Six Reasons To Consider Dumping SPY

by on September 27, 2010 | ETFs Mentioned:

The meteoric rise of the ETF industry–and to a smaller extent the introduction of index mutual funds some 20 years earlier–has had the effect of transforming indexes from hypothetical measures of performance into investable assets. As indexing strategies continue to gain more widespread acceptance with all levels of investors, the scrutiny of construction and maintenance methodologies underlying ETF-linked indexes has intensified considerably. And where potential pitfalls have been uncovered and explained, ETF issuers have been quick to provide investors with a number of alternatives.

Investors are creatures of habit, so it isn’t surprising that the majority of equity ETFs are linked to indexes that utilize long-standing, familiar construction strategies. The indexes underlying many of the most popular ETFs are market cap-weighted benchmarks, meaning that the companies with the largest equity value generally receive the largest weightings. But while the ETF industry has helped to reinforce the popularity of these indexes–such as the S&P 500, Russell 1000, and MSCI Emerging Markets–it has also given increased visibility to alternative weighting methodologies [see our database of ETF indexes]. A number of firms, including Rydex and State Street, offer ETFs linked to equal-weighted indexes. WisdomTree has carved out a niche in the earnings-weighted and dividend-weighted arena, while RevenueShares offers a suite of ETF products that determine individual security weightings based on top-line revenue. And in recent months the RAFI methodology utilized by several PowerShares funds has gained more widespread acceptance.

When allocating assets to equity ETFs, most advisors and investors focus primarily on the type of exposure desired: large cap domestics, small cap internationals, etc. Few give much, if any, consideration to the ideal weighting methodology to accompany the desired exposure. But the rules used to both select index components and allocate individual security weightings can actually account for a significant portion of the total return generated by an ETF.

YTD Performance
ETF Weighting Return
SPY Market Cap 1.98%
RWL Revenue 2.64%
EPS Earnings 3.40%
DLN Dividend 4.30%
PRF RAFI 4.51%
RSP Equal 5.24%
EQL Equal Sector 2.03%
*As of 9/23/2010

The adjacent table shows the year-to-date for the S&P 500 SPDR (SPY) and six other large cap equity ETFs that offer generally exposure, but put a unique twist on the weighting methodology afforded to component stocks. So far in 2010, SPY has lagged behind each of these alternatives, in many cases by a fairly wide margin. This table hopefully helps to highlight an important fact for ETF investors: the weighting methodology employed can have a major impact on bottom line returns, even when the components stocks are nearly identical [also see Three Low Volatility ETF Options].

The holdings of RSP and SPY are identical, but the lesser-known Rydex ETF gives each an equal weighting. That seemingly minor tweak is the primary reason for the more than 325 basis point gap in year-to-date returns. The same goes for RWL–its holdings are identical to SPY but weighted according to top line revenue–which has created a gap of nearly 70 bps. The overlap between SPY and the remaining funds isn’t perfect, but generally comes pretty close [see Beyond SPY: Nine Alternatives To S&P 500 ETFs].

  • RevenueShares Large Cap ETF (RWL): This ETF is comprised of the same stocks in the S&P 500, but individual security weightings are determined based on top-line revenue. Revenue-weighting essentially overweights stocks with low price-to-revenue multiples and underweights those with high P/R metrics. That strategy has obvious appeal in certain environments, and delivered pretty solid returns in 2010.
  • WisdomTree Earnings 500 Fund (EPS): This ETF seeks to replicate the WisdomTree Earnings 500 Index, a benchmark that measures the performance of earnings-generating companies within the large-cap sector of the U.S. equity market. Unlike RWL, this ETF doesn’t hold the exact same stocks as the S&P 500, but the overlap is significant. To understand the appeal of this strategy, consider two hypothetical companies with identical earnings but different market capitalizations. Earnings weighting would give the companies an equal allocation, while cap-weighting would have a bias towards the larger market cap (i.e., the company with a higher P/E ratio).
  • WisdomTree LargeCap Dividend Fund (DLN): This ETF is similar to the aforementioned EPS, but instead of utilizing earnings to determine components and weightings DLN relies on cash dividends. As such this fund has obvious appeal to investors looking to follow a value strategy, and may have the added bonus of steering clear of companies that have been “cooking the books” (since it’s difficult to fudge cash dividends paid).
  • PowerShares FTSE RAFI U.S. 1000 Portfolio (PRF): Much like the S&P 500, the index underlying this ETF is designed to track the performance of the largest U.S. companies. The difference is in the definition of “largest” utilized to determine components and individual weightings; whereas the S&P relies on market cap, the RAFI methodology developed by Rob Arnott uses a “fundamental score” to compute size. Stocks are selected and weighted based on four fundamental measures of size, including book value, cash flow, sales, and dividends–thereby incorporating some of the elements of revenue and dividend weighting [see Does Your Portfolio Need A RAFI ETF?].
  • Rydex Equal Weighted S&P ETF (RSP): This ETF from Rydex is relatively simple to understand; it holds each of the components of the S&P 500 ETF, but gives an equal weighting to each (i.e., each stocks makes up 0.20% of total assets upon rebalancing). That means that RSP maintains a significantly lower concentration of assets; the top 10 holdings of SPY make up about 19% of assets, compared ot just 2% for RSP.
  • ALPS Equal Weight Sector ETF (EQL): This ETF takes a different approach to equal weighting, giving equivalent allocations to each of the nine primary sectors of the U.S. economy. That means that relative to SPY, EQL will maintain larger weights in materials and utilities, with smaller positions in tech and financials. There are some potential advantages to an equal-weighted sector approach: the potential adverse impact of a crash in any one industry is lessened somewhat, while the opportunity to participate in a rally in any sector is improved [see Why EQL May Be A Better S&P 500 ETF Than SPY].

Of course, the period of time highlighted in the above table is relatively short–less than nine months. There have been stretches during which cap-weighted funds outperformed these alternatives, and there will no doubt be similar stretches in the future. Each methodology has both advantages and potential drawbacks. But the figures above should demonstrate very clearly that the weighting methodology selected has a material impact on bottom line return [see Ten ETFs Every Advisor Should Know (But Most Have Never Heard Of)].

Disclosure: No positions at time of writing.