Dual-class share structures involve two or more classes of authorized common stock, with one class having the traditional one vote (or, in the rare case of Snap, no votes) per share, and the other class having multiple votes per share. High-vote shares are usually allocated to certain pre-IPO stockholders – the founders, a combination of founders and pre-IPO investors, or all pre-IPO stockholders – and allow those stockholders to maintain effective control of the company for some period after completion of the IPO, even as economic ownership of the company becomes widely dispersed among new public stockholders.
The reasons for implementing dual-class can differ among companies. For some, founder control – based on the “visionary founder” theory – is paramount and ensures that the founder is permitted to achieve their innovative vision for the company by protecting against activist shareholders and other hostile actions. In other cases, allowing the pre-IPO investors to maintain control permits the company to ease its way into being a mature public company by keeping control in the hands of early investors who best understand the long-term vision of the company and out of the hands of index funds, day traders, and arbitragers who may be more focused on short-term results. Later-stage private companies may implement a dual-class structure long before an IPO as they seek to ensure voting control in light of liquidity transactions for founders and existing investors. Private equity-sponsored companies, family-owned companies, and controlled companies spun out from a parent company may also find a dual-class structure to be beneficial in order to allow the relevant stakeholders to maintain control for a time after an IPO or strategic carve-out.
Dual class staying power
Dual-class structures have long been controversial because they give a concentrated number of company insiders increased relative power over the business, even if they only hold a small percentage of the stock. Proponents of the structure argue that it provides founders, major investors and management with greater autonomy to implement business plans with minimum sway of short-term investors, who are usually more interested in quarterly profits – therefore offering the company a better chance at realizing its long-term vision and value. On the other hand, critics like to point out that information asymmetry is inherent within this unequal voting structure, and management has less accountability to the stockholders with a smaller economic ownership (which may become even more diluted over time if the company issues additional shares), so corporate governance risks tend to be greater due to limited board regulation.
That said, the April 2024 decision in In re Match Group Inc., Derivative Litigation may keep a check on fiduciary duties of controlling stockholders. In Match, the Delaware Supreme Court held that controlling stockholder transactions (other than freeze-out transactions) in which a controlling stockholder “stood on both sides of a transaction with the controlled corporation and received a non-ratable benefit” would be held to the entire fairness standard, a more stringent standard than the more deferential business judgment standard of review that applies to other transactions and board decisions.
While the holding in Match primarily centered on the standard of review in controlling stockholder transactions (you can read about it in much more detail), the holding may impact the fiduciary duties of controlling stockholders in dual-class companies in a number of circumstances, such as in compensation decisions for an executive who is a controlling stockholder. In Match, for example, the controlling shareholder owned 24.9% of Match Group’s common stock and all of the high-vote class B shares, giving it 98.2% of Match Group’s voting power. Despite their disproportionate voting power, controlling shareholders must keep in mind that they owe fiduciary duties to minority shareholders, and any transactions where there may be a conflict of interest will be scrutinized under the more stringent entire fairness standard.
In order to avoid the entire fairness standard, a company must follow the conditions set forth in Kahn v. M&F Worldwide Corp., as outlined below:
- Form an independent committee to negotiate the transaction.
- Obtain a vote of the majority of the minority (i.e., non-controlling) stockholders.
- Structure the transaction as an arm’s length negotiation.
- Ensure that the controlling stockholder does not receive a non-ratable benefit that is not shared with the minority.
These requirements were originally used in the context of a merger involving a controlling stockholder, so a stockholder vote was already an element of the transaction. Following these conditions every year to approve the compensation of a CEO who is deemed a controlling stockholder is simply not practical.
While there has been a slight dip in the prevalence of multi-class companies going public between 2021 (the last year, we might say, of a hot IPO market) and now, multi-class structures continue to remain a feature in tech IPOs. Based on data from Jay Ritter of the University of Florida, 32% of all IPOs in 2021 had a multi-class share structure, compared to 46% of tech IPOs in 2021. Similarly, in 2023, 26% of all IPOs were multi-class, compared to 44% of tech IPOs that year. And though the tech IPO market has seemingly opened only in fits and starts over the past couple of years, dual-class structures remain a feature in many of them. In 2023, Klaviyo and Oddity both went public with dual class, and in 2024, Reddit, Ibotta, Rubrik, Tempus AI, Onestream and ServiceTitan all had multi-class structures.
