Restraint of trade clauses and what makes them enforceable

A restraint of trade clause is a promise in an employment contract that limits what an employee can do after the job ends. The best known version is the non-compete clause, which stops a former employee from working for a competitor or starting a rival business for a set period. There are also non-solicitation clauses, which stop an employee from taking clients or co-workers, and confidentiality clauses, which protect information gained during employment.

The clause exists to solve a real problem for a business: a key employee walks out the door taking client relationships, inside knowledge and sometimes staff with them. The restraint is the employer's attempt to keep those assets in place for long enough that the business can recover. But the law treats these clauses with suspicion, because they also stop a person from earning a living. That tension, between an employer's legitimate interests and an employee's freedom to work, is what the whole area of law turns on.

This article explains how restraint of trade clauses actually operate in Australia: the starting presumption against them, who has to prove what, the factors a court weighs, how enforcement plays out in practice, and the federal reform that will soon change the landscape for many workers.

The starting point: restraints are void unless reasonable

Australian law inherited a common law rule that is over a century old: a restraint of trade is prima facie void because it is contrary to public policy. It is only enforceable to the extent that it is reasonable in the interests of the parties and in the interests of the public. In other words, the signed contract is not the end of the story. An employer cannot simply point to the clause and demand compliance. The clause only has legal effect if a court is persuaded it is reasonable, and that assessment is made against the particular facts, not in the abstract.

The High Court confirmed the shape of the test in Amoco Australia Pty Ltd v Rocca Bros Motor Engineering Co Pty Ltd [1973] HCA 40, adopting the classic formulation that the onus of establishing that an agreement is reasonable as between the parties lies on the person who puts the agreement forward, while the onus of establishing that it is contrary to the public interest lies on the person alleging that. Translated into an employment context, that means two things. First, the employer must prove the clause protects a legitimate business interest and goes no further than reasonably necessary. Second, if the employer succeeds, the employee still has a chance to show that enforcing the clause would be against the public interest.

There is no one size fits all answer to whether a restraint holds up. The same clause that is enforceable against a sales director who held the key client relationships may be void against a junior administrator. Everything depends on the role, the market and the wording of the clause itself.

The two-stage burden of proof

Because the clause starts out void, the burden of proof is deliberately stacked against the employer who wants to enforce it. The employer must establish two things at the first stage: that the clause protects a legitimate business interest, and that the clause is reasonable in scope between the parties. This is a real evidentiary burden, not a formality. The employer has to point to something specific that the employee could take and use against the business.

Only if the employer clears that first hurdle does the second stage arise, where the employee can argue that even a reasonable restraint is contrary to the public interest. That argument usually relies on broader concerns, such as the employee's capacity to earn a living in their trade or the public benefit in competition. In practice most cases are decided at the first stage, because courts are reluctant to enforce restraints that go beyond what the employer actually needs.

What counts as a legitimate business interest

The first question a court asks is whether the employer is protecting something the law recognises as worth protecting. Three categories are well established in employment cases:

• Confidential information and trade secrets: client lists, pricing structures, product plans, marketing strategies and other information that is genuinely secret and valuable, as opposed to general know-how the employee picked up along the way.
• Customer connection and goodwill: where the employee built personal relationships with clients, so that their departure creates a real risk those clients will follow them to a competitor.
• Workforce stability: protecting against the poaching of co-workers who are essential to the business.

The category of interest determines what the clause can legitimately restrain. If the concern is confidential information, a non-compete may be hard to justify when a confidentiality clause would do the job. If the concern is customer connection, a non-solicitation clause targeted at the clients the employee actually dealt with will usually be easier to defend than a blanket non-compete.

What the law does not protect is competition for its own sake. An employer cannot restrain an employee merely to avoid rivalry, and it cannot use a restraint to stop an employee from using the general skills and experience they developed during their employment. Those belong to the employee. Factors like the seniority of the role, the level of remuneration, and whether the employee was the public face of the business are relevant because they indicate how much real threat the employee poses to the protected interests. A senior figure who personally serviced major clients poses a very different risk from a back-office worker who never met a customer.

The reasonableness test: duration, area and scope

Even where a legitimate interest exists, the clause must be no wider than is reasonably necessary to protect it. Courts assess reasonableness along three dimensions: the duration of the restraint, the geographic area it covers, and the scope of activities it restrains. A restraint that is too long, too wide or too broad in what it prohibits is unlikely to be enforced at all, because a court will not rewrite a bad clause into a good one.

The recent decision in DXC Eclipse Pty Ltd v Wildsmith [2023] NSWCA 98 shows how strictly courts apply this. In a business sale context, a non-compete restraint of seven years with an extremely broad, near-international geographic area was held unreasonable in both scope and duration, even though the purchaser had paid for goodwill. The case also confirms that an acknowledgement in the contract that the restraints are reasonable cannot rescue a clause that is objectively too wide.

Because reasonableness is so fact-specific, drafters often use cascading or ladder clauses. Instead of one restraint, the contract contains a series: a non-compete for 3, 6 and 12 months; a geographic restriction covering a suburb, a city, a state and the whole country; each expressed to apply only if the wider one is invalid. That way, if the court strikes down the broadest version, a narrower one can still survive.

