BritNed's claim against ABB for losses resulting from the high voltage cable cartel is the first cartel claim to reach final judgment. The £13 million award is significantly lower than the figure that BritNed claimed, and interestingly reflects the lost efficiency, and the cost savings to ABB resulting from its participation in the cartel, rather than an intentional overcharge. The 202-page judgment of the High Court (Mr Justice Marcus Smith) provides some important pointers for other claims, and in particular for how courts are likely to assess expert and factual evidence.
BritNed, a JV between National Grid and TenneT, the operators of the GB and Dutch electricity transmission systems, owns and operates the BritNed interconnector, a 1,000 MW undersea cable connecting the GB and Dutch systems. BritNed procured the cable element of the interconnector from ABB.
In April 2014, the European Commission issued a decision in the Power Cables case, fining international manufacturers of high-voltage power cables €302 million for taking part in a cartel that shared markets and allocated customers between them, from 1999 to 2009. ABB, which would otherwise have paid a penalty of €33 million, paid nothing because it received immunity under the Commission's cartel leniency programme, as the first to come forward to provide details of the cartel. The General Court rejected a series of appeals against the decision in July 2018, although a number of the companies involved have now appealed to the ECJ. The claim against ABB was able to proceed to trial because ABB did not appeal against the decision.
BritNed claimed damages for the loss suffered, under three heads:
The judge noted that in English law, competition infringements are approached as statutory torts – claims for non-contractual liability. To succeed, a claim must show an infringement of competition law and harm or damage caused by that infringement. The fact of the damage must be established on the balance of probabilities (the English standard of proof in civil cases), whereas the measure of the loss takes into account a wide range of risks and possibilities. The fact that it may not be possible for a claimant to prove the exact loss is not an obstacle to recovery – the court may take a "broad brush" approach and adopt various estimates, assumptions and approximations.
The fact of the infringement was established by the Commission decision. The judge concluded that the overcharge that he should assess was the difference between the price agreed between BritNed and ABB and the price that would have been agreed – whether with ABB or another supplier – in the absence of the cartel. The price that a third party would have offered would be very difficult to determine.
An interesting aspect of the judgment – typical in English judgments where there may be lengthy examination of witnesses, but less familiar to those in civil jurisdictions - was the judge's assessment of the different types of evidence before the court. They included factual witness evidence, ie. the evidence of 7 individuals who gave evidence before the court on the facts, documentary evidence, the Commission decision and the evidence of expert witnesses – in this case an expert economist appointed by each party.
A key witness was Hans-Åke Jönsson, ABB's General Manager of its high voltage cables business at the time of the infringement. He was the only witness who was aware of the cartel while it was in operation. The judge recognised that Mr Jönsson was involved in the cartel, and this inevitably affected his testimony. The judge therefore treated his evidence with a relatively high degree of caution. Each witness was assessed for reliability in this way, with different degrees of reliability being attributed to each of them. Other ABB witnesses were assessed as being reliable, and their evidence particularly influenced the judge's conclusion that ABB had costed the project competitively.
The judge acknowledged that he was bound by the findings in the operative part of the Commission's decision (the single article finding an infringement, naming the participating undertakings and setting out the penalties imposed). He also concluded that he was bound by those of the 1078 recitals of the decision that constituted part of the essential basis of the operative part, ie. that set out the key findings underlying the operative article. However, the judge adopted a cautious approach to the recitals that did not constitute part of the essential basis of the operative part. He was prepared to accept a non-binding statement by the Commission only "where it seem[ed] to [him] that it [was] a finding [he could] properly make on the evidence as a whole".
Inevitably the evidence had gaps, which the judge bridged with his "broad brush". However, he rejected BritNed's invitation to draw adverse inferences from the absence of material.
The EU competition damages Directive provides that cartels are to be presumed to cause harm. Although BritNed accepted that the Directive did not apply in this case (because the facts pre-dated the date of entry into force of the Directive), it argued that the principle of effectiveness required that the judge should presume harm. The judge rejected this suggestion too. He observed that he would analyse the evidence before him. If the evidence was strong, there would be no need for a presumption, but if it was weak, he saw no reason why a presumption should bolster a weak case. It seems unlikely that, had any presumption been applied, it would have resulted in a different outcome.
