Most startup founders never think about dividends, and for good reason: venture-backed companies reinvest everything, and the standard investor consent schedule makes paying a dividend a reserved matter anyway. But plenty of Australian tech companies do reach the point where a distribution is on the table — bootstrapped SaaS businesses throwing off cash, profitable services arms, founder-controlled companies weighing salary against dividends against Division 7A loans. When that day comes, the question of whether the company can pay a dividend is answered by section 254T of the Corporations Act 2001 (Cth) — a provision that is routinely described as a “solvency test” and routinely misunderstood.
From Profits to the Three-Limb Test
For over a century, the rule was simple to state: dividends could only be paid out of profits. The Corporations Amendment (Corporate Reporting Reform) Act 2010 replaced that profits test with effect from 28 June 2010. Since then, section 254T has prohibited a company from paying a dividend unless all three of the following are satisfied:
- The net assets limb — the company’s assets exceed its liabilities immediately before the dividend is declared, and the excess is sufficient for the payment of the dividend;
- The fair and reasonable limb — the payment is fair and reasonable to the company’s shareholders as a whole; and
- The creditor protection limb — the payment does not materially prejudice the company’s ability to pay its creditors.
Two details in the fine print matter more than the headline. First, under section 254T(2), assets and liabilities are calculated in accordance with accounting standards in force at the relevant time — even for a small proprietary company that has never otherwise had to apply them. A founder-run company that keeps tax-basis books can’t simply eyeball the bank balance; leases, employee entitlements and deferred revenue all land on the balance sheet the accounting standards way before the test is applied. Second, the test is a continuing prohibition on payment, tested immediately before declaration — so it needs current management accounts, not last June’s financials.
Why “Solvency Test” Is the Wrong Name
The 2010 change is usually described as a shift from a profits test to a solvency test. It isn’t, quite — it’s a net assets test with solvency-flavoured limbs bolted on, and the difference cuts both ways:
- A company can be cash-flow solvent but fail the test. A startup with strong recurring revenue and cash in the bank, but accumulated losses that leave liabilities exceeding assets on an accounting-standards balance sheet, cannot pay a dividend — even though it could comfortably fund one.
- Conversely, passing the net assets limb proves nothing about solvency. The creditor protection limb still has to be satisfied independently, and paying a dividend is expressly treated as incurring a debt for the purposes of the insolvent trading prohibition in section 588G — so directors who cause an insolvent (or nearly insolvent) company to pay one are personally exposed.
Treasury proposed fixing this in 2014: an exposure draft of the Corporations Legislation Amendment (Deregulatory and Other Measures) Bill would have replaced the net assets test with a pure solvency test and clarified the interaction with the capital reduction rules. The dividend provisions never made it into the legislation as ultimately passed, and have never returned. The imperfect 2010 drafting is still the law in 2026.
The Trap for Funded Startups: Capital Is Not Distributable Cash
Here’s the version of this problem that actually shows up in startup land. A company raises capital, the money sits in the bank, assets exceed liabilities — and someone asks whether the company can distribute some of it. On a naive reading of section 254T, the net assets limb is satisfied. But a “dividend” funded from share capital rather than profits collides with two other regimes:
- The capital maintenance rules. Section 254T removed the profits requirement but did not authorise distributions of capital. A payment to shareholders that is, in substance, a return of share capital must comply with the capital reduction procedure in Part 2J.1 — section 256B requires that the reduction be fair and reasonable, not materially prejudice creditors, and be approved by shareholders. The interaction between section 254T and the capital reduction rules is one of the acknowledged loose ends of the 2010 amendments, and the conservative (and near-universal) practice is to pay dividends only where there are profits or retained earnings to support them, and to use a formal capital reduction or buyback for anything else.
- Tax. The ATO’s view in Taxation Ruling TR 2012/5 is that a dividend paid in compliance with section 254T out of current or retained profits is assessable and frankable — but a distribution debited against the share capital account is generally not frankable, and capital management structures designed to dress capital up as frankable dividends attract anti-avoidance scrutiny. The corporate law test changed in 2010; the tax system, in substance, still cares about profits.
So the practical rule for founders is narrower than the statute reads: no profits, no ordinary dividend — whatever the bank balance says. If the goal is returning capital or cashing out a departing holder, the right tools are a reduction of capital or a share buyback, each with its own procedure.
Mechanics: Determine, Don’t Declare
The process points are simple but frequently botched:
- Check the constitution first. Section 254T is a floor, not the whole rule. Many constitutions — especially older ones, or ones adapted from pre-2010 precedents — still say dividends may only be paid out of profits. That restriction binds even though the statute moved on. And a shareholders’ agreement will usually make dividends subject to investor consent, while preference share terms may confer dividend rights or priorities that constrain what’s “fair and reasonable to shareholders as a whole”.
- Determine rather than declare. Under section 254U (a replaceable rule), directors may determine that a dividend is payable and fix the amount, time and method of payment. Under section 254V, the company only incurs a debt when the time fixed for payment arrives — until then the determination can be revoked. But if the constitution provides for declaring dividends, the debt arises on declaration and can’t be walked back. For a company anywhere near the solvency line, that drafting difference is the difference between an escape hatch and a locked door.
- Paper the solvency analysis. The board should minute up-to-date accounts, the section 254T analysis limb by limb, and a forward-looking cash flow supporting the creditor limb. If things later go wrong, that record is what stands between the directors and personal liability — for insolvent trading, and for breach of the duties of care and good faith that apply to every exercise of the dividend power.
The Bottom Line
Section 254T lets Australian companies pay dividends without profits in theory, but the test is a net assets test built on accounting standards, not the pure solvency test it’s often called — and the capital maintenance rules and the tax system both still punish distributions that aren’t backed by profits. For startups the working rules are: check the constitution and the consent schedule before the balance sheet; don’t treat raised capital as distributable; prefer a board determination over a declaration; and document the three limbs properly. Dividends are the rare corporate action that can create personal director liability with a single board resolution — treat them with matching care.
This article is general information only, not legal or tax advice — whether your company can and should pay a dividend depends on its accounts, its constitution and its shareholder arrangements. Viridian Lawyers advises Australian founders, startups and investors on corporate structuring, governance and distributions. If a dividend, buyback or capital return is on your board agenda, get in touch before the resolution is passed.
