Shareholders' Agreement vs Constitution: What Decides If You Don't Sign One

Do three co-founders need a shareholders' agreement, or does the constitution cover it?

No, the constitution doesn't cover it. A company's constitution is a public, structural document, and on its own it does not decide what happens if a co-founder leaves, wants to sell their shares, or two founders can't agree on something and neither will back down. Malaysian company law does have an answer for those situations, just not one written with your specific company in mind: absent a shareholders' agreement, the constitution's default provisions and statutory minority-protection rules decide transfers and disputes instead, and that default applies whoever eventually drafts an agreement, even if nobody ever does.

What the constitution actually decides by default

A company's constitution is public: it's part of the official record, and anyone can pull a copy. It's also structural: it sets out the basic mechanics of running the company, in terms similar to what every other Sdn Bhd's constitution says, rather than provisions written around your founders' specific arrangement. What it isn't built to do is arbitrate a dispute between three named individuals about who gets what. When there's no shareholders' agreement to say otherwise, the constitution's own default provisions plus the Act's statutory minority-protection rules step in and decide instead. It's a general-purpose answer, applied to a situation the founders never actually discussed.

What a shareholders' agreement adds that the constitution doesn't

Four things, mainly, and they're the ones that actually start disputes:

• Vesting: shares that become fully yours over time. Leave in month three and, without a vesting schedule, you can still walk away holding a full founder's stake.
• Transfer restrictions: who a founder is allowed to sell shares to, so nobody wakes up with a stranger as a business partner.
• Deadlock: what happens when two owners are stuck on a decision and neither side can move it forward.
• Reserved matters: decisions that need more than a simple majority, usually because they change the company in some structural way.

Most agreements also add drag-along and tag-along rights. If the majority decides to sell the company, drag-along pulls the minority into the same deal; tag-along lets the minority insist on selling on the same terms instead of being left behind. None of this sits in the constitution. It's private, and it's written around the founders' actual situation rather than the generic one the Act assumes.

The awkward part is that a clean 50/50 split between two co-founders (the arrangement that looks fairest on day one) is exactly the one with no built-in tie-breaker. Deadlock provisions exist because an even split means neither side can outvote the other, by construction, and without a clause saying what happens next, the default is that nothing moves until one founder buys the other out under whatever pressure the standoff creates.

The moments that actually force the question

Nobody sits down and drafts a shareholders' agreement purely because it seemed like a prudent abstract idea. It tends to get written (or should have been) around one of four situations:

• A co-founder wants to leave. Without vesting, they leave holding whatever stake they were granted on day one, whether they contributed two months or two years.
• A co-founder wants to sell their shares (to a co-founder, to an outsider, or back to the company) and there's no agreed process for how that happens or at what price.
• A new investor is coming in. Multiple share classes, liquidation preferences and anti-dilution terms are genuinely bespoke to what that specific investor is asking for, and none of it exists in a standard constitution.
• Two founders disagree on something structural and neither can outvote the other. That's a deadlock, and without a deadlock clause, resolving it usually means one side buying the other out under pressure, or the company simply stalling.

Most founder disputes actually happen at exactly this stage (two or three founders, ownership never documented beyond the incorporation paperwork), not once a company has grown to many shareholders and the need is obvious to everyone involved.

Does it need to be rewritten every time something changes?

No. A shareholders' agreement is designed to evolve, not to be replaced. New shareholders, a new funding round or a new governance structure get added through amendments to the agreement that already exists, not a full rewrite starting from page one. That matters more than it sounds: it means the version you sign at formation doesn't have to anticipate everything, because the mechanism for updating it is amendment, not demolition.

Isn't the constitution legally sufficient on its own?

It's sufficient in the sense that a company can operate without ever signing a shareholders' agreement. Plenty do, for years, without incident. It isn't sufficient in the sense of protecting three specific founders from the situations above. A constitution is public and covers the structural basics; a shareholders' agreement is private and covers control, exits and disputes in far more detail. Most founders don't discover the gap between the two until they actually need the second document and find out it was never written.

Why do these agreements get more expensive once investors are involved?

Because at that point the document stops being a standard structure and becomes a negotiation. Multiple share classes, liquidation preferences and anti-dilution provisions are bespoke to the specific terms a particular investor is asking for. There's no template answer to "protect my downside on a down round" the way there's a fairly standard answer to ordinary founder vesting. Two or three founders splitting equity early on follow a much more predictable structure, which is also why that earlier stage is the cheaper and faster one to get properly documented.

When to actually get one written

At formation, or before a new shareholder joins, not after a disagreement has already started, which is when the conversation stops being about what's fair and starts being about who wins. In practice it tends to happen at one of three points: at formation itself, somewhere between six and twenty-four months in once the founders have an actual working relationship worth documenting, or right before an investor, a key hire, or an employee share option pool changes who holds a stake in the company.

What a shareholders' agreement doesn't cover

It's worth being clear about the edges of the document, too. Incorporating the company is a separate step that happens before any of this, not something a shareholders' agreement does for you. Employment contracts for the people the founders go on to hire are a separate document as well, handled once there's an actual hire to write one for. And amendment cost (what it takes to update the agreement later, once the company has changed) isn't a fixed number quoted up front, because it depends on what's actually changing and how many parties are involved by then.

Getting a shareholders' agreement drafted

OCTIS drafts shareholders' agreements as a fixed, one-off document — agreed before signing, not billed by the hour — with tiers that scale from two or three founders splitting standard provisions through to investor-ready structures with multiple share classes and negotiated liquidation preferences. See the shareholder agreements service for what's covered at each stage, and what changes once an investor is in the room.