Has the Court of Appeal misapplied the per se provisions of the Commerce Act?
January 23, 2019
Last year, the Court of Appeal released its decision in the Hamilton real estate agencies cartel case, Commerce Commission v Lodge Real Estate Limited. The agencies held a meeting in response to a dramatic increase in Trade Me advertising costs, and as a result each of the agencies elected not to self-fund those costs. In practice, this would mean that the costs would often be passed individual house sellers.
Usually when competitors (or even non-competitors) make an arrangement among themselves that the Commission takes exception to, the Commission needs to prove to the Court that the arrangement has some sort of anticompetitive effect. The Lodge case gives rise to an interesting point of competition law – when is alleged ‘cartel’ conduct so bad that the Commission doesn’t have to prove that it actually lessened competition. In other words, when does the law simply assume that there is an anticompetitive effect?
The Commerce Act is actually pretty clear on this. It says, basically, that if the arrangement impacts on the prices the competitors set then the anticompetitive effect is assumed. In the trade we call this per se liability. The policy justification is – again, over-simplified – that it is not worth wasting everyone’s time proving the anticompetitive effect because it’s so obvious. Competitors only make arrangements about their prices to keep those prices high, obviously.
But the Lodge case didn’t involve an arrangement about prices. The alleged arrangement (the agencies disputed that there even was an arrangement in the first place) dealt with a cost item – Trade Me listing fees. The Commission argued that co-ordinated treatment of an input cost had occurred, which required the parties to the arrangement to pass that cost on to house sellers directly. This would, of course, inevitably increase the price of each real estate agency’s services.
Here things get a bit tricky. The broad thrust of the Commission’s argument makes sense as a matter of law and economic theory. Costs are recovered through prices somehow, so an arrangement about an input cost must affect prices, and so per se liability follows. However, this line of logic depends on precisely what was discussed and agreed by the real estate agencies. That is, the facts matter, and they might undo this neat little theory.
That’s what happened in the High Court. In that forum, the Judge was astute enough to ask what was actually agreed among the parties. The agencies had in fact agreed (to the extent there was any agreement) to move from self-funding the Trade Me fee to what they called a ‘vendor funding’ model. That’s an unhelpful label, because its sounds like – and the Commission seems to have assumed – that the real estate agencies would simply invoice the sellers directly for the full costs of the Trade Me fees. But when the Judge asked what the term ‘vendor funding’ actually meant, his Honour discovered something interesting. All that term meant in the real estate industry was that the treatment of the Trade Me fees would be negotiated with each individual seller on each transaction. All the term really meant was that self-funding was no longer the default option.
Given this rather specific meaning of ‘vendor funding’, the High Court found that there could be no impact on prices. The agencies might have equally ‘agreed’ that funding of Trade Me listing fees would be determined by the market on a case by case basis, or simply that there would be no standard practice in respect of Trade Me listing fees for future transactions. As a result, the High Court determined in its judgment that “[o]n any individual transaction in the supply of real estate sales services or real estate advertising services, the full range of price setting options remained”. There was no effect on prices, so there could be no per se liability.
I think the High Court got this completely right, both by choosing to engage in close factual analysis and, on the basis of that analysis, making the only credible legal finding available to it. I said as much in an earlier post at the time the High Court judgment came out. However, I was in the minority. I have since chatted with several competition lawyers and economists who believe the High Court got it wrong.
Those lawyers and economists would have been very pleased with the Court of Appeal judgment, as it overruled the High Court on precisely this point. Having read the Court of Appeal judgment on this point quite carefully, I stand by my original assessment. In fact, I think the Court of Appeal has got this wrong and its judgment on this point will be vulnerable if there is an appeal. Here’s a brief overview of some of the key reasons why.
First, the Court of Appeal showed no interest in what the factual evidence said about what was actually agreed, and whether that agreement might impact on prices (as required by the Commerce Act for per se liability). It felt it sufficient to claim that any agreement relating to input costs would “plainly” and “obviously” impact on prices. This reluctance to engage in factual analysis may stem from the fact that per se liability is intended to avoid close factual analysis in most circumstances – that’s why it’s called ‘per se’. But this legal ‘short cut’ only applies to analysis of anti-competitive market effects. The Commerce Act specifically requires an inquiry into pricing effects. Not undertaking that analysis is potentially a misapplication of the relevant law.
Second, the little analysis the Court of Appeal does engage in misunderstands the findings of fact in the High Court. The Court of Appeal frames the High Court’s factual findings as to the nature of the agreement as involving a default practice of having sellers fund Trade Me fees, with some residual discretion applying in specific cases. If this was what the evidence said had happened, then I agree that per se liability is squarely on the cards. But, as set out above, there was no default practice established by the agreement. The High Court in fact found that the agreement was of a kind where there could be no pricing impacts at all. The Court of Appeal seems to have overlooked because it did not purport to contradict the High Court's factual finding, just the legal consequences of that finding. Once this error of fact is corrected the Court of Appeal’s argument that it can simply assume per se liability falls away.
Third, the Court of Appeal justified its reluctance to inquire into the actual impact on prices with reference to overseas case law, which it said adopted the same approach. But it misinterpreted this case law. For example, the Court of Appeal cited this passage from a United States case, Plymouth Dealers’ Association:
The test is not what the actual effect is on prices, but whether such agreements interfere with “the freedom of traders and thereby restrain their ability to sell in accordance with their own judgment” […] The competition between the Plymouth dealers and the fact that the dealers use the fixed uniform list price in most instances only as a starting point, is of no consequence. It was an agreed starting point; it had been agreed upon between competitors; it was in some instances in the record respected and followed; it had to do with, and had its effect upon, price.
On its face, this passage looks like it establishes a test for determining whether a finding of per se liability can or should be made, and that's what the Court of Appeal thought it meant. But when this passage is read in context, it is actually a description of the legal implications of a finding of per se liability. It does not establish a legal test. In fact, there was clear factual evidence before the court in Plymouth Dealers’ Association of actual price effects, and this factual evidence
(not the 'test' in the above passage) is what motivated the liability finding. As a result, the case cannot support the Court of Appeal’s approach of just assuming that per se liability applies.
There is much more to say about the Court of Appeal’s judgment, but there already seems to be enough here to suggest that the Court got the wrong result because it adopted an incorrect methodology – one that relied on assumption rather than factual analysis of the available evidence. Word about town is that there is an appeal to the Supreme Court in the works. If that eventuates, we can only hope that the Supreme Court endorses the High Court’s careful assessment of the actual facts regardless of the ultimate legal result it reaches.