The inclusion of non-voting or limited voting shares in stock market indices

ICGN Viewpoint

November 2017

Introduction

This Viewpoint follows an earlier ICGN report  on how differential ownership structures allow control over publicly listed companies that are disproportionate to the economic interests and cash flow rights of investors. ICGN’s long standing policy position is that the optimal share structure for companies wishing to benefit from access of public capital should be one vote for each share within a same class. ICGN’s general view is that the mismatch of ownership and control through differential ownership rights can create significant governance risks for minority shareholders such as management entrenchment and a lack of accountability.

This new Viewpoint is an extension of ICGN’s policy position and explores whether or not non-voting or limited voting shares should be included in benchmark stock indices. This is of particular concern given the inclusion of such shares in passive investment strategies that are index-based. ICGN’s normative position is that dual class shares should not be included in benchmark stock market indices.  However ICGN also recognises that companies with dual class shares are listed on many stock exchanges around the world and that it could be disruptive to markets if drastic measures are taken to cleanse all benchmark indices of dual class shares.  

This issue warrants an aspirational vision as well as pragmatic considerations, including having minimum minority voting rights thresholds, introduction of sunset provisions for dual class arrangements and encouraging index providers to not permit additions to existing indices for new differential share issues. In this regard, ICGN applauds the leading index providers for their public consultation on these issues and we believe positive progress is being made, while recognising that there remains scope for the index providers to strengthen further their positions on this issue.

Differential ownership pros and cons

The protection of minority shareholder rights is an ICGN policy priority. Risks and opportunities for minority shareholders exist in both dispersed and controlled ownership structures. With good governance and capable executive management, both forms of ownership can provide investors with sustainable risk-adjusted returns on capital. However differing risks present themselves as well in both ownership structures.

In a widely dispersed share ownership model the key agent of concern is executive management and its potential ability to exploit private benefits of control at the expense of minority investors and the company’s own long-term interests. In a controlled company, the agency problem mainly relates to potential lack of alignment between the controlling owner and minority shareholders, where the controlling owner is in a position to gain private benefits from control—again at the expense of outside shareholders. 

Corporate governance mechanisms can be introduced to mitigate risks in both controlled and dispersed share ownership. However, share ownership structures that allow disproportionate control via voting rights over publicly listed companies relative to economic interests are a particular risk for good governance. They can lead to entrenchment of control and an erosion of accountability to the detriment of a company’s investors and stakeholders – and can threaten the company’s own long-term success.  In this regard, ICGN’s Global Governance Principles  emphasise the importance of equal voting rights for all shareholders and call for explanation of any divergences, as well as commensurate extra protections for minority shareholders. 

We understand that differential ownership rights are viewed by some companies and regulators as a viable way to encourage long-term share ownership and to protect companies for the short-term pressures of the financial markets. For example, this contributed to the rationale for differential ownership rights arrangements being legally encouraged in France though the Florange Act. It may also account for the presence of dual class share structures in the high tech sector of California’s Silicon Valley. 

ICGN is supportive of the need to reinforce long term perspectives in investing, and we recognise that the influence of short-term investment horizons can work against a company’s ability for sustainable value creation. In early stages of a company’s life cycle there may be justification in some cases for differential ownership to provide “patient capital” for developing companies. However the circumstances of companies change as a company’s life cycle evolves. We believe that differential ownership structures are a flawed long-term solution to this potential of market short-termism. This form of ownership separates economic ownership from control and can produce unintended and undesirable consequences. As such, ICGN has successfully advocated against the introduction of such structures in a number of jurisdictions globally. 

Differential voting rights and benchmark indices

The growth of index-based passive investment strategies adds a new dimension to this issue, and has given rise to concerns around the inclusion of companies in benchmark indices with shares or share classes with differential voting rights. To cite one example, the recent IPO of Snap Inc. in the US, offering no voting rights to minority shareholders, prompted considerable controversy as to whether its non-voting shares should be included in the US benchmark indices.

