Discretionary Decision-Making and the S&P 500 Index

Discretion is an integral part of how indices, including stock market indices, are constituted, according to professors Rauterberg and Verstein and Robertson (here and here), and the S&P 500 index is no exception.

The S&P 500 is a market-weighted compilation of the share prices of common stock issued by 500 companies that are considered to represent blue-chip America. It is governed by the S&P’s U.S. Index Committee (“Committee”), which has almost total discretion in determining the companies that constitute the S&P 500.

In our new article, we find that the Committee’s discretionary decision-making has resulted in the approximately $4.6 trillion of funds that track the S&P 500 providing sub-optimal market returns to their investors.  As a result we do not find the S&P 500 index to be desirable for either tracking or benchmarking purposes, even though our proposed legal disclosures should mitigate any potential legal liability for its continued use.  We base our argument on two critical decisions made by the Committee: the delayed inclusion of Tesla into the index and the 2017 policy decision not to include any more companies with multi-class (“dual-class”) share structures.


On its first day of trading ( June 29, 2010), Tesla closed with a market capitalization of approximately $2.2 billion. This was almost twice the market capitalization of Eastman Kodak, then the S&P 500 company with the smallest market value.  Since then, the market value of Tesla has skyrocketed.  On December 18, 2020, the closing price of Tesla stock was $695 per share.  This meant that the value of Tesla at the close had a market value of $659 billion, or 2.051 percent of the closing value of the S&P 500 without Tesla.  This market value was used when determining the market cap weighting of Tesla when it was included in the index on the following Monday.  The market cap weighting used for Tesla’s inclusion was 1.69 percent.

Empirical Evidence on Tesla

Of course, we could not have expected the Committee to add Tesla on the first day its stock traded, well before the company achieved blue chip status.  However, what if the Committee had a rule allowing Tesla or any other company to be included after recording an average market value greater than the S&P 500’s minimum requirement for four quarters in a row?  If so, then Tesla would have been eligible for inclusion on June 29, 2011.  At that time, the company had a market value of $2.9 billion, 2.1 times the value of the smallest member of the index, RadioShack.

If the Tesla inclusion had occurred on June 29, 2011, the S&P index fund investor would have gained an additional 24 basis points (0.24 percent) in return per year while adding only 4.7 basis points (.047 percent) to the annualized standard deviation of the index.

Implications of Tesla Exclusion

Perhaps the Committee delayed Tesla’s inclusion to avoid adding volatility to the returns of the S&P 500.  However, it would not have been unreasonable to have included Tesla on a much earlier date, allowing its inclusion to make the S&P 500 more representative of the market, based on market value. Moreover, it is hard to understand why the world’s leading electric car maker and the largest U.S. automaker by market value was left out of the S&P 500 for so long.  Whatever the reason, if the Committee would have included Tesla several years before it did, the returns of funds that track the S&P 500 would have significantly increased.

Dual-Class Shares

Some companies may issue common stock with dual-class shares.  Yet, despite dual-class shares arguably being the wealth enhancing result of private ordering, activists have criticized them numerous times over the past century. Driving this perennial disdain for dual-class shares is the belief of many individuals and institutions that the issuance of common stock must only allow for one share, one vote, with no exceptions.

The most recent calls for the elimination of dual-class shares arose from Snap Inc.’s 2017 issuance of non-voting stock.  This offering created an uproar among institutional investors that have a strong focus on shareholder democracy, giving  them and incentive to once again call for a sharp reduction in the use of dual class shares.  As a result, the S&P Dow Jones Indices (July 2017) and FTSE Russell succumbed to pressures from these investors and changed their indices to limit the presence of dual-class shares. The S&P excluded all new dual-class share offerings from the S&P Composite 1500 and its components: the S&P 500, S&P MidCap 400, and S&P SmallCap 600.  So far, the Committee has implemented this rule without exception.

Empirical Evidence on Dual-class Shares

To measure the effect of this decision, we built a portfolio of companies with dual-class shares that were not in the S&P 500 index as of January 1, 2021, despite the fact that their market capitalization exceeded that of the smallest S&P 500 index stocks at the end of each of the past four quarters.

A cap-weighted composite of these multiple-class stocks returned 196.5 percent since August 2017 compared with a total  return of 64.2 percent for the S&P 500 index.  This performance gap may not be entirely due to the exclusion decision because some of these stocks may have been ineligible for inclusion due to other considerations such as not meeting the float requirement, a lack of positive earnings, etc.  Moreover, our test was limited by the lack of historical data on companies with multiple class shares that were delisted or acquired  prior to January 1, 2021.

More specifically, excluded dual-class share companies that met the S&P 500’s market cap requirement represented, on average, 2.2 percent of the index’s market capitalization since August 2017. These stocks outperformed the S&P 500 by 80.5 percent from August 2017 to December 31, 2020.  We can therefore estimate that their exclusion from the S&P 500 led to an approximately 1.7 percent (2.2 percent of 80.5 percent) reduction in the value of funds that closely track the index.

What Motivated the Exclusion?

