UK Court of Appeal grants claimants permission to appeal against Granville LCD cartel damages judgment

The Granville v LG cartel damages claim was one of the first few cartel claims to go to a full trial. Although the claimants were successful, they have now been granted permission to appeal on issues of overcharge and downstream pass-on, arguing that the judge made significant errors in his reasoning on overcharge and failed to apply the correct test in relation to downstream pass-on. Osborne Clarke act for the claimants. 

The claims

The claimants, who were UK computer manufacturers and sellers, claimed for losses they had suffered as the result of a price-fixing cartel relating to liquid crystal display (LCD) computer screens. In 2010, the European Commission had fined a number of manufacturers of LCD panels used in laptops and LCD monitors for colluding to artificially increase the prices of the LCD panels. The claimants' claim was therefore a "follow on" claim that was based on the findings as to the liability of the defendant manufacturers set out in the Commission's decision.

The claimants, as indirect purchasers of LCD screens, argued that they had suffered loss as a result of having purchased LCD monitors and notebooks at artificially increased prices. By the time the trial took place in October and November 2023, the only "live" defendant remaining was LG (the other remaining solvent defendants had settled with the claimants).

Commercial Court judgment

In a judgment dated 8 February 2024, the judge found in favour of the claimants and determined that the cartel had caused the claimants losses of around £4.5 million. This sum was calculated on the basis that the overcharge caused by the cartel was 8% for LCD panels used in monitors and 4% for panels used in laptops.

As to downstream pass-on, the court concluded that 65% of the overcharge had been passed on to the claimants' customers but the claim in respect of lost sales (caused by increased sale prices) significantly reduced the financial impact of the downstream pass-on on the claim (by about 50%).

Given that the judgment was one of the first few full trials and judgments in a cartel damages claim, it has been widely commented on in competition law circles. This appeal will therefore be of significant interest to competition litigation practitioners.

Appeal arguments

While the claimants were successful on all the main issues at trial and were awarded their costs, they were granted permission to appeal on the basis of arguments relating to: (i) the judge's adoption of the wrong approach to the assessment of damages, having referred to the need to "err on the side of under-compensation"; (ii) significant errors in the judge's reasoning and assessment of the expert evidence in relation to the issue of overcharge; and (iii) the judge having failed to apply the "direct and proximate causative link" test in respect of downstream pass-on before considering the likely amount of any such pass-on.

Assessment of damages

In competition cases, due to the difficulties with quantifying damages precisely, it is well established that the court applies a "broad axe" when undertaking its damages calculation. This means that when calculating the loss suffered, judges must do the best they can with the evidence (expert and factual) available. 

When applying the "broad axe" test, in Granville, the judge made repeated reference (in relation to both overcharge and pass-on) to the statement in Asda v Mastercard [2017] 4 CMLR 32 that that it was preferable to "err on the side of under-compensation".

However, that case had been held to no longer be good law, and was rejected by the Court of Appeal in Britned v ABB (CA) [2020] 4 CMLR 7. Britned clarified that the correct approach was neither to overestimate nor to underestimate when applying the broad axe. This error was drawn to the attention of the judge by the claimants at the draft judgment stage, and the judge then included a new footnote in the final judgment confirming that he had considered whether any of his conclusions as to the loss suffered would have been altered by applying the correct "broad axe" test (of giving "full compensation, no more, but certainly no less") and concluded that they would not.

Despite this, the claimants were granted permission to appeal on this ground. In granting permission, Lord Philips stated that it was "troubling" that the judge had left an incorrect statement in the body of the judgment and that it was "concerning that the footnote is not entirely coherent and may still reflect, as the claimants contend, an unbalanced approach".

Overcharge: econometric models

In respect of overcharge, the claimants' appeal focuses in particular on argument that the judge should have preferred a regression model that took into account the effect of price persistence in LCD panel prices. The claimants also argue that the judge failed to take into account the lack of robustness of defendants' regression model (which was very sensitive to reasonable adjustments to its specification) and the inherent implausibility of its overcharge estimates taking into account the known cartel price increases.

Price persistence is the effect whereby the price charged in a particular period is, at least in part, determined by the price charged in previous time periods. The arguments as to price persistence relate to the apparent contradiction that in the judgment the judge accepted the evidence of the claimants' expert that there was price persistence in LCD panel prices, but then elsewhere in the judgment concluded that there was no evidence of price persistence, and that none was referred to by the claimants' expert.

The claimants' argument was therefore that the judge had failed to take into account his own conclusions on price persistence when considering whether or not to adopt an expert model that took price persistence into account. The judge had highlighted himself that the question of price persistence was the "critical question", and whether or not an expert model including price persistence was used had a significant impact on the level of overcharge used in the damages calculation.

Regression model's 'robustness'

As to the issue of "robustness", the claimants contend that the robustness of a regression model is an important factor that must be taken into account when assessing its reliability. If the results generated by a model are very sensitive to reasonable changes to its specification, this undermines the reliability of the results.

The claimants' position is that the judge failed to take into account the evidence that the defendant's model was very sensitive to a wide range of small changes in its specification; furthermore, in almost all cases, the small changes to specification resulted in much higher overcharge estimates.

The claimants also contend that the judge failed to consider the inherent plausibility of the results of the regression models in the context of undisputed evidence as to the level of actual price increases in LCD panel prices (which amounted to approximately US$100 in the first few months of the cartel).

Downstream pass-on

As to downstream pass-on, the appeal is based on the argument that the judge did not apply the correct legal test for downstream pass-on, which requires that the defendants establish (on the balance of probabilities) a "direct and proximate causative link" between the overcharge and the prices the claimants charged to customers. Further, they contend that had the correct test been applied, there could have been no finding of downstream pass-on based on the judge's factual findings. 

The claimants argue that the judge overly relied on generalised economic theory in circumstances where the specific business models of the claimants were inconsistent with pass-on. In particular, the claimants' prices were significantly impacted by psychological price points (the need to price at £599, £999 etc), and claimants had a strong financial incentive to maintain sales volumes of PC packages irrespective of profit margin, because it gave them the opportunity to sell highly profitable warranties as a point-of-sale add-on.

Osborne Clarke comment

This case will be of general interest to all competition litigation practitioners. It is likely to provide important guidance on the correct approach to evaluating rival econometric models for estimating overcharge. Further, it is expected to provide clarity on the application of the direct and proximate causative link test for pass-on. These are two of the key issues encountered in cartel damages claims.

The Osborne Clarke team of Lisa Oakley, Siubhan McDonald and Katie Tradelius, led by partner Andrew Bartlett are acting for the claimants in this case.