Equity investing

A HEADLINE on Bloomberg claims (in typical Bloomberg-ese) there was "No Lost Decade for S&P 500 as Big-Cap Bias Masks Rally". The idea is that, if you equally weight stocks, then they have risen 66% over the last decade.

That may be true and I have argued in the past that there's a lot of scope for alternative ways of weighting portfolios than market values. The Robert Arnott approach, which weights stocks by "fundamentals" like sales and dividends, avoids the peril of market-value weighting, which leads investors to allocate most money to those stocks that are most fashionable, and thus likely to be too expensive.

The problem with this approach is that, by definition, not all investors can follow it. If we all equally weighted stocks then, well, all stocks would be equally-weighted. As it is, many stocks are quite small and illiquid so are difficult for big managers to own. Take the Fidelity Magellan fund which had $100 billion at its peak; divided equally among 500 stocks, that would be $200 million per holding. For the smaller stocks, the fund would have 20% of the equity; as it got in and out, it would move the price substantially.    

The reason we use market cap-weighted stocks to measure the long-term performance of equities is that they represent the performance of the average investor, before costs. Ten years ago, investors were making big bets on tech stocks; they were wrong because they were overpaying. Indeed, the average mutual fund investor has done even worse than the index suggests; because investors chase hot mutual funds which, in turn, tend to own hot stocks.

So yes, equally-weighted funds are a neat idea (although since they've done so well in the recent past, they may underperform in the future). But they aren't what 99% of people own. For most equity investors, it was indeed a lost decade.