The hallmark stock indices that are supposed to provide a quick gauge of market activity, and upon which myriad products are built, are increasingly looking broken.
Eddy Elfenbein at CrossingWallStreet notes the shocking fact that one Dow company, Caterpillar, has accounted for 40% of the index’s gain so far this year. Because it’s a price-weighted index, and because Caterpillar trades at $79/share — more than three times that of fellow Dow-component Microsoft — it has a ridiculously large influence on the index.
Not that the other indices are immune.
The NASDAQ 100, upon which the popular QQQQ ETF is based, is more and more based on Apple alone, an issue that Mike O’Rourke of BTIG highlights in his latest note.
Once or twice a year, we feel the need to make note of the poor design of the Nasdaq 100 Index. It usually comes up when Apple makes a run to new highs as it has recently done. The latest rally further extended Apple’s lead as the index’s number one component, now accounting for 20% of the index. This is a world where ETFs are coming under renewed scrutiny for numerous reasons. One example is professional investors recognise that levered ETFs can be very dangerous if used beyond a very, very short time horizon. Some brokerage houses preclude their retail clients from transacting in the products. Nobody would suspect to look at one of the oldest, most widely held and most active ETFs as potentially having a problem. The fact is that due to the construction of the Nasdaq 100, the QQQQ arguably fails to meet the goals laid out in its registration documents filed with the SEC.
If you look at the ETFs’ original offering documents, they explain the Nasdaq 100’s conversion into a “modified capitalisation-weighted index” in December 1998. This is intended to be a cross between a capitalisation weighted index and an equal weighted index. In the world of index construction, these are opposites that should be mutually exclusive. The resultant mess of attempting to mix oil and water is the current disproportionate weighting structure of the Nasdaq 100 where 1 name constitutes 20% of the index and the other 99 names comprise the 80% balance.
The problem back in December 1998 was that without using the modified weighting structure, the top 5 companies (Microsoft 22.5%, Intel 12.8%, Cisco Systems 9.5%, MCI Worldcom 8.5% and Dell 6.1%) would have comprised 59.4% of the weighting in the index. Today, Worldcom is defunct and under the old structure, the remaining 4 surviving companies would constitute 18.8% of the index. The current top 5 market capitalisation companies (Apple 10.7%, Microsoft 8.6%, Google 6.7%, Oracle 5.4% and Vodafone 5.4%) would make up 38.6% of the index if it was capitalisation weighted. The actual top 5 (Apple 20.1%, Qualcomm 4.7%, Google 4.3%, Microsoft 4%, and Oracle 3.1%) comprise 36.1% of the index weight today. The modified capitalisation structure winds up making the rich richer and the poor poorer as it creates a self reinforcing momentum from indexers. As Apple has outperformed over the past decade, its weight grew notably, and as the 5 leaders of 1998 underperformed their weighting shrank in similar fashion. The modified capitalisation structure has benefitted Nasdaq 100 indexers as Apple’s strong performance has pulled the index within 11% of its 2007 peak as opposed to the 37% the S&P 500 needs to rally to recover the 2007 highs. However, if Apple falters, then watch out.
The SEC registration statement says the modified capitalisation-weighted methodology was expected to “(1) retain in general the economic attributes of capitalisation weighting; (2) promote portfolio weight diversification (thereby limiting domination of the Index by a few large stocks); (3) reduce Index performance distortion by preserving the capitalisation ranking of companies; and (4) reduce market impact on the smallest Index Securities from necessary weight rebalancings.” The facts speak for themselves – the modified capitalisation weighting structure fails on the first 3 objectives. It probably fails on the fourth, because the majority of smaller companies are biased to have slight overweight to where they would be in a capitalisation weight structure.
What is clear and noteworthy is that over time the market cap top 5 leadership has changed as businesses have risen and fallen. The market cap top 5 is nearly equivalent to the actual current weighting, but more diversified. Therefore, if the Nasdaq 100 were market capitalisation weighted, it would do a much better job of meeting the 4 goals of the modified cap structure, than the modified structure does itself. This is a credit to the Nasdaq as an exchange itself, representing the progress it has made as it matured over the past decade. It is time its flagship index takes the training wheels off and begins to represent reality.