When Snap went public in March 2017, it sold the public a class of shares with no votes at all, an arrangement corporate-governance scholars flagged as the first no-vote listing at IPO on a U.S. exchange since the New York Stock Exchange generally barred multi-class common structures in 1940 (Harvard Law School Forum on Corporate Governance, 2017). The two co-founders retained roughly 88.5 percent of total voting power (CNBC, 2017). Buyers were invited to fund the enterprise and explicitly excluded from governing it.
Snap was an extreme case of a common design. At Meta, Mark Zuckerberg's super-voting Class B shares gave him control of almost 60 percent of votes with economic exposure to about 14 percent of the company (CNBC, 2018). The structure separates two things that classical shareholder capitalism bundled together: owning the value and steering the ship. The result is a cap table that reads like a feudal muster roll. At the top, founders whose authority is constitutionally insulated from removal. In the middle, preferred investors: knights with negotiated privileges, liquidation preferences, board seats, information rights. At the bottom, employees holding common stock, the class that is last in preference, least informed, and, in the dual-class variant, votes least or not at all.
The vision defense, and its discount rate
The standard defense of dual-class structures is protection of the founder's long-term vision from short-term markets. Sometimes the vision is real. But the argument leans on an attribution bias that organizational research identified decades ago. Meindl, Ehrlich and Dukerich (1985) called it the romance of leadership: the tendency to over-attribute organizational outcomes to the person at the top. In their studies, the same executive was rated far more charismatic when the company happened to be doing well; charisma was something followers conferred, not something leaders possessed. A governance structure that makes the founder unremovable is a bet that this attribution is accurate, placed at the one moment (a hot IPO) when the romance is strongest and the discount applied to it is weakest.
There is also a scope confusion in the defense. Insulating strategy from quarterly earnings pressure is one argument; insulating a person from all accountability to owners and employees, permanently, is another. Dual-class charters typically deliver the second while advertising the first. Institutional investors noticed: the Council of Institutional Investors objected to Snap's structure before the listing and petitioned index providers to exclude no-vote companies (Harvard Law School Forum on Corporate Governance, 2017).
What the employee actually holds
For employees, the structure has a specific consequence: the equity that recruiters describe as ownership carries the economics of ownership (diluted, deferred, junior) without its constitutional content. A common-holding employee cannot vote out a leader, cannot compel disclosure, and in a no-vote structure cannot vote at all. She is exposed to the decisions of the controlling founder in both her portfolio and her paycheck, a concentration of risk no financial adviser would recommend, with no corresponding voice in the decisions that drive it.
This is where the feudal vocabulary earns its keep, not as insult but as description. Feudal orders were not lawless; they were elaborately legal. Everyone's rights were specified, unequal, and hereditary in practice. A dual-class charter does the same work: it writes the hierarchy into the corporate constitution itself, beyond the reach of anything the lower tiers can do at the ballot.
Power that must be re-earned
The alternative is not shareholder populism; it is accountability with a renewal date. As Ricardo Semler has told it, Semco had every employee anonymously evaluate their boss twice a year. Two low scores in a row and the manager lost the function, a rule that applied to Semler himself, with a return to well-paid technical work as a legitimate exit. Separately, teams could veto a manager outright. His premise was simple: anyone who controls other people's days has to keep deserving to.
Set that against the dual-class charter and the design question becomes visible. One system says authority is a role, held on performance, reviewed by the governed, and losable. The other says authority is a property right, acquired at founding, and reviewable by no one. Companies are free to choose the second; capital markets have priced and accepted it. But the choice should be named for what it is. It is not a governance detail. It is the difference between a leader and a lord, and every employee holding voteless common stock already knows which one signs her grant.
Sources
- Harvard Law School Forum on Corporate Governance (2017). Snap and the Rise of No-Vote Common Shares. Snap's March 2017 IPO was the first no-vote listing at IPO on a U.S. exchange since the NYSE generally barred multi-class common in 1940; CII objected before the listing. https://corpgov.law.harvard.edu/2017/05/26/snap-and-the-rise-of-no-vote-common-shares/
- CNBC (2017/2018). Coverage of Snap and Facebook voting structures. Snap's co-founders retained ~88.5% of voting power; Zuckerberg controlled almost 60% of Facebook votes with ~14% economic exposure. https://www.cnbc.com/2018/03/20/shareholders-wont-force-zuckerbergs-hand-in-facebook-management.html
- Meindl, Ehrlich & Dukerich (1985). The Romance of Leadership (Administrative Science Quarterly). Observers over-attribute organizational performance to leaders; perceived charisma tracks results, not traits. https://doi.org/10.2307/2392813