The Suitability of the S&P 500 Index as a Benchmark
[The following is excerpted from the 1998 S&P 500 Directory.]
The use of the S&P 500 as the proxy for the overall stock market predates the widespread adoption
of the Capital Asset Pricing Model (CAPM) in the 1970s. As the amount of money invested in the
equity markets grew in the 1950s and 1960s, the need for a capitalization-weighted, broad-based
market indicator that reflected how people actually invest in equities became self-evident. By
convention, the S&P 500 Index, already well-known to academics and to professional money managers,
was used as the market portfolio in tests of the CAPM. Betas of individual stocks were then
calculated against the S&P 500 Index, which by definition had a portfolio beta of 1.00.
These days, it is difficult to find an equity manager who cannot tell you how its portfolio's
performance compares with the S&P 500. Some companies, such as Fidelity, even make a portion of
their management fees contingent on whether their funds outperform the S&P 500. In the Magellan
Fund's case, its fee is raised or lowered by 0.20% of assets, depending upon how it fares against
the S&P 500's total return over a rolling three-year period.
However, it is no longer possible for an investment management firm simply to claim that it beat
the S&P 500. The adoption of performance presentation standards by the Association for Investment
Management and Research (AIMR) and their inclusion as of January 1, 1993, as Standard III F of the
AIMR Standards of Professional Conduct, has focused attention on the need for properly defined and
utilized investment benchmarks. The firm must use a benchmark that parallels the risk or investment
style the client's portfolio is expected to track.
A valid benchmark should be a passive representation of the manager's investment process, and the
manager's portfolio should fall inside the manager's benchmark universe, according to Jeffery V.
Bailey, CFA, of Richards & Tierney, Inc. Speaking at the November 1994 AIMR conference on
Performance Evaluation, Benchmarks, and Attribution Analysis in Toronto, Bailey outlined six
requirements for a valid benchmark:
- The benchmark must be unambiguous.
- The benchmark must be an investable, passive alternative.
- The benchmark must be measurable.
- The benchmark must be appropriate for the manager.
- The benchmark should be a reflection of the manager's current investment opinions.
- The benchmark must be specified in advance, i.e., before the manager's performance review period begins.
In order for an index to work as an investment benchmark, it must provide an unbiased model of the
market segment it is intended to represent. The S&P 500 has evolved over the years to fill that
need. When the S&P 500 was introduced in 1957, all of its components were listed on the New York
Stock Exchange. The S&P 500 was designed as a sample drawn from that pool to represent the market
performance of the leading companies in the leading industries in the United States. Unlike the
Fortune 500, which simply ranks the largest 500 publicly traded companies in the United States in
terms of sales, companies have never been chosen for the S&P 500 simply because of their size.
However, the initial 500 stocks comprised 90% of the market value of all the companies on the Big
The academic community quickly adopted the S&P 500 as the benchmark used in the evaluation of
investment theories. As part of his work in developing the Capital Asset Pricing Model, William
Sharpe created the notion of the market portfolio, one of the major characteristics of which is
that each asset in that portfolio is held in exact proportion to its market value.9 That also
happens to be the basis upon which the S&P 500 is constructed, a fact noted not only by the
researchers who used it to develop modern portfolio theory but also by the creators of indexed
As both the U.S. economy and the equity markets grew, Standard & Poor's expanded the selection pool
for the S&P 500 to include stocks traded on the American Stock Exchange and over the counter on the
Nasdaq quotation system. The S&P 500 remains broadly based, containing stocks in more than 100
different industry groups ranging alphabetically from Aerospace/Defense to Waste Management. As
noted above, its role as the best benchmark of the stock market's performance is reflected by its
inclusion in the Composite Index of Leading Economic Indicators. As of December 31, 1996, the
stocks in the S&P 500 had a total market capitalization of $5.626 trillion. A year later, the
market capitalization had reached $7.555 trillion.
