16 Dec 2011

(by Bruce Lyons) On 16th April 2010, the OFT imposed a record total fine of £225m on two tobacco companies and ten retailers who they decided had engaged in practices that breached competition law. The heart of the case was around ‘parity and differential requirements’ (PDR) of the form that, say, Imperial Tobacco requires that the retailer should not price Imperial’s brand A more than 3 pence higher than Gallagher’s brand B. Rather loosely, I call this ‘price matching’ for ease of exposition but it should be taken to include ‘restoring price differentials’. Earlier this week, the Competition Appeals Tribunalin a decision that will see the OFT repaying the fine and paying huge costs. Even more importantly, what does it mean for the enforcement of antitrust policy in the UK?

Was the decision right?

This was not a straightforward hard core horizontal price-fixing cartel, though the OFT fine was at a level that suggests an equivalent offence. The theory of harm was essentially that a set of unilateral vertical agreements had the object of restricting competition by removing the incentive for a retailer to cut retail price in response to a wholesale price cut by one manufacturer. The key feature of a PDR is that it has a direct horizontal effect even though it is not an agreement between horizontal firms. Even if only one manufacturer has a PDR, both are disincentivised to compete at the wholesale level. A wholesale price cut that does not move retail price differentials would not reap the reward of a higher market share. The theory of harm seems quite plausible. It has sound theoretical roots and no efficiency defence is mentioned in the appeal. So why, given the visible smoke of PDRs, was the OFT unable to prove to the CAT that there was a fire burning a hole in the pockets of consumers?

The problem was the evidence. The OFT relied mainly on written PDR agreements, signed by manufacturer and retailer, which rewarded the compliant retailer with bonuses or discounts. However, the CAT listened to a number of cross-examined witnesses provided by Imperial and the retailers, none of whom said that these agreements raised prices at the retail level. The inability of the OFT to find a single business witness (indeed, even the original whistleblower from Sainsbury’s) who supported their precise theory of harm was decisive to the CAT, who threw out the case.

The OFT appears to have made three significant errors. It should have obtained signed witness statements supporting its interpretation of PDRs before reaching its original decision. It should have been satisfied with an effects based argument and not bet everything on the object of the agreements being to raise price – even if that approach would have resulted in a less headline grabbing fine. And it could have been much cleverer in wording its decision more broadly.

Unfortunately, the CAT did not reach a decision on whether tobacco PDRs lead to high prices, or whether similar practices in other markets could yet be found anticompetitive. The CAT followed the law and precedent, but this led them to throw out the OFT’s original arguments as inconsistent with the evidence they presented. They found the gap too great to be recovered in the level of review provided by the law (which is not a full review of the merits of PDRs).

Wider implications

So, yes, the CAT was probably right to quash the case given the OFT’s phrasing of the decision and the evidence it provided. However, this does give cause for a non-lawyer to reflect on the system that led to this. I offer the following thoughts.

  1. The OFT is required to adopt a balanced investigative (administrative) approach right through to the decision stage. The OFT investigates, decides and punishes. There was no clear separation of roles within the OFT and probably no effective challenge to an incomplete case. It needs a genuinely independent decision panel.
  2. To protect the parties subject to this monolithic process, the appeals system is not pure judicial review but has an element of a merits appeal. I say an element because it is not a full re-hearing of the evidence with a view to an independent decision, but there is a merits review only of the evidence relevant to the narrow grounds of appeal. This is important because it means that if the OFT does not get its specified theory of harm exactly right, there is limited flexibility for the CAT to consider a modified theory of harm. In this case, the OFT wanted to argue that even if the PDR did not fully require the retailer to match prices, it had the effect of the retailer following the manufacturer’s instructions or requests to price match. This is a fairly subtle change, but the CAT thought the whole line of questioning of the witnesses would have been different under this theory of harm. It did not see fit to go through all the hearings again.
  3. The OFT started its original investigation as an impartial inquisitor and ended it in an uncertain position between inquisitor and prosecutor before the CAT. At the CAT, it had one QC lined up against six appellants supported by no less than seven QCs (including two for Imperial Tobacco). Each witness provided by Imperial was cross-examined by the OFT and a retailer, and each witness provided by a retailer was cross-examined by the OFT and Imperial. My legal colleagues assure me that this imbalance would have no effect if the OFT has a good case. I remain more open minded.
  4. The appellants also had seven expert economists lined up against two for the OFT, only one of whom dealt with the theory of harm. The CAT threw them into a proverbial hot tub so they could come back with the extent of their agreement and disagreement. It does not seem they were impressed by the outcome. When the OFT suggested that the experts would not find it difficult to review a slightly revised theory of harm, the CAT’s response was: ‘We regard this as a triumph of hope over experience, given the number of experts involved in these appeals and the absence of much common ground between them so far.’ [#92]

This last point is a depressing confirmation of how a battle of the experts before a tribunal (which has no similar experts of its own) can lead to a cancelling out of the economic evidence as far as the court is concerned. The economic effects of business practices, which should be central to an antitrust appraisal, face the danger of becoming a side issue.

[My wider views in relation to the reform of antitrust enforcement can be found in an and in the to the BIS consultation.]