One stock, one vote: the rule Google and Meta break

2026-08-21 · By Lubin Danilo, founder of Lubin Investment

The rule of one share, one vote is not universal. At Alphabet and Meta, founders hold shares worth ten votes each and keep control with a fraction of the capital. At Ferrari, loyal shareholders earn an extra vote instead. Here is how it works, and why.

The rule everyone assumes, until you read a prospectus

When you buy a stock, you almost always assume the same implicit rule: one share, one vote. The more shares you hold, the more say you get at the shareholder meeting, always in exact proportion to the money you put in. That is the norm on nearly every stock exchange, and it is the rule that justifies calling shareholders owners: whoever puts the most capital at risk logically gets the biggest say over who sits on the board, the long term strategy, and any merger.

Yet this rule has exceptions worth several trillion dollars of combined market value. Take Alphabet, the parent company of Google. According to its latest proxy statement filed with the US stock market regulator, the SEC, as of April 6, 2026 Larry Page alone controlled 27.4% of total voting power, and Sergey Brin 25.3%, for a combined 52.7%. A majority reached without either of them needing to own half the capital: their shares, called Class B, carry ten votes each at the shareholder meeting, versus a single vote for the ordinary shares anyone can buy under the tickers GOOGL and GOOG.

The most telling detail is the difference between GOOGL and GOOG itself, which confuses almost everyone the first time they notice it. Both track the same price, the same company, the same quarterly results. But GOOGL (Class A) carries one vote per share, while GOOG (Class C) carries none. This third class was created in 2014, during a stock split, for a specific purpose: letting Alphabet keep issuing new shares, to pay employees in stock or fund an acquisition, without ever diluting Larry Page and Sergey Brin's voting power. A share, in this specific case, is no longer a slice of power at all: it is purely a slice of the company's value, the right to a proportional share of profits, with zero say over who runs it.

How to keep control while owning only a fraction of the capital

Meta, the parent company of Facebook and Instagram, pushes this logic even further on the economic side. According to the company's 2025 proxy statement filed with the SEC, Mark Zuckerberg holds roughly 99.8% of Class B shares, the ones worth ten votes each, yet his actual economic stake in the company, the share of profits and assets he would receive if it were liquidated, is only about 14%. The result: he controls roughly 61% of total voting power with less than a seventh of the capital. That figure is not fixed: at the 2012 IPO, his control stood at just under 56%. The increase since then comes mostly from Meta's large annual buybacks of ordinary shares, which mechanically shrink the number of Class A shares outstanding without ever touching Zuckerberg's Class B share count, so his own voting power never shrinks along with it.

This imbalance is not an accident, it is a deliberate decision made before the IPO, and the reasoning behind it deserves to be understood before it is judged. The argument founders who choose this structure repeat, echoed for years in financial literature, is protection from short term pressure. A board re-elected every year by a scattered shareholder base might be tempted to punish an expensive bet whose payoff only shows up a decade later. Zuckerberg cites exactly this reasoning to justify the massive spending poured into Reality Labs since 2020, Meta's virtual reality and metaverse division, a business that has racked up more than 60 billion dollars in operating losses over that period according to Meta's own quarterly results, without a single shareholder vote ever able to challenge that choice.

What matters technically is that a share class is nothing more than a box checked in a company's bylaws at the time it is founded or goes public. No law forces a company to issue shares with equal votes: it is a choice, disclosed in the prospectus, that an investor implicitly accepts by buying the stock. The catch is that the choice is made once, at the exact moment a company most needs to win outside investors' trust, and it is nearly impossible to undo afterward: giving up ten-vote shares once you already hold them almost never happens without outside pressure.

Loyalty instead of birthright: Ferrari and Berkshire each rewrite the rule differently

Ferrari chose an entirely different mechanism when it separated from the Fiat Chrysler group in 2015. Rather than creating a share class reserved by nature for founders, the company set up a loyalty voting program open, on paper, to any shareholder. The principle: whoever registers their ordinary shares in a dedicated register and holds them without interruption for three years receives, on top of their original share, a special voting share that grants a second vote. These special shares cannot be traded separately, carry almost no extra economic rights, and disappear automatically if the shareholder sells the underlying position. At the time of the 2015 separation, this mechanism gave Exor, the Agnelli family's holding company, about 33.4% of voting power, and Piero Ferrari, founder Enzo Ferrari's son, about 15.4%, with roughly 51.2% held by public shareholders as a whole.

The philosophical difference from Alphabet or Meta is real, even though the outcome, voting power concentrated among a few long standing players, looks similar on the surface. At Google and Meta, the extra power is attached to a share class that only founders and their heirs can ever hold: no new shareholder can acquire it, no matter how patient they are. At Ferrari, the mechanism rewards how long you hold, not who you are: any individual investor who buys Ferrari shares today and holds them without interruption for three years earns, in theory, the exact same voting bonus as Exor or Piero Ferrari. In one case, power passes by birth; in the other, it is earned through patience.

