Elon Musk owns roughly 42% of SpaceX’s equity. But when the company’s S-1 goes public this month, he’ll control 79% of every shareholder vote. The gap between those two numbers is the most expensive lesson in corporate governance that most founders never learn until it’s too late.
SpaceX filed its confidential S-1 with the SEC on April 1, 2026, setting up what’s expected to be the largest IPO in history at a $1.75 trillion valuation. The public prospectus is expected between May 18 and May 22. When retail investors start buying Class A shares at roughly $525 apiece, they’ll get an economic stake in a rocket company. What they won’t get is any meaningful say in how it’s run.
A dual-class share structure is a corporate setup where a company issues two or more classes of stock with different voting rights, allowing founders to retain majority voting control even when they own a minority of total equity. SpaceX’s version gives Musk Class B shares worth 10 votes each while public investors get Class A shares worth one vote each. The S-1 states that Musk “can only be removed from our board or these positions by the vote of Class B holders,” which means the people buying shares on Nasdaq won’t have the power to fire him. Ever.
This isn’t a quirk of SpaceX. It’s the playbook that built the most valuable tech companies in the world. And whether you’re raising a seed round or prepping for a Series C, the decisions you make about your cap table right now determine whether you’ll have this option later.
Last updated: May 2026
Quick answers
What is a dual-class share structure?
A dual-class share structure is a corporate setup where a company creates two classes of stock with different voting weights. Founders typically hold Class B shares worth 10 or more votes each, while public investors get Class A shares worth one vote. This lets founders control board decisions and company strategy even after they’ve sold most of their economic stake to outside investors.
How does Elon Musk control SpaceX with a minority stake?
Musk holds Class B shares carrying 10 votes per share, giving him roughly 79% of total voting power despite owning about 42% of SpaceX’s equity. The company’s S-1 filing specifies that only Class B holders can remove him from the board, meaning public shareholders buying Class A shares at IPO have no governance power over his position.
Can early-stage founders set up dual-class shares?
Yes. The groundwork for dual-class structures is typically laid years before an IPO, often at the Series A stage. Founders negotiate super-voting provisions, anti-dilution protections, and board composition terms into their earliest financing agreements. The cap table decisions made during seed and Series A rounds determine whether this option exists later.
How dual-class shares actually work
The mechanics are straightforward. A company creates two classes of common stock. Class A shares get one vote per share. Class B shares get 10, 20, or in some cases even more votes per share. Founders and early insiders hold Class B. Everyone else, including public investors, institutional funds, and retail buyers, gets Class A.
The math is simple. If a founder owns 15% of total shares but those shares carry 10 votes each, they can hold majority voting power over investors who collectively own 85% of the company’s economics. Venture capital investors have complained about this imbalance for decades, but the most successful founders keep using it because it works.
SpaceX’s structure is a textbook implementation. Musk’s Class B shares carry 10 votes each. At 42% equity ownership, those 10x-weighted votes translate to 79% of all shareholder votes. Public investors buying Class A shares at the IPO will own a piece of the company’s financial upside. They won’t own a piece of its decision-making.
This isn’t theoretical. The S-1 language is explicit: Musk can only be removed by Class B holders. That clause means SpaceX’s board answers to one person. Period.
Which founders have used this playbook?
Musk didn’t invent this. He copied the best.
Mark Zuckerberg controls 61% of Meta’s voting power while holding just 13% of the company’s total shares, according to Meta’s April 2026 proxy filing. He owns 99.7% of all outstanding Class B stock, and each B share carries 10 votes. That means every proposal presented to Meta shareholders passes or fails based on what Zuckerberg decides. He is the sole deciding vote, regardless of what BlackRock, Vanguard, or any other institutional investor wants.
Google’s co-founders pulled it off in 2004, more than two decades ago. Larry Page and Sergey Brin together control 52.7% of Alphabet’s voting power through Class B shares, per the company’s April 2026 proxy statement. Their economic ownership? Single digits. Page holds 5.8% of total outstanding shares. Brin holds 5.4%. They added a third class of stock (Class C, ticker GOOG) in 2014 that carries zero votes, letting them issue shares for acquisitions and compensation without diluting their grip.