Operation of the cartel
To address a perceived excess of capacity among cable suppliers, members of the cartel sought to maintain price levels and allocate bids. Asian (Japanese and Korean) manufacturers would not compete in Europe, and European manufacturers would not compete in Asia. Within Europe, there was further allocation. The parties held meetings to allocate projects. The cartel also included "compensation" mechanisms:
"Thus, if one member of the Cartel forwent a particular opportunity to bid (either by not bidding at all or by putting in an uncompetitive bid), that member would in due course receive “compensation” (generally in the form of being the favoured bidder in another project). For example:
ABB was not involved in the cartel from the start, but was seen as an important member, although it did not take part in the European/Asian meetings. In contrast, it was aware of the allocation of projects within Europe, and it was clear to Mr Jönsson that his role included continuing the cooperation between the manufacturers.
The BritNed tender
BritNed sought prices for 3 lots, each consisting of one engineering, procurement and construction (EPC) contract: 1 each for the cable and converter stations and 1 for the cable and converter stations together. BritNed received expressions of interest from 6 manufacturers. However, after Nexans and Prysmian failed to submit compliant bids for the cable element of the project (which the judge found was a result of the cartel), BritNed was left with no competitor to ABB in either a combined package or a lot consisting of just the cable contract. BritNed used a number of strategies to seek to optimise its position in the negotiations, with some degree of success.
A key question was whether those involved in the bid within ABB knew that it was the only bidder for the cable-only and combined lots. The judge found that certain officers within ABB (Mr Jönsson and another) knew of the cartel and were involved in the bid. They knew that while competition for the cable could not be ruled out, Nexans and Prysmian had accepted the allocation of BritNed to ABB, and were therefore very unlikely to bid. However, other individuals within ABB were not aware of the cartel and of the allocation of BritNed to ABB. Because the detailed pricing of the cartel was in the hands of these unaware individuals, Mr Jönsson and the other involved executive could not influence the direct costs. However, they could influence the common costs allocated to the project, and had the potential to influence the tender indirectly.
BritNed's economic expert compared the price of projects during the cartel with the price of projects after the cartel. The judgment explains clearly and thoroughly the economic and statistical analysis used, also drawing on the European Commission's staff working document on quantifying harm. BritNed's expert estimated the overcharge at 25.4%. In contrast, ABB's expert compared the price of BritNed alone with the price of other projects after the cartel, focusing in particular on ABB's costs of supply. He found no overcharge.
The judge preferred ABB's approach to BritNed's. He noted that the approach adopted by BritNed's expert was significantly more complicated than that of ABB's expert, and therefore inherently more prone to error. It relied heavily on proxies rather than actual figures, because of concerns about the reliability of ABB's costs. However, Mr Justice Marcus Smith was prepared to accept that ABB's direct costs – on which its expert's assessment was based – were reliable and capable of being used as the basis for the analysis. The judge also had other concerns about BritNed's model, such that it was appropriate to disregard it entirely.
The judge first set out an interesting assessment of the type of loss that may be recovered in a cartel damages claim. A cartel does not, by definition, involve a single participant – it is the collective failure to compete that is the wrong targeted by Article 101. This collusion prevents, restricts or distorts competition. To require a claimant to show monetary harm is to impose too great a burden. Instead, the claimant must show that the defendant's conduct in some way restricted or reduced the level of the claimant's consumer benefit, in other words that the claimant has suffered as a result of the prevention, restriction or distortion of competition. This might take the form of an increased price, but it might also take the form of a reduction in the number of suppliers properly participating in a tender process. The judge found that this was what had happened in BritNed's case – but for the cartel, BritNed would have been presented with a different commercial environment, with different tenderers tendering on different terms. The judge concluded that this cause of action had been made out.
This is an interesting approach. The judge adopted a rather wider view of the types of harm caused by cartels than the traditionally accepted direct overcharge or increase in the headline price of the product or service concerned. Instead, his approach appears to accept that other types of harm (such as a loss of efficiency, as in this case, as will be seen below) should be treated as a consequence for which damages could be recovered.
