Medical rosters and the Trade Practices Act

Warren Pengilley
Med J Aust 2003; 178 (7): 337-340. || doi: 10.5694/j.1326-5377.2003.tb05226.x
Published online: 7 April 2003


  • Medical rosters are not free of trade practices problems, notwithstanding assurances by the Australian Competition and Consumer Commission (ACCC).

  • Neither the ACCC nor the recently convened Wilkinson Committee has applied rigorous legal principles in interpreting the Trade Practices Act 1974 (Cwlth) to reach its conclusions.

  • The Australian law should be changed to bring it into line with that of the United States and New Zealand.

The Australian Competition and Consumer Commission (ACCC) has condemned the Australian Medical Association's view that medical rosters may breach the Trade Practices Act 1974 (Cwlth) as a collective boycott.

I examine here whether roster systems are illegal under the Trade Practices Act as collective boycotts and aim to answer the simple question "Is the ACCC right in what it says about roster arrangements?". I do not discuss price-fixing arrangements, which are illegal under quite different provisions of the Trade Practices Act to those covering collective boycotts.


The Australian Medical Association (AMA) and the ACCC have come to different conclusions regarding the legality of roster agreements.

On 29 August 2001, the Prime Minister announced an Inquiry to review the impact of the Trade Practices Act on the recruitment and retention of medical practitioners in rural and regional Australia. The Wilkinson Committee's report became publicly available on 10 November 2002.1 The Inquiry took place against a background of public assertion by the ACCC that, in relation to the legality of rosters, the AMA and other groups continued to perpetuate "alarming and misleading comments".2

The ACCC is on record to the effect that:

The ACCC has not targeted ... roster arrangements in the past and will not in the future.3

The Wilkinson Committee had put to it that the ACCC could prosecute doctors for participating in a roster but chose not to do so. This view was stated in the following words:

It is not sufficient for the ACCC to say they will 'look the other way'. This is akin to a policeman saying he will look the other way when you do 100 km per hour in a 60 km per hour zone — it is still illegal, and he could change his mind at any time. (p. 64)1

The Wilkinson Committee concluded that there was no question of the ACCC choosing to ignore illegal conduct. The ACCC itself denied that it was a question of "turning a blind eye",2 stating that a roster developed between doctors for the purpose of ensuring availability of medical services or to provide appropriate breaks for doctors does not breach the Trade Practices Act and therefore is not illegal.1,2 The Committee, citing the views of the ACCC,4 reached the conclusion that:

... doctors may collaborate in devising rosters but may not collude in devising rosters. An agreement that was set up for an anticompetitive purpose rather than to ensure the availability of medical services and to provide appropriate breaks for practitioners would be an unlawful roster, not a genuine one. Such a roster could fall foul of the boycott provisions of the Act and could lead to an investigation by the ACCC. It follows from this that doctors may collaborate together if the effect of that collaboration is to provide arrangements that will serve their patients better. Any form of collusion in that process which may result in restricting care options, would be a clear breach of the Act. (p. 65; original emphasis)1

Several issues arise from the above discussion:

  • Is the ACCC correct in what it says? It is not a question of "turning a blind eye", but of interpreting the law. The ACCC has to be in a position to guarantee legally that rosters are not in breach of the Trade Practices Act.

  • Has the Wilkinson Committee got it right? Although the Wilkinson Committee, by and large, accepts the reasoning of the ACCC, there are some separate points which the Committee makes. Again, it is not a question of "turning a blind eye".

  • Is ACCC authorisation a satisfactory solution?

The legislation

The law that makes collective boycotts illegal is section 4D of the Trade Practices Act 1974 (Cwlth). Section 4D(1), in summary, provides, relevantly to rosters, as follows:

  • a provision of contract, arrangement or understanding entered into between competitors is illegal if

  • the provision has the purpose of preventing, restricting or limiting the supply of services to particular persons or classes of persons ("persons" includes any entity, and a hospital comes within this term)

  • by all or any of the parties to the contract, arrangement or understanding.

Virtually every roster is an arrangement between competitors (and thus fulfils criterion (1)). It also fulfils criterion (3). The debate centres around criterion (2). A crucial word in criterion (2) is "purpose". The difficulty involved in this word is the nature of the purpose involved and, in particular, whether the limitation or restriction on supply has to be the sole purpose of the arrangement. Specifically, section 4F of the Act provides that the purpose test is satisfied if the relevant limitation purpose is a purpose and such purpose is a substantial purpose. The Act thus recognises that there may be more than one purpose in conduct. Therefore, if a restriction on the supply of medical services is a substantial purpose, the Trade Practices Act is infringed, even though there may be one or more other purposes of the arrangement in question.

