As the ETF industry has continued to gain traction in recent years, indexes have been transformed from hypothetical performance benchmarks into essentially investable assets. As such, investors have become increasingly familiar with many of the indexes that are linked to the most popular exchange-traded products and the methodologies used in the construction and ongoing maintenance.
In most cases, the strategies used to determine index components and assign weightings is straightforward and easy to understand. Many of the most popular equity ETFs are linked to cap-weighted indexes, meaning that the companies with the largest market capitalization–in some cases float-adjusted–are given the largest weighting. Equal weighted indexes have become increasingly popular in recent years; these products also feature a relatively simple methodology for determining components and weightings. Other index weighting and construction methodologies, including those that use various fundamental factors such as revenue, earnings, and dividends, are relatively easy to grasp [see the Guide To ETF Index Weightings].
But the composition of some ETFs may be a bit baffling. The line of HOLDRS products from Merrill Lynch (which are perhaps best thought of as distant cousins of true 1940 Act ETFs) feature some bizarre rebalancing rules that can lead to huge concentrations among a handful of individual stocks. The iShares Barclays Aggregate Bond Fund (AGG) holds 700 or so individual securities–less than 10% of the bonds that comprise the underlying index. And then there’s the PowerShares QQQ (QQQQ), one of the most popular U.S.-listed equity ETFs.
Under The Hood
QQQQ–better known as the “cubes”–has become a popular option for investors looking to gain exposure to growth companies and the tech sector in general. The fund finished 2010 with more than $22 billion in assets, making it the seventh-largest U.S.-listed ETF by AUM. QQQQ is based on the Nasdaq-100 Index, a benchmark that includes 100 of the largest domestic and international nonfinancial companies listed on the Nasdaq Stock Market by market capitalization. That methodology seems simple enough, but a look at the composition of the ETF may have some investors scratching their heads [read Weighting Methodologies: An ETF Report Card].
|QQQQ Weight||Market Cap ($B)|
Tech giant Apple makes up nearly 21% of QQQQ assets, a considerable weighting but not totally uncommon among certain ETFs. Exxon Mobil (XOM), for example, accounts for about 22% of the Vanguard Energy ETF (VDE) and 24% of the iShares Dow Jones U.S. Energy Sector Fund (IYE).
What is strange about QQQQ is that the weightings assigned to individual securities don’t seem to be consistent with the respective market capitalizations–something one would expect from a cap-weighted index (the Nasdaq 100 Index is calculated under a “modified market capitalization-weighted methodology”). After Apple, the next four largest components have an aggregate market cap of about $680 billion, more than twice Apple. Yet these four companies make up just 16% of QQQQ, considerably less than Apple’s individual weighting [see all the ETFs that offer exposure to AAPL].
The explanation for the quirky weights in QQQQ goes back to 1998, when a Nasdaq subsidiary was planning to introduce an ETF linked to the Nasdaq 100 Index (Nasdaq transferred sponsorship of QQQQ to PowerShares in 2007). In order to be eligible for a certain tax status, mutual funds and ETFs can’t have more than 25% of holdings in a single security. When Nasdaq was planning an ETF linked to the Nasdaq 100 Index in the late 1990s (it was originally called the Nasdaq-100 Tracking Stock), Microsoft made up more than 25% of the underlying index–presenting a hurdle to launching an ETF.
In order to make the underlying index ETF-compliant, Nasdaq undertook an interesting modification process:
- Divided the 100 components into large companies (making up 1% or more of the index) and small companies
- Cut Microsoft’s weight to 20%, and cut the weightings of big cap companies by the same proportion
- Increased the weighting of the largest small companies to 1%
- Increased the remainder of the small companies by the same proportion
At the time, Apple was nothing like the tech giant it is today; at the end of 1998, the stock made up just 0.9% of the index. Unadjusted, the company would have accounted for just 0.4% of the Nasdaq-100 Index. Microsoft made up 14.5% of the index at the end of 1998; MSFT would have accounted for 22.5% prior to the weighting adjustment.
Over the next 12 years, Apple stock skyrocketed; AAPL gained more than 3,000% between the end of 1998 and the end of 2010. In any cap-weighted index, an appreciation of stock price leads to an increased weight in the index. And thanks to the 1998 adjustment that approximately doubled the weighting, the impact of the meteoric rise of Apple stock has been exacerbated over the years. In early 2011, AAPL makes up close to 21% of the QQQQ, thanks in no small part to the rebalance that took place more than a decade ago [see Apple & Verizon Team Up On iPhone: How Will ETFs Respond?].
Microsoft hasn’t matched the success of Apple over the last decade or so; for the 12 years ended December 2010, MSFT stock was up less than 1%. And just as the one-time adjustment has amplified the increase in Apple’s weighting, it has also amplified the contraction of Microsoft in the index.
That’s the back story behind QQQQ’s seemingly bizarre weightings; a plan to reduce the weight of one tech giant in 1998 has contributed to the huge weighting of another company some 12 years later. The quirky modified market cap weighting methodology doesn’t necessarily detract from QQQQ as an investment, though of course it is important to consider the impact that changes in AAPL shares will have on the ultra-popular ETF [also see For ETF Investors, Little Decisions Have A Big Impact].
[For more ETF insights, sign up for our free ETF newsletter]
Disclosure: No positions at time of writing, photo is courtesy of Yutaka Tsutano.