There is also an important difference between the states. New South Wales is the only state with legislation dealing directly with the validity of restraints. Section 4(1) of the Restraints of Trade Act 1976 (NSW) provides that a restraint of trade is valid to the extent to which it is not against public policy, whether it is in severable terms or not. That lets a NSW court read a clause down, keeping the part that is reasonable rather than throwing the whole clause out. In the other states the common law applies, and the usual approach is that an unreasonable part of a covenant can be severed, but the court will not rewrite what remains.

How enforcement actually plays out

Enforcement is triggered by a departure. An employee resigns and joins a competitor, sets up their own business, or starts contacting clients and staff. The employer then has a series of escalating options, and each step is governed by its own timing pressures.

The first step is usually a demand letter. The employer puts the former employee on notice of the clause, identifies the conduct it says is a breach, and demands that it stop. This is often enough to resolve matters, particularly where the employee has legal advice and can see the clause may be enforceable.

If the employee does not back down, the employer's real remedy is an application to the Supreme Court for an interlocutory injunction. This is urgent work. Injunction applications are often brought within days of discovering the breach, because every week the former employee spends competing at a new employer causes further damage. The court decides whether there is a serious question to be tried about the enforceability of the clause and where the balance of convenience lies, meaning which side would suffer more from the court getting the interim decision wrong. At this stage the court is not finally deciding the clause's validity, but in practice the injunction application is where most restraint disputes are won or lost, because the employee who is stopped early rarely comes back to fight the full case.

To succeed, the employer must be able to point to evidence of a real risk of detriment, not mere speculation. The evidence that commonly features in these applications includes an employee emailing client lists to a personal address before resigning, copying client files or internal documents, soliciting clients to move their business to a competitor, and disclosing or misusing confidential information. The stronger this kind of documentary evidence, the harder it is for the employee to argue the restraint is unnecessary.

Damages: proving the loss

An injunction stops conduct, but damages compensate for loss. If the employer has not sought an injunction, or the breach has already happened and finished, the employer may claim damages for breach of contract. The catch is that damages must be proved. A court will not award damages unless the employer can show the breach caused a loss, and that loss does not have to be purely monetary. It can include lost or threatened revenue from an existing client base, or the erosion of confidential information such as future designs and marketing plans.

In practice, this is why injunctions dominate restraint of trade litigation. Loss is often genuinely hard to quantify, particularly where the damage is to relationships and reputation rather than revenue. An employer who waits until the damage is done may find itself with an enforceable clause and nothing it can prove it lost. That is a strong reason to act quickly when a key employee leaves.

The reform horizon: non-competes from 2027

The law described above is about to change for many workers. In the 2025-26 federal Budget, the government announced reforms to worker restraints, and the Treasury consultation on the reforms ran from 25 July to 5 September 2025. The centrepiece is a proposed ban on non-compete clauses for low- and middle-income workers, alongside bans on no-poach agreements between businesses and wage-fixing agreements. The reforms are expected to take effect from 2027, after legislation passes parliament.

The consultation also asked whether reform is needed for non-compete clauses for high-income workers, for non-solicitation clauses, and for restrictions on employees holding multiple jobs, so the final shape of the changes is not yet settled. The government's non-compete clauses page sets out the evidence behind the push. An e61 Institute survey found one in five Australian workers have a non-compete clause, including lower-wage workers in childcare, labouring and clerical roles, and the ABS found one in five businesses use them. Treasury also cites research finding workers at businesses with widespread use of non-compete clauses have lower job mobility, and that similar businesses without them pay about 4 per cent more on average.

The practical point for a business owner is that this reform does not remove the need to think about restraints. Non-solicitation clauses, confidentiality clauses and no-poaching of staff remain lawful and will continue to do most of the work of protecting a business after a reform passes. What is targeted is the blunt instrument of the non-compete for lower-paid workers, and that means the enforceability of any particular clause will depend even more on the worker's role and pay.

Getting the assessment right: where a lawyer helps

There are three moments when a restraint of trade clause deserves professional attention, and the cost of getting each of them wrong is different.

The first is before the clause is signed. For an employee about to accept a role, or an employer drafting a contract, the scope of the restraint should be negotiated while there is still goodwill between the parties. A clause that is calibrated to the role, with sensible cascading limits, is cheaper to agree to than to fight over later.

The second is on the way out. An employee who has resigned to join a competitor should have the clause reviewed before acting, not after receiving a demand letter. An employer who has lost a key employee should get advice within days, because the evidence that supports an injunction, such as copied files and client communications, needs to be preserved early, and delay can be fatal to an urgent application.

The third is in the dispute itself. Whether you are the employer seeking an injunction or the former employee resisting one, the assessment of whether the clause is reasonable is a legal question that turns on evidence about the role, the market and the interests at stake. A lawyer can tell you where the balance of convenience is likely to fall before you spend money on litigation.

The one question that matters most is whether the clause is calibrated to something real. A boilerplate restraint that bans everything for a year across the whole country is the least likely clause in the contract to survive contact with a court. A clause tied to the specific clients, information and staff the employee could actually take is the one that will be enforced. Get that assessment made early, and you will know whether the piece of paper is a real protection or just a comfort blanket.