I uncovered and interesting article, “The S&P 500 is a mutual fund – and a bad one,” written in 2002. As someone who has considered buying shares of SPY to make up the major part of his US portfolio, this article, claiming that the S&P 500 is a bad mutual fund was a must read. In it, Jon D. Markman says:
“Unlike most index publishers, such as the Nasdaq and Dow Jones, Standard & Poor’s adds and subtracts stocks from its three broad indexes — the largecap 500, the Midcap 400 ($MID.X) and the Smallcap 600 ($SML.X) frequently in accordance with a largely subjective list of criteria that includes market capitalization, liquidity and their representation of industrial sectors.” [emphasis mine]
The S&P 500 is supposed to be representative of “the market” (not the US economy which consists of 306 privately-owned US companies with revenues of at least $1 billion) and one of the ways they do this, is by matching the sector allocation of the entire US stock market (small-caps and all, almost 10,000 stocks as of 2005) to the index. In 2000, according to Mr. Markman, the S&P had a 14% weighting in technology stocks by market capitalization whereas the entire U.S. market had a weighting of 18% in technology stocks by market cap. The people at S&P proceeded to add the tech stocks with the next-largest market caps to its index, removing stocks in industries that had now become “over-weighted” compared to the entire US market sector allocations. Many of these technology-related stocks were at their peaks in 2000, and extremely over-valued. The S&P then held on to these stocks as they plummeted and lost sometimes greater than 80% of their peak values.
In their quest for ultimate sector diversification, they have performed one of the most banal rebalancing operations I have ever seen: buying tech stocks based on high momentum and high valuation. Standard portfolio rebalancing operations would normally involve selling stocks which have recently advanced to such a large degree so as to have overweight positions, compared to their original weightings. The tech stocks which were already in the S&P500 index had advanced significantly and the S&P500 was overweight in these positions compared to a few years ago, and probably should have sold off some shares in the technology sector; however, the broader market (including the large number of high tech companies barely post-IPO and with no earnings) just happened to have advanced even more (to make up 18% of the market vs. 14% for the S&P500), thereby triggering a technology stock buying spree for the S&P500 index managers.
This would be enough to put any holder of SPY in an uproar. Especially if SPY was a mutual fund with a mutual fund manager. Only the SPY (or S&P500 Composite) is in fact NOT a mutual fund, as the title of the above article claims, it is just an index. It was designed to be some representative figure of the value (as currently being traded) of the US stock market. This article, although it is misguided in laying blame on the people at S&P, highlights some key downsides of indexes, especially the ones based on market value, and sector allocation. There are also many other silly things which go into factoring how much of each stock is held, such as liquidity and available float. You get what you pay for I guess. ETFs which follow the major indexes are popular because of their low fees and the ability to beat the returns of a significant number of equity mutual funds. But the indexes are not mutual funds, and they are not supposed to be smart. They follow a passive strategy and stick to it.
There are other indexes out there which do not follow this type of methodology. Two of which I know of are the S&P Equal Weight Index (replicated by the Rydex S&P Equal Weight ETF), and also the Dow Jones Canada TopCap Value Index (replicated by TD Select Canadian Value Index Fund ETF). From Rydex’s website,
Equal weighting also offers increased diversification as compared to its cap-weighted counterpart. The composition of the securities of the S&P 500 Index combined with quarterly rebalancing avoids over concentration in popular (or momentum) sectors, such as the domination of the Information Technology sector that occurred in the S&P 500 in 1999 – 2000.
The S&P Equal Weight Index is just another alternative, and a good one if you want to avoid what happened to the S&500 in 1999-2000. I would not be surprised if the S&P Equal Weight Index was born out what happened in the late 1990s bull market, in the same way that the S&P TSX 60 capped index was created after Nortel became a huge percentage of the TSX 60 index during that same period.
This article was first published on October 31, 2005