The ACCC’s loss of another case concerning section 46, the abuse of dominance provision, may well add to the pressure for the Competition Policy Review to deliver a final recommendation in March 2015 consistent with its earlier draft recommendation that section 46 should be amended to introduce an “effects” test, subject to a new defence if the conduct has a rational business purpose and is in the interests of consumers. Consultant, George Raitt discusses the possible impact of the proposed test.

The Federal Court in its recent decision essentially concluded that Pfizer had substantial market power, until its patents expired, and took advantage of that market power in implementing direct to pharmacy distribution arrangements that bundled its patented product Lipitor with its generic atorvastatin, which was launched into the market prior to patent expiry. However, the Court concluded that Pfizer did so for the purpose of meeting competition, and not a proscribed purpose of preventing or deterring competition in a relevant market. Nothing, of course, could prevent Pfizer’s patent expiring, and the Court concluded that suppliers of generics would not likely be deterred from entering the market post-patent expiry.

The intuition underlying the Court’s decision is that once Pfizer’s patent expired there was nothing Pfizer could do that would create an insurmountable “barrier to entry” and accordingly the process of competition should be left to take its course. Generic suppliers have their own distribution arrangements, which encourage pharmacists to promote generics, with which Pfizer must compete. Pfizer’s estimates of the impact of patent expiry on sales of its products in Australia in the 5 years after patent expiry indicated that sales would reduce from about $1M pa to $200K pa (worst case) or with the direct to pharmacy distribution arrangements would reduce to $500K pa (best case).

The Court accepted the ACCC’s contention that, for the purposes of assessing “market power”, the market is “atorvastatin”. The Court’s reasons seem slightly confused but it may be inferred from the repeated statement “a pharmacist presented with a prescription for atorvastatin was required to supply atorvastatin” is intended to suggest that generics did not form part of the market for the purpose of assessing market power. Given the Court’s finding on “purpose”, it became unnecessary to solve the conundrum that market power pre-patent expiry existed in relation to Lipitor and post-patent expiry the relevant market would be generic atorvastatin. This conundrum suggests that, by definition, the required nexus between the exercise of market power and harm to competitors in a relevant market could not be demonstrated.

The decided cases indicate the “purpose” is subjective and is to be distinguished from “effects” (though of course a person is taken to intend the natural and probable consequences of their actions, so purpose and effect are in some ways linked).

While section 46 does not currently contain an “effects” test, the ACCC also argued the case under section 47, which does have an “effects” test. Curiously, however, the ACCC did not contend that the conduct had an anti-competitive effect. We might reasonably conclude, therefore, that the introduction of an “effects” test into section 46 would not have altered the outcome of the case on the present facts. Perhaps, however, the ACCC might say it would have argued “effects” if section 46 had contained an “effects” test. Perhaps the ACCC felt that it has a better chance forensically of proving “purpose” and may have sought to avoid gathering evidence on market effects. The possible application of the proposal by the Competition Policy Review is, nevertheless, considered below.

The Pfizer case may be contrasted with the recent Cement Australia case, where the ACCC failed in its section 46 case but succeeded under section 45 on the basis that Cement Australia’s purposes included a substantial purpose of excluding competitors. 

It is noteworthy that section 45 has an “effects” test, but the Court considered that any anti-competitive effect of Cement Australia’s exclusive supply contract was dissipated by market factors. Unlike many areas of law where “causation” is a wellestablished requirement, the “effects” test that appears in section 45 refers to “effects or likely effects”. It has been held that a “likely effect” is one which has “real chance or possibility” of occurring. That is, the test is prospective, and must be decided without the benefit of waiting to see what happens. Although the trial in Pfizer occurred during 2014, it would not have been necessary for the purposes of an “effects” test to enquire what actual effects may have occurred in the market after expiry of Pfizer’s patent in May 2012. Nevertheless, there was some evidence before the Court of customers switching between generics post-patent expiry.

In the Pfizer case, the Court recognised the problem of “dual purposes”, in that any action to secure sales of Pfizer’s products at one and the same time harms competitors, who are thereby foreclosed from making the same sales. Section 4F provides that one anti-competitive purpose among others will suffice, provided it is a substantial purpose. However, the Court concluded on the facts, having regard to the evidence and credibility of Pfizer’s witness, that Pfizer did not have a substantial purpose of preventing or deterring competition. The Court was satisfied that Pfizer’s overwhelming purpose was to ensure the survival of its generic product post-patent expiry. Thus the Court in Pfizer neatly side-stepped the “dual purposes” issue that the Court in Cement Australia resolved in the ACCC’s favour.

