Dividend Weighted Index Funds

There's an interesting article in today's Wall Street Journal claiming that, over the past 40 years, a dividend-weighted index fund would have outperformed a market-capitalization-weighted index by over a full percentage point per year with lower volatility.

To anyone who isn't a finance geek, this sounds pretty esoteric. In the context of finance theory, however, this is huge: akin to proving that continents move or (in the analogy the author uses) that the Earth goes around the Sun instead of the other way around.

According to the Capital Asset Pricing Model (CAPM), the official Received Wisdom of finance theory from about the 1960s until the mid-90's, it is impossible to systematically outperform a market-weighted index fund over the long term and with lower volatility.

In recent years, though, CAPM has come under attack on various fronts--mostly from people who have found systematic biases in the way people think about buying and selling stocks (also known as Behavioral Finance). I've also seen some intriguing research showing that one of the mathematical assumptions of CAPM turns out to be more critical than most people believed, and that the mathematical proof of CAPM breaks down completely if the assumption is even mildly violated.

(For math-finance geeks intrigued by the previous statement, here's a quick summary: CAPM assumes that all market participants can have unlimited leverage both positive and negative. In other words, there are no limits to either margin buying or short selling for anyone. In real life this isn't true, of course, since some people can't buy on margin or short-sell at all, and everyone faces at least some limits. Theorists have generally assumed that this real-life restriction didn't change the results of the theory. It turns out, though, that violating this assumption makes the Efficient Frontier nonlinear for any market participant who has a margin restriction, and if any market participants have a nonlinear Efficient Frontier, then the market portfolio will no longer lie on the Efficient Frontier. Sorry for no link, but I read this in a dead-tree journal some months ago.)

The defense of CAPM against all these various attacks has been that, even though stocks may be mispriced from time to time, there is no systematic way to profit from those mispricings without taking on more risk than you'd assume by buying an index fund. It's a sort of Heisenberg Uncertainty Principle of the stock market: stock prices might be wrong, but there's know way to know if they're too high or too low.

But in light of the holes in CAPM, I've been wondering lately if there might be some better way to put together a stock index than by weighting by market capitalization. Market-cap-weighted indexes have been traditional, but there's nothing special about them other than the fact that they are supposed to be the most efficient (i.e. highest return at the lowest risk) under CAPM. If CAPM is wrong, though, then there could easily be other ways to construct indices which do better.

So, it seems, is the case. The claim in the WSJ article is that some other methods of weighting stock indexes do, in fact, outperform a market-cap index.

There are some caveats to this. First is that the author of the article works for a company that sells stock funds which weight by dividends, so there's a conflict of interest.

The second is that there may be something special about the period of time from the mid-60's to today. Four decades is a reasonably long time to test a theory like this one, but the markets have been known to have multi-decade trends in the past. Just because this strategy would have worked in the past doesn't mean it will continue to work in the future.

Finally, dividend weighting is effectively a form of value investing. Studies in the past have shown that value investing can outperform growth investing over time, and it tends to be less risky, but there will be years when value investing underperforms and even looks stodgy. So don't expect to impress your buddies at the country club by talking about your dividend yields.