Berkshire Hathaway illustrates a third case, often misunderstood because it gets lumped in with the other two. Berkshire's Class B share carries one ten-thousandth (1/10,000) of a Class A share's voting power, but one fifteen-hundredth (1/1,500) of its economic value: the two ratios are deliberately different. Warren Buffett created this Class B in 1996, not to concentrate his own power, but for almost the opposite reason, a defensive one: packaged investment trusts had started reselling small, expensive slices of Berkshire A shares to the public without Buffett's consent or getting a cut himself. By creating a cheaper B share himself, he cut the ground out from under those schemes, while making sure, through a voting ratio deliberately set below the economic ratio, that decision making power stayed concentrated among long-standing Class A holders rather than diluted across a potentially much larger base of B shareholders. Unlike Alphabet or Meta, this power is not locked in for the benefit of any one individual: Buffett himself holds no separate class, only a large number of A shares, bought like any other long-term shareholder.

What it means for you, and how I read it in my own screen

This structure is far from universally accepted, and the institutional debate has already made a full U-turn. In 2017, following Snap's IPO with shares carrying zero votes at all for the public, index provider S&P Dow Jones simply banned companies with multiple share classes from entering the S&P 500 and its other indices. Google and Meta, already index members, were grandfathered in, which is why they still sit in the index today. But in April 2023, S&P Dow Jones reversed course and reopened its indices to these structures, as long as at least 5% of voting power remains with the public. In parallel, the Council of Institutional Investors, a group of large US asset managers, now pushes for a mandatory sunset clause within seven years or less of an IPO, after which all shares revert to one vote each. Some companies have already adopted this voluntarily, with different timelines: three years for EVO Payments, seven years for Smartsheet, ten years for Fastly.

The most honest argument in favor of these structures is the long horizon one: a leader shielded from being voted out at every annual meeting can afford to lose money for several years on a bet they consider strategic, like Zuckerberg on Reality Labs, without fearing a hostile takeover led by a shareholder unhappy with the current quarter. The most honest argument against it is that the same protection applies regardless of how good the bet actually is: nothing in the mechanism itself distinguishes a visionary founder from a simply stubborn one. An ordinary minority shareholder loses, in both cases, the one concrete lever they usually have: the ability to vote against the board if results deteriorate for good.

That is exactly why I never treat a multi-class share structure as an automatic flaw in my quality screen. My criteria measure a company's profitability, growth and financial discipline, not how its voting rights are distributed. But that numerical neutrality is precisely what makes the qualitative reading more important, not less: when one person or one family keeps the final word indefinitely, how they allocate capital, something I cover in my analysis of management, matters more than for a company whose board stays accountable to its full shareholder base every year. That is the same logic behind why I always look at a company's moat, its durable competitive edge, as a filter separate from who holds power, and why I applied that exact lens to Ferrari, where the brand's moat does not erase a real industrial dependency that loyalty voting does nothing to fix. My full methodology explains how I systematically separate these two questions, the quality of the business and the quality of its governance, before I even look at the price.

FAQ

Do all shares of the same listed company always carry the same vote?

No. Companies like Alphabet or Meta issue several share classes with different voting rights (one, ten, or even zero votes per share), while granting the same economic claim on profits. Ferrari instead rewards how long you hold shares rather than a fixed class.

How do I know if a stock I own carries less voting weight than another class?

Check the IPO prospectus or the company's investor relations page: it lists every share class, its ticker, and the number of votes attached. A different ticker for the same company (GOOGL versus GOOG, for example) often signals a difference in voting rights.

Is Ferrari's loyalty voting program really open to any shareholder?

Yes, on paper: you just need to register your ordinary shares in the dedicated register and hold them without interruption for three years. In practice, long standing holders like Exor and Piero Ferrari benefit the most, simply because they have held their shares the longest.

Why did Google create a share class (GOOG) with no voting rights at all?

To keep funding acquisitions and paying employees in stock without ever diluting its founders' voting power. This third class, created during a 2014 stock split, carries exactly the same economic rights as ordinary shares, but zero votes at the shareholder meeting.

About the author

Written by Lubin Danilo, founder of Lubin Investment. A self-taught individual investor, I find fundamental analysis fascinating, and it has delivered excellent results. For three years now, my performance has beaten the S&P 500. But analyzing every stock took too much time: sites with incomplete data, calculation methods and criteria never aligned with mine. And spotting the best stocks was just as time-consuming, even with my own well-defined checklist. So I put my software development background to work to build this software, base my investment strategy on its results, and share it with people who share the same passion as me. It judges a company's quality and its price separately, using criteria drawn from the financial literature (Warren Buffett, Michael Mauboussin, Aswath Damodaran).