The Competition Policy Review draft report adopts much of the ACCC’s 2014 submission that section 46 should be changed to prohibit conduct that has “the purpose or likely effect of substantially lessening competition” with, however, the addition of the following: the accused corporation would have a defence if it proves that the conduct in question would be rational for a corporation that did not have substantial market power and the conduct would be likely to have the effect of advancing the long-term interests of consumers.

The first element of the defence, “rational decision”, re-opens the question of the hypothetical standard by which conduct is assessed under the current “taking advantage” requirement, i.e. is the conduct possible in a hypothetical competitive market in which market power is absent?

The second element of the proposed defence has been widely criticised: economists might interpret the long term interests of consumers as “economic efficiency” but this is unclear, given the ACCC wears a “consumer protection” hat, and it is possible that the defendant might have to prove that long-term prices to consumers will be minimised as a result of the conduct in question.

It remains to be seen what final changes the Review may recommend to section 46. Hypothetically applying the draft recommendations to the facts and findings of the Pfizer case, we are faced first with the fact that the ACCC did not allege any anti-competitive effect, despite the case being put additionally under section 47. Can we infer that the ACCC did not consider there to be any such effect (i.e. that Pfizer may have merely “attempted” to lessen competition)?

Or can we infer that, as apparently anticipated by Pfizer and reflected in the Court record, generics would be likely to reduce Pfizer’s sales in the shortterm to anywhere between 20%-50% of pre-patent expiry levels, and that there would be active reprice competition and consumer switching between generics? 

There does not appear to be clarity around forensic methods of establishing competition effects. Nevertheless, there is strong intuitive appeal for a conclusion that, post-patent expiry, the market for atorvastatin was at least workably competitive despite the distribution arrangements of Pfizer (and its competitors).

If the ACCC were able to demonstrate an actual or likely anti-competitive effect, Pfizer would bear the burden of establishing the proposed defence. It could be concluded that Pfizer would likely satisfy the “rational decision” defence but that the “long-term interests of consumers” element is imponderable, absent facts concerning the relative efficiency of Pfizer and other generic manufacturers post-patent expiry. The reverse onus of proof is offensive to general notions of justice and fairness given that the matters which must be proved by the defendant are hypothetical and virtually incapable of definitive proof. It is clear, however, that the forensic application of section 46 and the proposed defence, if the proposed changes are adopted, will make section 46 cases vastly more complex navel-gazing exercises than they are at present.

It may be useful to further test the facts of Cement Australia and Pfizer against a possible approach to section 46 using a possible prohibition of market manipulation. In Cement Australia, it seems reasonable to infer that the ACCC considered that the defendant was attempting to corner the market for fly ash in order to become the sole supplier in the region of concrete grade fly ash, so forcing up the costs of its competitors (in concrete markets) gaining access to that raw material. The evidence however is focussed on the defendant’s ability to price without constraint, not the actual effect on prices and supplies to competitors, or prices and competition in the concrete markets. The evidence was clear that collection and processing of raw fly ash involves capital investment and risk. Adopting a “market manipulation” approach, we might therefore ask: 1. Did the defendant restrict supply to competitors? 2. Was price commensurate with costs of production and normal return on capital invested? 3. If price is thought to be above a competitive level, is this caused by restriction of supply? 4. If the defendant has obtained a monopoly profit, is this due to competitive activity (e.g. by competitive tendering to remove power stations’ waste fly ash)?

In Pfizer, it seems reasonable to infer that the ACCC considered that the defendant was attempting to extend its patent monopoly by locking in distribution arrangements. This seems inherently unlikely to succeed. Adopting a “market manipulation” approach, it seems obvious that the defendant, the “incumbent monopolist”, did not attempt to restrict supply, and prices were subject to competitive forces. It seems that incumbent and new entrants will compete on price, for volume, and the outcome will be determined by “survival of the fittest”. The Court’s judgment seems intuitively sound – the law should not intervene.