Snap took the most aggressive approach anyone’s tried at IPO. When Evan Spiegel and Bobby Murphy brought Snap public in March 2017, they sold Class A shares carrying zero voting rights. Not reduced voting rights. Zero. Public investors got an economic ticket and nothing else. The co-founders held Class C shares worth 10 votes each, giving them 88.5% of voting power. It was the first time a U.S. company had IPO’d with non-voting public shares since the NYSE banned the practice in 1940.
| Company | Founder(s) | Equity owned | Voting control | Vote ratio |
|---|---|---|---|---|
| SpaceX | Elon Musk | ~42% | ~79% | 10:1 |
| Meta | Mark Zuckerberg | ~13% | ~61% | 10:1 |
| Alphabet | Page & Brin | ~11% combined | ~52.7% | 10:1 |
| Snap | Spiegel & Murphy | ~45% combined | ~88.5% | 10:1 (public = 0 votes) |
What happens when founder control goes wrong?
The SpaceX S-1 looks like a founder’s dream. WeWork’s S-1 in 2019 looked the same way. It became a case study in why unlimited founder power can destroy a company.
Adam Neumann held shares carrying 20 votes each at WeWork, giving him majority control over every corporate decision. He used that power to lease buildings he personally owned back to WeWork, trademarked the word “We” and charged the company $5.9 million for the rights, and installed his wife as chief brand and impact officer with the authority to appoint his successor. When WeWork filed its S-1 in August 2019, institutional investors read it and recoiled.
The company cut Neumann’s super-voting shares from 20:1 to 10:1 in September 2019, trying to salvage the IPO. It wasn’t enough. WeWork pulled its S-1 on September 17, and Neumann resigned as CEO the next day. SoftBank eventually paid him $1.7 billion to leave, an exit package that The Washington Post called “a lesson in giving founders too much control.”
The difference between Musk and Neumann isn’t the structure. It’s the execution. Both used dual-class shares to maintain control. Musk built SpaceX into a company that generated $16 billion in revenue and landed reusable rockets. Neumann couldn’t show a path to profitability. Dual-class shares amplify whatever the founder does with the power. Good judgment becomes unstoppable momentum. Bad judgment becomes an unaccountable disaster.
Theranos sits in a similar cautionary category. Elizabeth Holmes maintained tight control over her board and governance structure throughout the company’s life. She handpicked directors, resisted outside oversight, and kept key decisions behind closed doors. The structure didn’t cause the fraud, but it removed every check that might have caught it earlier. When the lies unraveled, there was no governance mechanism in place to hold her accountable before the SEC and DOJ stepped in. For founders evaluating dual-class structures, WeWork and Theranos are the same lesson told two different ways: concentrated power without transparent execution creates the conditions for catastrophic failure.
How does SpaceX’s IPO compare to Saudi Aramco’s record?
SpaceX is targeting a $75 billion raise at a $1.75 trillion valuation, which would make it the largest IPO in history by a wide margin. Saudi Aramco raised $29.4 billion at a $1.7 trillion valuation in December 2019. SpaceX’s raise would be roughly 2.5 times that amount.
The most unusual part of SpaceX’s approach is the retail allocation. CFO James McNeil stated in January 2026 that SpaceX is committing to 30% retail allocation, approximately $22.5 billion of the $75 billion total. Standard IPOs reserve 10-15% for retail investors. SpaceX is offering three times the typical share, which McNeil described as making retail “a critical part of this” offering.
Bank of America, Citigroup, Goldman Sachs, JPMorgan, and Morgan Stanley are underwriting the deal. The roadshow is expected to begin the week of June 8, with a Nasdaq debut targeted for June 12 under the ticker SPCX. The implied share price sits around $525 based on $1.75 trillion divided across approximately 3.32 billion fully diluted shares.
The governance question matters more for SpaceX than for most IPOs because the company also merged with xAI, Musk’s artificial intelligence venture. That merger means public investors aren’t just buying into a rocket company. They’re buying into a conglomerate spanning space launch, satellite internet (Starlink), and AI research, all under a single founder’s control. The complexity of that business mix, combined with the governance structure, has drawn criticism from pension funds and institutional investors who argue that dual-class shares at this scale represent a concentration of power without precedent in the public markets.
Can early-stage founders plan for this?
Yes. And they should start earlier than most realize.
Dual-class structures aren’t created at IPO. They’re negotiated into the company’s formation documents, often at the Series A stage. The decisions founders make during their earliest fundraising rounds set the constraints for every option that comes later. Once you’ve given up board control or allowed investors to block a dual-class conversion, the door closes permanently.