Those passive funds with a mandate to invest based on an index composition are contractually obliged to hold the securities in the index, notwithstanding any corporate governance concerns about differential ownership rights. This results in a “captive market” of passive investment funds, and ultimately distribution to retail savers and pensioners. This can put index providers in the difficult position of having to invest in companies they might not otherwise invest in on the basis of disproportionate voting rights. 

Index provider consultations

ICGN applauds S&P/Dow Jones,  FTSE Russell and MSCI for undertaking public consultations in 2017 on the treatment of non-voting shares in their benchmark indices. Each consultation had its own distinct set of questions and varied in scope and detail. But all were aimed at understanding and addressing investor and company perspectives on these points. The main thrust of these consultations was to get a sense as to whether or not there should be index exclusions based on differential ownership rights, and, if so, under what conditions. In particular, there was an attempt to find scope for compromise to determine if certain mitigating features might be acceptable, such as sunset provisions or a minimum level of voting power by minority shareholders.

Consultation reactions: positive, but far reaching enough?

At the time of writing, both S&P/Dow Jones and FTSE/Russell already have announced changes to their index policies relation to this debate.  MSCI has chosen to broaden its consultation on this issue to a wider range of market participants before reaching its final conclusions.

In the case of S&P/Dow Jones, it was announced that their composite indices would exclude companies with multiclass. While quite far reaching, this approach grandfathers the existing index constituents, so the impact will be limited to new listings. A significant group of issuers in the US market will be unaffected. 

In the case of FTSE/Russell, it established 5% voting rights threshold as a minimum criteria—if nothing else this eliminates the potential for zero-vote share classes being included in indices, such as in the case of Snap Inc. Like S&P/Dow Jones, FTSE/Russell will grandfather existing issuers in this category with less than 5% voting rights. But it takes the additional measure of giving issuers a five-year time period to introduce a 5% threshold, at which point the issuer would be removed from the index if this threshold were not achieved. 

The fact that both S&P/Dow Jones and FTSE/Russell have agreed to take action at all is a positive development. It reflects recognition of the merits of the issue and its concern to investors. At the same time, there is more that could have been done; it is clear that that the index providers do not want to be disruptive to markets, particularly where multi-share structures are well established.

So while we welcome the direction of FTSE/Russell’s change, in practical terms a 5% voting threshold provides little, if any, material support to minority shareholder rights and protections. There could be more done in this domain.  And while we also welcome S&P/DowJones’ proposal regarding future listings, we believe there could be scope for implementing tougher standards the incumbent companies: S&P/Dow Jones could reconsider some its grandfathering arrangements – perhaps providing a time frame to introduce sunset provisions or other changes to the capital structure as a requirement for index membership. 

Pending the results of this broader consultation, MSCI has “temporarily” made unequal share structures ineligible for new introductions to its indices while leaving unaffected dual class shares currently in its indices. This is also a welcome recognition of the merits of this problem, but its longer term solution remains an uncertainty. 

Does index exclusion deprive passive investors of opportunity?

ICGN is aware that exclusion of companies with differential rights structures could limit the investment opportunity set for passive investors—and could therefore have the potential, at least in the near term, for greater systematic risk (through lesser diversification in a modern portfolio theory context) and weaker investment returns, as compared to an index that does not provide such exclusions. This has prompted some institutional investors, including some ICGN members, to discourage index exclusion based on differential ownership structures. 

We understand and appreciate this logic, but taken over a longer time frame, and with a view towards the systemic integrity of financial markets, we believe that this perspective may be unduly focused on near-term outcomes. Particularly with a view towards promoting long-term value creation and sustainable capital markets we believe there is a more compelling economic logic for either excluding differential ownership structures from indices or introducing risk mitigants such as minimum minority voting thresholds.

