moderate comments because of an annoying pest spammer who keeps posting American football matches links.