Whether the rule should be one share, one vote or multiple-voting shares (dual-class stock) is, in international legal and economic terms, highly controversial. The European Union’s 27 Member States as well as the United States all have very different rules. Lately the EU has attempted a harmonization. On February 14, 2024, the European Council agreed on a Multiple Voting Rights Directive, for inclusion in the Listing Act, that permits multiple voting shares as a minimum harmonization. The European Parliament agreed on 24 April 2024. This is a step in the right direction but falls short in that it is limited to the SME growth market. The EU Member States are trending towards broader approval of multiple-voting shares. Germany recently executed a complete turnaround in a law of 11 December 2023 that largely permits multi-voting shares, but only with limits and clear restrictions for listed companies to protect other shareholders. France followed suit with a law of June 2024, and an Italian law of March 2024 went in the same direction. The reason for these laws has been concern for national stock exchanges, the promotion of start-ups and the competitiveness of the respective country. Economically, the question is still highly controversial. Traditionally, agency theory predominately preferred the one-share, one-vote principle due to the risks for non-multiple-voting shareholders. More recent views tend to see and emphasize more the overall advantages. A recent article of ours discusses the legal situation in the 27 EU member states in comparison with the United Kingdom and the United States and deals with the pros and cons of multiple-voting shares from an economic and legal policy perspective:

Under the European Directive of 2024, the only companies the law of the EU member states must permit to introduce multiple-voting shares (as per their articles of association) are those contemplating an initial listing on an MTF. The mandate that multiple-voting shares must be authorized relates only to the MTF admission phase. The crucial idea is that the founders and original owners of the shares should be able to exercise significant influence beyond an IPO despite the eventual listing of the company’s shares. But the Directive contains various mandatory safeguards. The Directive stipulates that the decision to introduce multiple-voting shares should be by a qualified majority of the general assembly as specified in national law. Multiple-voting shares must be open to all legal entities. The Directive itself also does not stipulate a maximum voting-rights multiplier. The Directive also specifies a series of optional safeguards that it groups into three categories, each a type of sunset clause, i.e., a clause that stipulates when the multiple-voting structure must expire under the articles of association: upon first transfer by the first owner to a legal successor (a transfer-based sunset clause); after a certain period of time (a time-based sunset clause); or upon the occurrence of a certain event (an event-based sunset clause).

The Directive is a step in the right direction. But in being limited to the MTF market (including the SME growth market), it falls short, and it would have been better to permit multiple-share structures for all listed companies, though not at the cost of overly-strict mandatory shareholder protections. Instead, the Member States should have been given regulatory leeway, with private ordering as the guiding principle for articles of association, but with certain limits and safeguards for listed companies. The directive should have specified a maximum multiple-voting ratio of up to 10:1. Limiting multiple-voting structures to ten years’ duration without the option to extend is too restrictive, and a good compromise is to permit at least one ten-year extension. It would have been unfounded to restrict the exercise of multiple-voting rights on resolutions concerning human rights and the environment in particular, as contemplated in the European Commission’s draft; the final version of the directive rightly abandoned this position. And it ought to be clear that the limits on “golden shares” developed by the European Court of Justice cannot simply be side-stepped through the use of multiple-voting shares.

The picture that emerges from the comparative law of multiple-voting structures is anything but uniform. Countries such as the United States, the Netherlands and the Scandinavian countries that mostly or fully permit multiple-voting rights stand alongside countries that mostly or completely reject them, as Germany did until 2023 and as Austria and Israel still do. Others have liberal regulations for unlisted companies while allowing multiple-voting structures for listed companies but with limits and safeguards for other shareholders. In a fundamental reform, Germany introduced a system of this kind in December 2023, that broadly allows multiple voting structures for unlisted companies but imposes clear restrictions for listed shares. Countries such as Italy and France that allow loyalty shares – generally with double-voting rights and subject to a holding period of at least two years – traditionally have not permitted multiple-voting shares for listed companies. However, this has shifted in both countries with enactments in 2024; in June, the French legislator passed a law allowing multiple-voting shares in the context of the first admission of a company’s shares to trading on a regulated market or an MTF. These shares can only vest in one or more named persons with a voting rights ratio of no more than 25:1 over an ordinary share in the case of admission to trading on an MTF. They are also limited in time (ten years plus an optional five-year renewal), and in the event of transfer are converted into ordinary shares, which can only confer double voting rights (loyalty shares). In regard to the appointment of auditors and approval of certain items (annual financial statements; amendments to the articles of association other than capital increases; related party transactions; say-on-pay votes), these shares only confer a single voting right, including in the case of takeover bids if the company’s articles of association so stipulate. As of March 2024, Italian law retains the loyalty-share regime (as per articles of association) that affords a maximum of two votes if held for no less than 24 months; now, however, the law also allows each share to be awarded an additional vote each year after the initial two years, up to a maximum voting-rights multiplier of 10. A significant liberalization is also imminent in the UK. Overall, it can be said that a clear majority of EU member states already allowed multiple-voting shares, and since 2023 leading countries such as Germany, France, Italy and the UK have either admitted multiple-voting shares for the first time or become (or are set to become) significantly more liberal. Comparative law offers important insights and guidance for regulators on multiple-voting share structures in this context, particularly with regard to limits and safeguards for non-multiple-voting shareholders. An overview of the 27 EU member states’ positions on multiple-voting shares can be found in the appendix to the above-mentioned article.