An infringement of section 4D is what is called in trade practices law a "per se" offence. In the present context, this means that any arrangement with the substantial purpose of limiting the supply of services by medical practitioners to a hospital is illegal. It is irrelevant that the arrangement is not anticompetitive.5 The market power position that parties may hold is likewise irrelevant. The arrangement is also illegal even if it is beneficial in effect (unless an authorisation on public benefit grounds is obtained from the ACCC). Further, it is the immediate purpose of the arrangement which is relevant, and not its long term "object".

Is the ACCC correct in what it says?
What the ACCC said to the Wilkinson Committee and why it is wrong

The ACCC, in its submission to the Wilkinson Committee, argued that the AMA's legal advice is wrong because it states that anticompetitive purpose is not a relevant element in section 4D. The ACCC says that "anticompetitive purpose (i.e. of preventing, restricting or limiting the supply of service) is central to the conduct prohibited by s.4D" (p. 22).6

The ACCC is guilty of some sleight of hand here. All that has to be shown for an infringement of section 4D to occur is that there is an agreed limitation of services.

Unilateral decisions made by doctors are, of course, legal. But the mathematical chances of doctors adequately covering a 24-hour period seven days a week without some arrangement between them must be infinitesimal. Necessarily fundamental to any roster system is that Doctor X and Doctor Y agree that Doctor X will work between certain hours. Necessarily, this agreement also involves the obverse (ie, that Doctor Y will not work those hours). Hence Doctors X and Y have made an agreement for which a substantial purpose is just as much to restrict or limit the provision of services by Doctor Y as it is to secure the services of Doctor X in his or her rostered hours. Indeed, no roster could function unless Doctor Y agreed not to provide services when Doctor X is rostered on duty. Rosters inherently involve an understanding that some will be on and some will be off.

An overall anticompetitive purpose is not relevant, and the ACCC is in error in asserting this to be an ingredient of a section 4D infringement. The point is subject to a specific Federal Court decision in the Kim Hughes case.5 In that case, the Western Australian Cricket Association (WACA) banned Kim Hughes from club cricket because he had led a "rebel" cricket team to South Africa, in defiance of the Australian ban on team sporting contacts with that country because of its apartheid policy. The Federal Court specifically held that this conduct was not anticompetitive in purpose. However, the WACA's conduct was illegal, because it came within the statutory provisions of section 4D and had the purpose of limiting or restricting the supply of services (club cricket) to Kim Hughes.

The ACCC's sleight of hand is that it regards anticompetitive purpose and the purpose of restricting or limiting services as the same. The Kim Hughes case shows the two are not the same. One fundamental difference between the two is that, to prove anticompetitive purpose, a purpose to affect competition in the market as a whole must be shown. In contrast, a restriction on the supply of services can be quite specifically targeted at an individual entity.

Is it all about "turning a blind eye"?

Of course, if everyone is happy there are no complaints. The ACCC would hardly be justified in spending resources to take action against a medical roster which delivered a benefit (a factor irrelevant to whether there is an infringement of section 4D) in providing an after-hours service. The ACCC has established enforcement guidelines, and such a case would not come within the guidelines.

But the ACCC has said that this is not its stance. Rosters, it asserts, simply cannot come within section 4D. This proposition cannot stand.

Hospital rosters will not normally be matters of great concern to anyone, assuming that the rosters adequately cater for hospital needs. This may not necessarily be the case. Consider two medical practitioners in a rural town who regard themselves as overworked and have long pressed the town's hospital to employ a resident doctor. As a method of forcing the issue, one doctor agrees with the other to service the hospital for only eight hours per day. The other doctor agrees to service the hospital only for a further eight hours per day. They both decline service to the hospital for the remaining eight hours per day. This would clearly give rise to antagonism and possible action by the hospital against the medical practitioners under section 4D. The ACCC, in blessing medical rosters, is not looking at such a case, although it is in such cases that section 4D action is most likely to eventuate.

Risk-taking assessments

In assessing risks, the AMA and its spokespersons must take into account that the ACCC does not make the law and is not the only enforcer of the law. In the above scenario, it may be the hospital which takes proceedings, or, if the issue becomes political enough, the Minister for Health. There are plenty of lawyers who would advise that there is, in this scenario, a highly arguable section 4D case against the doctors involved in the roster.

Whether the ACCC would proceed in the above circumstances if pressed politically or otherwise to do so is a matter on which views will differ. However, the risk is that the conduct is illegal, even if its enforcement is dormant, and the ACCC could change its mind at any time.

Has the Wilkinson Committee got it right?