Although the judge concluded that some individuals within ABB knew of the cartel and knew that ABB would face limited competition when tendering for BritNed, that knowledge did not translate into a direct influence on the direct costs of the project, which were compiled honestly and competently with a view to putting together a competitive bid. While those involved in the cartel had the opportunity to influence the common costs, in fact critical pricing decisions were taken by an individual who, the judge was satisfied, was not aware of the cartel and who priced the project competitively. However, the judge concluded (at paragraph 449) that:
"449. …the effect of the Cartel was to insulate ABB from inefficiencies in its own product. Had there been a properly competitive environment, ABB would have faced technical solutions from others involving less copper and perhaps less insulation. As a result, one of two things might have occurred:
450. I find that there was an overcharge to BritNed arising out of this baked-in inefficiency. As I have noted, this is not something that would be picked up by [ABB's expert's] margin analysis, nor either of his other two methods of assessment. This is because the technical issue was one that would not have ended with the cessation of ABB’s participation in the Cartel. It would have persisted into the post-Cartel period. This finding is confirmed by the fact that, post-Cartel, most of these projects were lost by ABB."
The judge considered that, applying a broad brush, this overcharge corresponded to the cost of the additional copper that ABB would have absorbed in order to retain the bid. The appropriate measure of this overcharge was 15% of the cost of the copper. This amounted to some £7.5 million. The judge then found a further overcharge in the form of a saving in ABB's costs resulting from the cartel, amounting to a further £5.5 million.
The judge also rejected BritNed's claim for lost profits – its claim that but for the cartel, it would have procured a 1,320 MW interconnector that would have delivered higher profits. He concluded that in this counterfactual world, BritNed would have procured a 1,000 MW cable as it did in the real world.
ABB argued that any loss suffered by BritNed had been reduced or eliminated by a regulatory cap on BritNed's rate of return, such that BritNed would not be permitted to retain any profit above the cap.
The cap finds its source in Regulation 1228/2003, Article 6(6) of which requires revenues resulting from the allocation of interconnector capacity to be used for guaranteeing the availability of capacity or maintaining or increasing capacity, or as income to be taken into account in calculating network tariffs. New interconnectors may apply for an exemption from this requirement (Regulation 1228/2003 has since been replaced by Regulation 714/2009, which contains similar provisions). BritNed applied for exemption, and received a 25-year exemption from the requirement in 2007. The European Commission, which has the power to review national exemption decisions, requested the addition of a condition under which, if the internal rate of return over the first 25 years was more than 1% above the internal rate of return estimated in the exemption application, BritNed was required either to invest the surplus in increasing interconnector capacity, or to apply it towards its parents' national regulated asset bases (ie. in reduction of network tariffs). ABB argued that this meant that any overcharge would cause BritNed no loss, because by increasing the project costs it would also increase the level at which profits would exceed the cap and would be paid out. However, the judge disagreed: there were public policy reasons why ABB should not be permitted to retain the overcharge.
BritNed claimed compound interest on the basis that it had incurred higher capital costs than it would have done in a competitive market. ABB countered that this loss was not BritNed's but that of its parents, who had provided the funding. The judge accepted ABB's argument, but awarded simple interest to BritNed.
A striking feature of the judgment is the judge's very thorough weighing of both economic and factual evidence. Particularly evident is the judge's very careful assessment of the knowledge and motivation of the individuals involved in the pricing decisions. The mere fact that certain individuals at ABB were aware of the cartel did not, in this instance, automatically mean that ABB's prices at that time incorporated an overcharge. Instead, the fact that the individual with primary responsibility for pricing the project credibly testified that he had sought to reflect ABB's direct costs competitively was highly persuasive. This granular approach is clearly very different from the broad brush view of the European Commission, which is not required to consider the effect of the overall collusion on individual projects. The Commission is also not usually concerned to differentiate between the degree of knowledge of individuals within cartel members, focusing instead on the undertaking as a whole. The judge's careful appraisal of the two very different economic models presented by the experts is also noteworthy. He set out a detailed critique of each one, before concluding that he preferred ABB's cost-based approach over BritNed's reliance on proxies and regression analysis. To the extent that future claims that reach trial adopt a similar approach, claimants and defendants alike will need to take a very detailed look at all of the evidence – factual and economic.
Clearly a claim in respect of a project such as this – a large individually-negotiated project – is very different from a claim in respect of quantities of homogeneous products purchased over a period of time, where proxies and statistical models may be more appropriate. In such cases, the outcome may be very different.
The judgment offers encouragement to both claimants and defendants, and will be perused very carefully in future claims. Whether it is the last word in this particular dispute remains to be seen, and other claims in respect of the same cartel are still pending.
Jun 01 2023