Unmasking pseudo-index funds - the S&P500 is not an index

I hadn't realized this until I started following the S&P 500's decisions recently, but it turns out that the S&P 500 is not truly an index, and funds mirroring it are not index funds, as explained in this article. This is particularly sad because there are about 750 billion dollars worldwide which mirror the arbitrary set of stocks, chosen by committee, in the S&P 500.

Basically, an index fund is a passive mirror of a particular asset class, like "All public US-based companies with readily available price information" (the Wilshire 5000, a total index), or "The 100 largest non-financial companies listed on NASDAQ" (the Nasdaq 100, QQQQ). The key to an index fund is that there is no human trying to outguess the market - you leave that to the hedge funds, and just bet on everything, with minimal transaction costs. If you do make guesses, you make broad guesses about asset classes like "Value will outperform growth", or "Small will outperform large". You never make guesses about individual stocks.

Yet that is exactly what the S&P 500 does. A committee of 8 people meets periodically and makes decisions about what stocks to add and drop, based on some broad rules about price and market cap, plus their own judgement of what rules matter. The composition of the "index" is not set by a formula but by human decisions - in other words, an actively managed mutual fund (although one with a light trading hand). Contrast this with QQQQ:

Unlike Standard & Poor’s, Nasdaq rebalances its leading index according to strict statistical criteria once a year, on the third Friday of December. Essentially, all 5,000 or so Nasdaq stocks are ranked from high to low by market capitalization, and the largest 100 that have been public for at least two years and trade at least 100,000 shares daily make the index. If an index stock falls to rank 101st to 150th in market capitalization at the time of the rebalancing, it is given a one-year grace period before facing expulsion. If it falls below the 150th rank in any year or fails to regain the 100th spot or better in its grace-period year, the stock is expelled and replaced by the next-highest ranked Nasdaq member. The only major exceptions are companies that are delisted from Nasdaq, as Adelphia Communications (ADELA, news, msgs) was a few weeks ago

While this particular policy has some issues - for example, the one-year grace period arguably lets total busts hang around too long - it is a passive index, and so funds mirroring it are index funds, unlike the S&P 500. Now, its not like the S&P 500 has a lot of churn. Mostly, it does buy and hold. Mostly, it reflects the largest companies. Mostly - but not completely, because it uses a subjective weighting and set of criteria to choose its set of stocks.

One of the most popular indices, reflected in some of the largest index funds, including some of the earliest index funds ever created - and its actively managed. John Bogle should be spinning in his grave...err...boardroom.

Share this

Thanks, Patri... this is one

Thanks, Patri... this is one of the most useful posts I've read this year.

There's a world of

There's a world of difference between an actively managed fund and an index fund, and the fact that the S&P is chosen in part using subjective human judgement rather than purely by statistical rules doesn't diminish that difference.

The biggest difference is that S&P index funds have low overheads compared to actively managed funds. The index may be managed, but the index funds aren't, so index funds save investors money on management fees.

If you believe in random walk or even weak-form EMH, then that's the only difference between an S&P index fund and an actively managed fund with a wide range of large-cap holdings. Each is as likely to have stocks that will go up or down... but the index fund will have a lower overhead because they have fewer overpriced analysts to pay.

Nice post. This is why you

Nice post. This is why you should ALWAYS read the prospectus.

I don't recommend index

I don't recommend index funds, they only worked when no one else was investing in them.

While I agree that the S&P

While I agree that the S&P 500 may not be a true "index" I wasn't a big fan of that article from msn.com. The writer seemed to indicate that the S&P 500's poor performance was evidence of its failings. However, I don't think that's a useful criterion by which to judge an index. He even implies that the S&P 500 compares unfavorably with the Dow Jones Industrial Average, though to my mind the DJIA (which uses stock prices to weight its constituents), is truly a strange beast.

Incidentally, DFA describes its funds as "passive" rather than as index funds. In the end it's just a name. Whether it's a committee which arbitrarily chooses the stocks, or a committee that chooses some algorithm to pick the stocks, they all seem like different points on a spectrum to me.

Gus Sauter (who is the uber-manager of Vanguard's index funds) wrote an interesting article on index construction here:

While you can decide whether to agree with him or not, he does show that there are a lot of ways to construct an "index".

The DJIA made sense before

The DJIA made sense before the days of computers and the value of the index had to be computed with an abacus or something like that.

But why do people still cling to a relic of a pre-computer past?

Well, here is the question:

Well, here is the question: how much is it not an index? In other words, if you applied the NASDAQ criteria to the S&P, what fraction of the companies would be different. If it's 99% the same, it simply isn't going to have an impact. Worth checking into, though.


Walt - its a small

Walt - its a small difference.

Carnival of the Capitalists

Carnival of the Capitalists 12/19/2005
Welcome to the Carnival of the Capitalists and my second time hosting the COTC. Note that several people tried to submit multiple posts - when that happened, I picked just one to include this week. Many thanks to Silflay Hraka