The Committee's conclusions cannot be accepted as an accurate synopsis of the law. Section 4D talks about "contracts, arrangements or understandings". Both "collaboration" and "collusion" come within this definition, and there is no legal basis for believing, as the Wilkinson Committee does, that one is legal and the other is not. The difference between the two terms is semantic, not legal. Further, the test in section 4D is not whether better patient care is provided. The test is whether there is a limitation on services. Likewise, the test of section 4D is not whether there is an anticompetitive purpose, but whether there is a purpose to restrict services. The two are quite different concepts. One can restrict services to individuals (and thus breach section 4D) but not be anticompetitive in the market as a whole, as the Kim Hughes case quite specifically concludes.

Therefore, I conclude that the Wilkinson Committee's view of the law is not correct for at least the above reasons. Nor is it appropriate to accept the ACCC's view that the Wilkinson Committee's conclusions should be accepted as those of the umpire.7 The Wilkinson Committee is not the umpire. The courts are the only umpires on legal matters.

I should make it clear that I make no comment here about the Wilkinson Committee report generally. No doubt the Committee has sought out relevant submissions and carefully evaluated the issues. However, there is a clear difference between policy recommendations and conclusions as to the interpretation of the law. The Wilkinson Committee may be correct in its policy recommendations. For example, it may be the case that the application of the Trade Practices Act is not impacting on the recruitment and retention of rural doctors. My observations here address only the issues of legal interpretation.

Is ACCC authorisation a satisfactory solution?

The ACCC has power to authorise conduct on public benefit grounds. On 9 May 2002, the Treasurer established a committee, chaired by former High Court Justice Sir Daryl Dawson, to review the administration and impact of the Trade Practices Act. In a submission to the Dawson review, the ACCC stated that:

The Commission considers that the Authorisation process effectively balances the need for a process that is flexible and responsive, broadly accessible, fair to all parties, efficient and timely, gives business certainty and has an appropriate framework for accountability.8

That view is certainly open to debate. But no matter how the ACCC administers the authorisation process, authorisation should not be needed for conduct that is not anticompetitive. A simple amendment to the law could ensure this result.

Authorisation is not a satisfactory alternative to an amendment to the law because:

  • it necessarily involves unwarranted expense (both filing fees and professional advice);

  • there are delays in obtaining authorisation;

  • the ACCC may grant authorisation for a limited time, thus making repeated applications necessary; and

  • the ACCC may use authorisation as a vehicle to impose conditions on conduct. In the case of conduct that is not anticompetitive, the authorisation process should not be able to be used for this end.

The ACCC sees authorisation as a solution to the concerns of medical practitioners. Nevertheless, ACCC authorisation is a very imperfect solution to a problem caused by a drafting deficiency in section 4D and a non-appreciation of what should, and what should not, be banned per se under the Trade Practices Act.

Where the law is deficient and where it should be amended

The problem of medical rosters is but an illustration of the general overkill of section 4D. An amendment to section 4D to make it a more sensible tool of competition policy would solve the difficulty of the legality of medical rosters and mean that the law applies equally to all. The Australian deficiencies are shown by reference to United States and New Zealand experience.

US law makes collective boycotts illegal per se, but differs from the Australian law in two basic respects.

  • The US law only bans per se a boycott by competitors of other competitors, actual or potential. Thus, the US law could only prohibit per se the boycotting by medical practitioners of other medical practitioners (eg, in relation to a group of doctors denying hospital access to a newly qualified medical practitioner when adequate hospital facilities are available). US law could not ban hospital rosters per se, as the target of the collective conduct is not a competitor of those engaging in such conduct (a hospital is not a "competitor", actual or potential, of medical practitioners). The Kim Hughes case would not be an illegal collective boycott in the US, as the target of the boycott (Kim Hughes, a cricketer) was not a competitor of those engaging in the boycott (the various Western Australian cricket clubs).

  • In assessing whether the per se test is applicable, the US courts give the arrangement an initial "quick look". The conduct is banned per se only if:

. . . an observer with even a rudimentary understanding of economics would conclude that the arrangements would have an anticompetitive effect on customers and markets.9

In Australia, the 1976 Report of the Swanson Committee,10 upon whose recommendations section 4D was enacted, accurately conceptualised at least the first of the above points in recommending a per se ban on collective boycott activity in Australia. However, the parliamentary draftsperson erred, and we did not enact what the Swanson Committee recommended. Had section 4D been enacted in accordance with the Swanson Committee recommendations, medical roster arrangements would not be banned per se under the Trade Practices Act.

New Zealand initially copied the Australian Act, including section 4D. However, in 1990, it changed its legislation to provide, as in the United States, that the target of a boycott had to be a competitor, actual or potential, of those engaging in the boycotting activity. In 2001, the New Zealand Commerce Act was amended to provide that parties to a collective boycott would not be in breach if, on a reverse-onus basis, they could demonstrate that the relevant activity did not have the purpose, effect or likely effect of substantially lessening competition.