Putting that figure into perspective, at the end of 1997, there were 7,710 U.S. equity stocks,
valued at $10.320 trillion, in the Standard & Poor's Stock Guide database. The S&P 500 accounted
for approximately 73% of the database's capitalization, and the S&P MidCap 400 Index accounted for
9%. The S&P SmallCap 600, designed as a representative sample of the more than 6,300 remaining
stocks, contained 3% of the database's capitalization, leaving 15% in terms of capitalization
outside of the Standard & Poor's indexes. Together, the 1,500 stocks in all three indexes comprise
the S&P Super Composite 1500, which was introduced on May 18, 1995. At the end of 1997, the Super
Composite had a market capitalization of $8.089 trillion, representing 85% of the total value of
the U.S. equities in the database.
Because of the Index's mandate to select leading stocks in leading industries, the S&P 500 has
evolved into a measure of large-capitalization stocks. The average market capitalization of the
companies in the S&P 500 at the end of 1997 was $15.109 billion, and the median valuation, the
point at which the 500 stocks could be split into pools of the 250 largest and 250 smallest
companies, was $6.881 billion. Furthermore, as has been the case since the S&P 500 was created,
most of the capitalization is concentrated in the largest companies; at the end of 1997, the top 50
companies accounted for 49.33% of the Index's total capitalization, up from 47.61% a year earlier.
However, that dominance compares to the situation at the end of 1986, when the top 50 stocks in the
S&P 500 contained 45.23% of its market cap. In 1986, the 50 smallest stocks in the S&P 500
contained just 0.60% of the Index market cap; in 1997, the smallest 50 contained 0.82% of the
This large-cap bias is not the result of a deliberate attempt to draft for size, but rather is a
reflection of the S&P 500's traditional mandate to include the leading companies in leading
industries. The companies that are added to the S&P 500 did not mushroom overnight; they grew into
their industry leadership positions. Furthermore, once added to the S&P 500, those leading
companies have tended to keep leading their industries.
All Standard & Poor's indexes are constructed with the aim of matching, as closely as is
practicable, the economic sector distributions of the securities universes from which they are
drawn. This allows comparison of Standard & Poor's sector weightings with those of actively managed
portfolios, so that a plan sponsor or pension fund consultant can determine exactly where a manager
is adding (or losing) value in the selection of stocks or the execution of trades. This procedure
also works across all Standard & Poor's equity indexes, because they are constructed to form a
The question of index rebalancing also must be addressed. Stocks in the S&P 500 can never become
too large. However, some stocks have faltered and become relatively small. (Sometimes, entire
industries-including many not generally perceived as cyclicals-fall into hard times; some companies
recover, others do not.) Rather than periodically (or prematurely) remove such outliers from the
Index, as some competing large-cap indexes do on an annual or even quarterly basis, Standard &
Poor's attempts to keep portfolio turnover at a minimum. Every change in a benchmark's components
adds costs for the investment managers replicating it, without necessarily adding value. Heavy
turnover also adversely affects the usefulness of historical data.
The AIMR standards require benchmarks to be consistently applied and to parallel the risk or
investment style that the client's portfolio is expected to track. For example, if a portfolio is
weighted to consist of 50% large-cap stocks and the remainder mid- and small-cap stocks, then the
Standard & Poor's indexes could be used to customize a benchmark. One solution might consist of 50%
S&P 500 Index with the remainder split between the S&P MidCap 400 Index and the S&P SmallCap 600
Index in accordance with the targeted weightings in the client's portfolio. The splits of the S&P
500 into the S&P/BARRA Growth Group and the S&P/BARRA Value Group, and of the S&P MidCap 400 into
its respective S&P/BARRA MidCap Growth and S&P/BARRA Value Groups also provide tools for the
creation of customized benchmarks for style-tilted portfolios. The total returns on each index
segment then can be combined to produce the total-return data for the benchmark portfolio.
Standard & Poor's determines the total returns on its indexes by using accrual accounting of
dividends as of their ex-dividend dates. Cash flows are evaluated daily by this procedure, in
accordance with AIMR's recommended practice. Furthermore, because Standard & Poor's accounts for
cash flows caused by corporate actions (such as spinoffs or new share issuances) daily as they
occur, Standard & Poor's index data provide a true time-weighted rate of return.