ICGN position on voting power and index eligibility

ICGN’s responses to these index provider consultations drew from our Global Governance Principles, our own policy work in ICGN’s Shareholder Rights Committee, and informal consultation with ICGN members. ICGN positioning takes a global perspective which is not based on any specific jurisdiction. From this policy foundation, and from our own consultations with ICGN members, we have opposed in our consultation responses the inclusion of low or non-voting shares in investible benchmark indices.  

ICGN recognises the reality that in many cases issuers of dual class shares are permitted  in indices in major financial markets. While that is not ICGN’s preferred position, in such cases safeguards should be considered, which could include sunset provisions for dual class/dual voting arrangements or ensuring a minimum level of voting rights for minority shareholders. 

We support the views of FTSE Russell and MSCI that listed voting power could be a key criterion for determining index eligibility, as this would offer basic protection and a reasonable degree of control over the future of the company to non-restricted shareholders. We would emphasize, though, that it is not the total listed voting power that matters, but the listed voting power held by non-restricted shareholders - i.e. voting power of the investable free float. 

In our response to the MSCI consultation, ICGN stated in principle that the listed voting power held by non-restricted shareholders should be at least 50% + 1 of the total voting rights (listed and non-listed). From this perspective, any company where listed voting power held by non-restricted shareholders (i.e. voting power of the investable free float) is below this threshold should not be eligible for inclusion into benchmark indices. 

We recognise that this 50% + 1 threshold is much higher than other threshold levels under discussion, including the 25% threshold discussed in the FTSE/Russell consultation. But we believe there may be an inappropriate comparison between free float and voting rights as these measures address two distinct issues. Free float fundamentally deals with market liquidity, and 25% free float (or less in some cases) may be sufficient for trading purposes. 

But market liquidity is not the same thing as the protection of minority rights, and a 25% voting rights threshold would still leave substantial practical limitations in terms of the interests of minority shareholders or improving the quality of benchmark indices. Hence, while ICGN presents a 50% + 1 target as a ultimate aspirational target for minority voting rights in dual class we believe that a 25% threshold is a more realistic near-term waypoint in that longer term journey, so we would be supportive of more immediate considerations by index providers to introduce 25% minority voting rights as a minimum threshold for index inclusion. 

Conclusion

Positive developments have been made in addressing the concerns about multiclass shares in benchmark indices, and it is very encouraging to see that the index providers are open to examining their policies and indeed have made some significant changes.  However encouraging, we believe that the changes announced thus far could have gone further in protecting investors from differential ownership in benchmark indices.  So there is yet more to be done here, and ICGN will continue being active in this debate.

In the meantime several practical steps can be taken by investors. These include:

Engaging with companies with existing dual class structures to introduce sunset provisions that will result in their elimination over time as the company evolves in its own life cycle.

Encouraging index providers and companies with differential ownership structures to provide for at least 25% minority voting rights on an aspirational journey over time to 50%+ 1.

To encourage index providers at least not to exacerbate the current problem by not adding new dual class shares into benchmark indices. 

We encourage ICGN members who support this policy positioning—particularly those asset owners and asset managers with passive index strategies -- to continue their own direct engagement with the index providers and companies..  It is important that the buy-side voice is strong and clear in this debate as a counterbalance to the incentives of the sell-side community to bring new dual class structures to the market. The role of the stock exchanges and regulatory authorities are also critical to preserve a regulatory “race to the top” – or at least avoid a “race to the bottom”.

 

About ICGN Viewpoints

ICGN Viewpoints provide opinion on emerging corporate governance issues and are intended to generate debate, whilst not defining a formal ICGN position on the subject. ICGN Viewpoints are produced by our member-led Policy Committees and we encourage dialogue by contacting the ICGN Secretariat or policy committee chairs as follows:

George Dallas, ICGN Policy Director: george.dallas@icgn.org

Eugenia Jackson, Co-Chair, ICGN Shareholder Rights Committee: Eugenia.Jackson@AllianzGI.com

Bram Hendriks, Co-Chair, ICGN Shareholder Rights Committee:

BHendriks@ktmc.com