Moreover, certain policy changes may reflect a more general warming toward dual-class companies. Notably, the S&P Dow Jones Indices announced in April 2023 that companies with multi-class stock are now eligible for inclusion in the S&P 1500 and its component indices, which include the S&P 500, reflecting a reversal of the prior policy that prohibited companies with multiple classes on its indices. In 2024, Yale evaluated the stock performance of all companies in the Russell 3000 with dual-class shares versus companies in the Russell 3000 with single-class shares, through October 3, 2024. The study found that the companies with dual-class shares, on average, outperformed the companies with single-class shares across both the short and the long term. Moreover, while some major tech companies – including News Corp, Alphabet and Meta Platforms – have faced shareholder proposals to eliminate the companies’ multi-class capital structures, such shareholder proposals have failed.
Key features
Below are the key features that we have seen in dual-class companies:
Voting ratios
The voting ratios between high- and low-voting stocks have not varied much in the last few years. The ratio typically ranges between 10:1 and 20:1, though in rare cases certain classes of shares can have a ratio as high as 50:1 (typically reserved for a very limited number of stockholders or executive officers). Ultimately, the ratio is just a means of getting the applicable stockholders the right amount of voting control over the desired period of time (in light of anticipated sell-downs, distributions, etc.), so a 10:1 ratio at one issuer could have the same practical effect as a 50:1 ratio at a different issuer.
Sunset provisions
As the name suggests, a sunset provision contractually ends the dual-class share structure. These automatic sunsets can be triggered by (i) the passage of time, (ii) an ownership threshold, where the high vote stockholders’ ownership falls below a specified threshold[1], or (iii) a combination of the two, where the sunset is triggered upon the earlier of the two specified events.
The market has debated the inclusion and length of sunsets for dual-class companies, but has generally coalesced around a sunset of seven years or less. Both Institutional Shareholder Services (ISS) and Glass Lewis will recommend voting against or withholding votes from directors of companies that have multi-class share structures without a time-based sunset of seven years or less; even with a sunset, dual-class companies get a ding on their governance scores simply for having more than one class of stock. The Council of Institutional Investors (CII) advocates against dual-class stock, maintaining a list of “dual class enablers” comprised of directors who sit on boards of companies with dual-class shares without a time-based sunset of seven years or less.
Nonetheless, according to data from CII, of the nine companies that went public in the first half of 2024 with unequal voting rights, five had a sunset, and four had no sunset – a nearly even split. Of the class of 2023 and 2024 tech IPOs, we saw sunsets of 20 years (Tempus AI), 15 years (ServiceTitan), 10 years (Rubrik), and seven years (Ibotta, Oddity and Klaviyo). Several of these sunsets were also combined with ownership-based sunsets, triggered when the high-vote stock dipped below anywhere from 33% of the outstanding shares (Oddity) to 20% (ServiceTitan), 8% (Reddit, which did not have a time-based sunset), or 5% (Ibotta). Lastly, several of these recent tech IPOs provided for a founder-based sunset, in which the dual-class structure sunsets after the death or incapacity of the founder and/or the founder is no longer providing services to the company. Such founder-based sunset appears to continue a line of founder-based tech companies that included this provision. Facebook may be the most famous example of this, although Tempus, Rubrik, Ibotta and Oddity all have similar founder-centric sunsets.
Transfer rights
When venture-backed dual-class company IPOs were still relatively new, there were three broad categories of permitted transfers of high-vote stock:
- Transfers to trusts or other legal entities not resulting in the holder losing control over the disposition or voting of the shares.
- Pledges of shares as collateral.
- Limited transfers of voting rights for ministerial or nonrecurring purposes, such as proxy voting.
Controversially, other permitted transfers included transfers to the holder’s family members, which led former SEC Commissioner Robert Jackson Jr. to call high-vote stockholders corporate royalty back in 2018. These days, additional permitted transfers include transfers to other limited partners so long as the holder retains control or voting power, and in some cases, transfers among other high-vote stockholders.
Recent discourse has been around exceptions for voting agreements. Our colleagues in Cooley’s M&A group provide an excellent analysis of voting agreement carve outs in dual-class company charters, with a discussion of why an express carve out permitting high-vote stockholders to enter into voting agreements in connection with change-of-control transactions is recommended.
[1] See for example Twilio, Fastly, MongoDB, Pure Storage, Yelp and Box, each of which went public with dual class but have since converted all remaining Class B shares to Class A shares when the high-vote stock ownership fell below the specified threshold, triggering an automatic sunset.