The result of these differences is that medical rosters in New Zealand and the United States are untroubled by competition law.

There is no policy reason why we should retain section 4D in its present form. An amendment to bring the section in line with United States and New Zealand law would be a complete solution to any trade practices roster problems. Such an amendment would also solve the problems of many others caught by section 4D overkill. The potential illegality of a joint venture between competitors, in which they each agree not to sell their jointly produced product except on agreed terms and agree not to compete against the joint venture, is but one example of the more general problem.


At the time of writing (December 2002), the High Court has not yet given its decision in the South Sydney Rugby League Football Club appeal,11 which is the first case before the High Court involving section 4D. What is said in this case may significantly affect present opinion in relation to section 4D. Nevertheless, at present, medical practitioners do have a trade practices problem in relation to roster arrangements. The statements of the ACCC and the findings of the Wilkinson Committee do not send this problem away. Authorisation is a procedure that should be unnecessary and would be unnecessary if our trade practices law were amended.

The AMA, quite reasonably, regarded its problem as being unique to doctors and argued its own case. In doing so, it gave the appearance of seeking special status under the Act. In this respect, the AMA's campaign was bound to fail. The ACCC could fend off the AMA's campaign on the basis that it was a self-interested "campaign against the Trade Practices Act".12

To say that the AMA's campaign was bound to fail does not mean that it was not, in principle, correctly directed. When it is realised that the AMA's situation is far from unique, the "self interest" aspect of the campaign changes. The AMA can quite properly join others in pressing for amendments to the Trade Practices Act to bring Australian law into line with that of New Zealand and the United States. Asking for an amendment to this end is, after all, only asking that an Australian drafting error now 25 years old be corrected, and that our trading laws be brought into line with those of two of our major trading partners, whose commitment to competition law principles is certainly no less than our own. Perhaps the recommendations of the Dawson Committee will provide much-needed amendment to the Trade Practices Act.

  • Warren Pengilley

  • School of Law, University of Newcastle, Newcastle, NSW.


Competing interests:

None identified.

  • 1. Inquiry into the impact of Part IV of the Trade Practices Act 1974 on the recruitment and retention of medical practitioners in rural and regional Australia (Warwick Wilkinson AM, Chairman). 10 November 2002. Available at: (accessed Mar 2003).
  • 2. Rural Doctors Association of Queensland unfortunately joins AMA in misleading medical practitioners and the public about genuine medical rosters and the T.P.A. ACCC Media Release 15/02. 29 January 2002. Available at: (accessed Mar 2003).
  • 3. Medical rosters do not breach Act: ACCC. ACCC Media Release 67/01. 30 March 2001. Available at: (accessed Mar 2003).
  • 4. General practitioners: a guide to the Trade Practices Act. Revised Draft March 2001. Canberra: ACCC, 2001. Available at: (accessed Mar 2003).
  • 5. Hughes v Western Australian Cricket Association Inc (1986) ATPR ¶ 40-736.
  • 6. Australian Competition and Consumer Commission. Submission to the Inquiry into the impact of Part IV of the Trade Practices Act 1974 on the retention and recruitment of medical practitioners in rural and regional Australia. 29 November 2001. Available at: (accessed Mar 2003).
  • 7. Prime Minister issues independent review report which backs ACCC views on doctors. ACCC Media Release MR 276/02. 10 November 2002. Available at: (accessed Mar 2003).
  • 8. Submission to the Trade Practices Act review (Dawson review). June 2002. Canberra: ACCC, 2002; 245. Available at: (accessed Mar 2003).
  • 9. California Dental Association v FTC 1999-1 Trade Cases 72529 (US Supreme Court).
  • 10. Trade Practices Review Committee (T B Swanson, Chairman). Report to the Minister for Business and Consumer Affairs. August 1976.
  • 11. South Sydney District Rugby League Football Club Ltd v News Ltd [2001] ATPR 41-824 (Federal Court). Appeal to the High Court argued on 6 Aug 2002.
  • 12. AMA 'wrong' on genuine rosters: GP's misled on breach of law. ACCC Media Release 42/01. 6 March 2001. Available at: (accessed Mar 2003).


remove_circle_outline Delete Author
add_circle_outline Add Author

Do you have any competing interests to declare? *

I/we agree to assign copyright to the Medical Journal of Australia and agree to the Conditions of publication *
I/we agree to the Terms of use of the Medical Journal of Australia *
Email me when people comment on this article

Online responses are no longer available. Please refer to our instructions for authors page for more information.