Three things matter at the early stage. First, incorporate with the flexibility to create multiple share classes later. Delaware C-corps make this straightforward, and most startup lawyers will include provisions for it in the initial certificate of incorporation if asked. Second, negotiate anti-dilution protections and maintain enough ownership that a 10x voting multiplier gives you majority control even after multiple dilutive rounds. If you own 15% at IPO and your shares carry 10 votes, you’ll control roughly 60% of votes. If you’ve been diluted to 8%, even 10x votes only gets you to about 45%. Not enough.
Third, control board composition from day one. Founders who lose their board majority early rarely get it back. Every investor director seat you concede is a vote you can’t override with super-voting shares because board decisions operate separately from shareholder votes. Page and Brin kept three Google board seats at IPO. Zuckerberg maintained majority board control at Meta’s IPO. That wasn’t luck. It was planned before their first outside check cleared.
What are sunset provisions and why do they matter?
Sunset provisions are the guardrails that limit how long dual-class structures last. They’re the price founders pay to get institutional investors comfortable with the power imbalance.
The most common trigger is time. Some companies include automatic expiration of the super-voting class after 7, 10, or 20 years. Google’s Class B shares don’t have a time-based sunset, which is part of why Page and Brin still control Alphabet more than 20 years after its IPO. SpaceX’s S-1 similarly includes no automatic expiration for Musk’s voting power.
Other triggers include the founder leaving the company, selling below a certain ownership threshold, or transferring shares to an entity they don’t control. These event-based sunsets give investors confidence that if the founder walks away, the power structure normalizes. According to Harvard Law’s corporate governance research, the trend is moving toward mandatory sunset provisions, with more exchanges considering time-based limits for new dual-class listings. A separate Harvard study found that in almost 60% of firms that go public, the founder is no longer CEO at IPO, and half of those who are CEO at IPO lose the role within three years.
For founders, the negotiation over sunsets is where the real power negotiation plays out. Every VC will ask for sunsets. The stronger your company’s traction and competitive position, the weaker the sunsets you’ll have to accept. Snap had effectively no investor protections. WeWork’s absence of sunsets was one of the red flags that spooked public markets.
What does this mean for founders raising money right now?
SpaceX’s S-1 is a live tutorial. Here’s what it actually teaches.
The cap table you build at your earliest funding rounds determines whether you’ll have the option to go public on your own terms. Musk didn’t decide to hold 79% of votes the week before SpaceX filed. That outcome was engineered across 22 years of financing decisions, starting with SpaceX’s founding in 2002.
Three tactical takeaways for founders in 2026:
1. Ask your startup lawyer about dual-class provisions before your first priced round. It costs nothing to include the flexibility in your certificate of incorporation. It costs everything to add it later when investors have blocking rights. Most founders don’t think about governance until Series C or later. By then, the available options have narrowed.
2. Guard board seats more carefully than equity percentages. Super-voting shares control shareholder votes. Board seats control operational decisions, executive compensation, and strategic direction. Losing the board while holding voting shares creates a split-brain problem where you can block shareholder proposals but can’t direct management. Travis Kalanick learned this at Uber when he retained significant equity but lost board control, leading to his forced resignation in 2017.
3. Build a company that justifies the structure. Zuckerberg’s dual-class works because Meta generates $40 billion in annual net income. Neumann’s didn’t because WeWork couldn’t show a path to profitability. Investors tolerate asymmetric governance when the founder delivers results. The structure doesn’t protect you from failure. It protects successful execution from short-term market pressure.
There’s a common objection worth addressing: VCs will push back on dual-class provisions, especially at the seed and Series A stages when founders have the least negotiating power. That’s true. But the founders who ended up with dual-class structures at IPO didn’t win the argument all at once. They established the framework early and reinforced it at each subsequent round as their companies gained traction and the power dynamic shifted. Page and Brin included dual-class provisions in Google’s original incorporation documents in 1998, six years before the IPO. Zuckerberg negotiated his voting control during Facebook’s early financing rounds, then strengthened it over time.
The founders who lose this option are the ones who don’t think about governance until their lawyers bring it up before the IPO roadshow. By then, the board is already set, the share classes are already fixed, and the conversion mechanisms either exist or they don’t. You can’t retrofit dual-class shares onto a single-class company without every existing shareholder agreeing. That rarely happens.
