The Problem With Indexing
Indexing your investments is a good way to increase your chances of doing better than average. Most active investors trying to pick winning stocks (and avoid losing ones) don’t beat the index. Those who do are hard to identify ahead of time. Once you think you’ve identified a market-beating manager, they often allow their fund to get too big, which typically reduces their chances of repeating their past success. And sometimes they’re on the cusp of retirement. Or they’ve lost analysts who’ve helped them be great. In other instances, you might be correct in thinking they can outperform over the long term, but you’re about to capture one of their fallow periods. Nobody ever said this game was easy.
Yet, there are problems with indexing too. “Indexing” generally means following what’s called a “capitalization weighted” index. That means the rank of the stock in the index is determined by how much the stock market values it. Market capitalization is the total shares of a stock outstanding multiplied by the price per share. But is taking the stock market’s verdict about where a stock should be ranked in an index reasonable? Many people think it isn’t. After all, you necessarily own more shares of the most favored and possibly expensive stocks in a capitalization weighted index.
When the first index fund – the Vanguard 500 Index Fund – came to the market in the early 1970s, academic finance was dominated by the notion that the market prices stocks properly – or accurately, based on available information – at all times. That made it seem reasonable to rank stocks by how the stock market says they should be ranked.
But there’s a case to be made that ranking stocks based on their underlying companies’ economic fundamentals such as sales, earnings, dividends, and book value might also be reasonable. And if stock markets don’t always set prices properly, if they are sometimes the victims of emotion gripping their participants, maybe focusing on economic fundamentals of the businesses is a smarter approach to indexing. And, after all, if you rank stocks based on companies’ underlying economic fundamentals, you’re arguably not showing any disrespect for markets. You might just be elevating the markets for companies’ goods and services over the market that trades those companies’ stocks.
An easy way to “break the link” (as Rob Arnott, founder of fundamental indexing, puts it in describing all “smart beta” strategies) between market capitalization and index rank of a sock is to start with a capitalization weighted index, but then rearrange it in a way that equally weights the stocks. No attention to underlying economic fundamentals needed.
This is the subject of a new paper published by Standard and Poor’s. First, the paper documents the outperformance of equal-weight indexes compared to their capitalization weighted equivalents. According to the paper the S&P 500 Equal Weighted Index has outperformed the S&P 500 Index by 2.1 percentage points. Moreover, this outperformance or “alpha” exists in international markets as well.
How much money is 2 percentage points annualized? It can amount to $50,000 in a scenario that’s not difficult to imagine for an ordinary investor. For example, if you invest $5,000 for 25 years and earn 5% annually on the investment, you’ll wind up with just under $340,000. If you invest the same amount for the same time frame, but earn 7% annualized instead of 5% annualized, you’ll wind up with just under $390,000.
The most obvious reason for the outperformance of the equal-weight index is its greater exposure to smaller stocks. The plain S&P 500 Index is concentrated at the top. Only 10% of the names – 50 stocks – account for nearly one-half of the index’s total weight, and the largest 30% of stocks – the top 150 – account for 75% of the total weight. Conversely the smallest 40% — the bottom 200 stocks – account for only 10% of the index’s total weight.
An equal-weight index will elevate the exposure to the smaller stocks in the index, and this “size bias” explains a “considerable portion of the S&P 500 Equal Weight Index’s long-term returns,” according to the paper. Another smaller part of the explanation is related to the fact that the equal weight index undergoes periodic rebalancing. This means that stocks that have done well are periodically sold and stocks that have done poorly are purchased to keep the fund’s allocation equally weighted to the 500 stocks. So equal weighted indexes have a kind of “anti-momentum” strategy. And while momentum is viewed as a legitimate factor that can beat a plain index, so does an anti-momentum strategy. In fact, performance of the equal weighted index relative to its capitalization weighted counterpart can indicate how momentum is faring overall in markets in a given period, according to the paper. When momentum is doing well, the equal weighted strategy tends not to do well, and vice versa.
Another feature of an equal weight index is how it treats the sector components of the index. During the technology bubble of the late 1990s, technology stocks began to overwhelm the index in the sense of becoming a larger and larger component of it. And although an “equal-weight index effectively allocates to each sector in proportion to the number of stocks held in that sector,” according to the paper, this is enough to mute the bad or overwhelming effects that a sector run-up can have.
Overall, the evidence suggests that an equal-weighted index captures a greater share of the stocks that achieve above-average returns. This doesn’t mean such a fund should be the only stock holding in a portfolio. For example, it can complement trend-following or relative strength strategies, helping to increase a portfolio’s volatility-adjusted returns when used that way. An equal-weight strategy can also mute the tendency of a low volatility strategy to go through periods of underperformance.
Although the paper doesn’t characterize a capitalization weighted index this way, its analysis begs the question of whether such an index is itself a kind of a momentum strategy. In any case, investors should be mindful of the problems inherent in plain capitalization weighted indexes especially at a time when indexing is garnering so many investor dollars. Capitalization weighted indexer will do fine over time, if investors can stick with the approach through thick and thin. But, at some point, they might wonder why they didn’t consider alternatives.