The comparative law findings say nothing about how multiple-voting shares are to be valued economically. An ample body of economic theory supplies arguments for and against multiple voting structures. Traditionally, the one-share, one-vote principle was considered preferable due to the risks for non-multiple voting shareholders (agency theory). Institutional investors and proxy advisers overwhelmingly reject multiple-voting shares. More recent opinions, however, tend to see social advantages and benefits for shareholders, stock exchanges, the economy and international competitiveness. Empirical studies remain relatively scarce and tend to contradict one another. Conventional (mostly older) studies applying agency theory have found greater control costs and loss of value associated with deviations from one share, one vote. More recent studies have found no loss of value or even gains, but also that the advantages of multiple-voting structures decline after seven to ten years and that the agency costs increase. But complete deference to private ordering of the articles of association also seems ill-advised from an economic point of view. The power of the IPO market does not appear to be sufficient to ensure optimal corporate governance in the case of multiple-voting rights. A balance must be found.

For legislators, the comparative law, economic theory and empirical findings surrounding multiple-voting share structures provide an important basis for decision-making. Multiple-voting shares should be registered and should be permitted by law on all markets. But in the case of listed shares, the authorization of dual-class structures must be accompanied by limits and safeguards for other shareholders. For unlisted companies, the articles of association should be unconstrained. However, the introduction of a multiple-voting structure (such as in an IPO) should require a qualified resolution on the articles of association and the consent of all affected shareholders. The measure need not be accompanied by a substantive justification. As for the limits and safeguards for non-privileged shareholders, a wide range of regulatory modules is already available. The article deals in particular with this arsenal, which so far has been under-researched.

Limits on the voting weight and occasions for the exercise of multiple-voting rights are essential. A multiple-voting right factor of 10 is common internationally. It should never be permissible to exercise multiple-voting rights on resolutions to appoint auditors, and while such is proscribed in German law, there is also much to be said for the stricter French solution, which covers approval of the annual financial statements, amendments to the articles of association other than capital increases, approval of related party transactions, and say-on-pay votes. The proscription could extend to resolutions to assert or waive claims for damages and possibly to take defensive actions against hostile takeovers. However, the exercise of multiple-voting rights to elect board members should not be prohibited; such a rule would sabotage the incentivizing purpose of multiple-voting structures.

Multiple-voting shares must be offered to all shareholders, not just to board or other executive-body members. In the case of founders and start-ups, the interest is in them being able to focus on their mission and exert influence without performing these kinds of corporate functions. A minimum shareholding or “skin in the game” is important and was included in the draft European directive but was dropped from the final legislation. There is much to be said in favor of 10% of the share capital for a multiple-voting shareholder. A good alternative option would be a provision – as in German stock corporation law – under which at least 50% of the share capital must be in shares that are neither multiple-voting shares nor non-voting preferred stock.

Multiple-voting rights serve to attract founders, capital and innovators. An unlimited duration or unlimited transferability of the multiple-voting right is contrary to the specific purpose of the multiple-voting structure. Comparative law as well as the economics literature shows various pros and cons of sunset clause options, which are highly controversial. German law provides a time limit of ten years with a one-off, ten-year extension by way of amendment to the articles of association (carried without the exercise of multiple-voting rights). This is a justifiable regulation. Another sound regulation would be to provide a second ten-year extension (this time carried by only the non-multiple-voting shareholders) for certain cases such as family companies. Under German law, multiple-voting rights expire upon transfer whether the shares are sold or pass by operation of law. This rule poses difficulties, at least in the case of family companies and possibly also in the group setting. It would be better to have legislation that restricted the free transferability of multiple-voting shares subject to amendment to the articles of association carried without multiple voting or with only non-privileged shareholders participating.

With multiple-voting rights, transparency or disclosure is an absolutely essential requirement for the protection of the other shareholders. The multiple-voting structure should be indicated in the articles of association as well as in the commercial register and on the company’s website. Follow-up regulations in takeover law as it concerns elements of stock corporation and capital market law seem necessary, in particular to clarify that passive attainment of control as a result of the termination of multiple-voting rights does not trigger a mandatory bid as provided for under European law.

Once legally authorized and adopted by companies, multiple-voting share structures raise considerable issues of application and practice and lead to a host of doctrinal questions for the legislators, courts and legal literature in European and other jurisdictions to resolve. The legislator’s decision to evaluate in a few years the effects of their imposed limits and safeguards for shareholders (in two years in the EU, four years in Germany) seems sound. In this evaluation, the legislator should bear in mind the numerous other ways of decoupling economic ownership from control.