Scott Hirst and Kobi Kastiel provide a number of reasons for an index provider not to exclude dual-class share companies from their indices.  They also point out, though, that “where particular arrangements are strongly disfavored by their clients [institutional investors], index providers have an incentive to respond to those preferences by adopting an exclusion rule.” This reality acknowledges what many people forget: Index providers are for-profit organizations and must act according to what is best for their own shareholders.

This is a particularly relevant issue for the S&P 500.  The index generates an incredible amount of licensing-fee revenue for the S&P Dow Jones Indices, making it particularly sensitive to the demands of its clients.  We estimate that the index provides S&P Dow Jones Indices with over $1 billion per year in revenue.

Given the S&P 500’s financial importance, any decisions regarding the exclusion of companies with dual-class shares must take into consideration how that decision will impact the retention of institutional investor clients.  Therefore, the feedback that the Committee received from these clients must have been enough to create an exclusion rule.

Implications of the Dual-Class Share Exclusion

The Committee, when dealing with the composition of the S&P 500, is in a much weaker bargaining position with its institutional investor clients than is commonly understood.  It should not, therefore, be surprising to see the Committee eventually giving into pressure to exclude from its index one or more groups of companies, such as fossil-fuel companies, that are not viewed positively by the public or the managers of institutions.  The result would be a serious blow to making the S&P 500 reflective of large cap stocks across all U.S. industries.

Main Recommendation for Enhancing Committee Performance

In determining the companies that make up the S&P 500, the Committee needs to use its discretionary authority to allow funds that track the S&P 500 to create portfolios of stocks that most accurately represent the market risk and expected returns of large cap, blue chip America.  It can do this by minimizing the amount of negotiation with external parties such as institutional investors.  The Committee is in a weak negotiating position, and so continuing to negotiate will erode the risk-return profile of the index, making it an increasingly sub-optimal for investors.

To enhance the returns to investors of funds that track the S&P 500, we recommend that the Committee implement a market value rule that can be used to justify the relatively prompt inclusion of companies like Tesla and rationalize the inclusion of additional dual-class share companies.  This new rule would go far in allowing the Committee’s decision-making to reflect the collective wisdom of active investors, not the clients it must negotiate with.

Our market value rule would require the Committee to periodically include any firm that has been in the top 500 U.S. companies by market value for four consecutive quarters.  Moreover, it would exclude any company that has been below the top 500 for four consecutive quarters.  Exceptions to this rule can be made if it is required to maintain 500 companies in the index.  However, the index may be allowed to go over 500 in order to include qualifying companies.  This would hopefully allow for a quick inclusion of a dual-class company like Zoom, whose market value has consistently been about equal to the minimum requirement and is now approximately 10 times greater than the requirement.

Legal Issues and Recommendation

We do not find any benefit in the Committee becoming a registered investment adviser under the Investment Advisers Act of 1940.  Most important, it is unclear how the fiduciary duties of investment advisers, being quite broad and undefined, would affect the discretion of the Committee in selecting the companies that make up the S&P 500. We fear that an unintended consequence of these duties might be that the SEC becomes a third party to how the S&P 500 is constituted.  This does not appear to be a desirable outcome.  Instead, we find an adequate alternative in the disclosure regime that our securities laws currently provide.

Disclosure Requirements for Investment Funds

We believe that S&P 500 funds need to provide an additional principal risk in both Items 4(b)(1)(i) and 9(c) of the SEC’s Form N-1A and, as a result, in their respective statutory and summary prospectuses.  This principal risk disclosure would focus on selection risk and would disclose that the Committee may not necessarily use its discretionary decision-making to constitute the S&P 500 with companies that are expected to most accurately reflect the risk-return profile of the market it is meant to represent.  We provide proposed wording for such a disclosure in our article.

Benchmarking Disclosures

The SEC’s Form N-1A requires the performance of an actively managed investment fund to be compared with an appropriate broad-based securities market index.  In practice, the S&P 500 is the benchmark index of choice.  As of the end of 2019, actively managed funds worth approximately $6.6 trillion were using the S&P 500 index for benchmarking. Yet, just because it is broad-based and widely used does not mean the S&P 500 is an appropriate index.  Because of its sub-optimal market returns, we do not believe it currently qualifies for purposes of Form N-1A.  The result would be to mislead investors into believing that the actively managed fund is doing better than it actually is.

Finally, because of the significant amount of discretionary decision-making that goes into the governance of the S&P 500, we find that funds that track the index should be considered actively managed.  Therefore, in order to conform to the instructions of Form N-1A, besides tracking the S&P 500 to provide investors information on tracking error, these funds also need to track an additional benchmark such as a total market index or an index that is trying to provide something similar to the S&P 500, e.g., an index that tracks the top 500 U.S. companies by market value.

This post comes to us from Bernard S. Sharfman, a senior corporate governance fellow at the RealClearFoundation and Vincent Deluard, director of global macro strategy at StoneX Group Inc. It is based on their recent article, “How Discretionary Decision-Making Has Created Performance and Legal Disclosure Issues for the S&P 500 Index,” available here.