WHEN the House of Lords decided OBG Ltd. v Allan [2007] UKHL 21, [2008] A.C. 1, some much-needed clarity was finally brought to the economic torts. Or so we thought. Within a year, in Revenue and Customs Commissioners v Total Network SL [2008] UKHL 19, [2008] 1 A.C. 1174, a differently constituted panel muddied the waters again (in several well-documented ways). But one aspect of the OBG decision concerning the unlawful means tort seemingly escaped unscathed: Lord Hoffmann's so-called “dealing requirement”. According to this, D's use of unlawful means against a third party with a view to causing loss to C will not enliven this tort unless those unlawful means also interfere with the third party's freedom to deal with C (OBG, at [57]).
This qualification to what counts as “unlawful means” attracted much academic criticism in the wake of OBG. But it remained unchallenged in the courts until Secretary of State for Health v Servier Laboratories Ltd. [2021] UKSC 24 [2021] 3 W.L.R. 370. There, the defendants had allegedly deceitfully obtained a patent in respect of a drug called Perindopril from the European Patent Office (“EPO”). The drug did not possess the required degree of novelty, but the defendants falsely claimed that it did. And sticking to this untruth, the defendants secured from the English courts injunctions preventing other pharmaceutical companies from marketing drugs of the same generic type. The claimants alleged that these orders, too, had been deceitfully obtained. Five years later, the truth was discovered and the patent was revoked. During the interim, however, the defendants had enjoyed a monopoly and charged the claimants a much higher price for Perindopril than could have been charged had other companies been allowed to enter the market. Alleging a commission of the unlawful means tort, the claimants – who had bought vast quantities of the drug – sought damages in excess £200 million.
Whether the patent had been obtained and enforced by means of deceits perpetrated against the EPO and English courts was not in issue. The defendants argued instead that the claim must fail since the claimants could not satisfy the dealing requirement. The claimants acknowledged an absence of dealings between themselves and the EPO/English courts, but nonetheless pursued their claim on two alternative grounds. They argued, first, that Lord Hoffmann's dealing requirement did not form part of the ratio of OBG (so the courts below had been wrong to strike out their claim). In the alternative, they argued that, even if the dealing requirement, properly construed, did form part of the ratio of OBG, it should now be departed from in accordance with the Practice Statement of 1966 ([1966] 1 W.L.R. 1234) (on the basis that it constituted an undesirable and unnecessary addition to the elements of the tort). The Supreme Court rejected both of these contentions and the action failed.
In relation to the claim that the dealing requirement formed part of the ratio of OBG, Lord Hamblen (who delivered the only full-length judgment) offered various “reasons” for regarding the dealing requirement this way. He noted, first, that it had been expressly endorsed by most of the members of the panel in OBG, albeit only in their general discussion of the tort's ingredients (at [65], [66], [70]). Second, he afforded salience to the fact that the dealing requirement was consistent with the outcomes of, and obiter dicta in, numerous earlier cases (at [64], [67], [68]). Third, he pointed to various Commonwealth cases in which the dealing requirement has subsequently been treated as an element of the tort (at [71]). Finally, he adverted to the fact that “the dealing requirement is consistent with and reflects Lord Hoffmann's concern that the tort be kept within reasonable bounds” (at [69]).
With respect, none of the “reasons” offers a proper basis for concluding that the dealing requirement formed part of the ratio of OBG. The general discussion of the tort by the other members of the panel in OBG is not where one would expect to find the ratio of the case. One would normally look to the leading judgment and the precise basis on which the appeal was decided. Similarly, Commonwealth cases are, according to the popular oxymoronic phrase, no more than persuasive authorities. Third, Lord Hamblen's reference to the outcomes of, and obiter dicta in, various earlier cases is, with respect, a red herring. For, by his own admission, “neither Allen v Flood nor any other pre-OBG authority holds that the dealing requirement is an essential element of the unlawful means tort” (at [89]). In fact, the only previous case in which any reference to dealings between the claimant and third party was made, was Quinn v Leathem [1901] A.C. 495. But what was said there was said obiter: Quinn was a lawful means conspiracy case. Finally, Lord Hoffmann's view that, ideally, the tort ought to be kept within reasonable bounds, suggests more a rule that should exist than one that does exist.
In only one of the three appeals heard in OBG could the dealing requirement conceivably have formed the ratio: the one in Douglas v Hello! But Douglas was formally decided as a breach of confidence case, not one involving the unlawful means tort.
The Supreme Court's decision not to depart from OBG in accordance with the Practice Statement was anchored to a perception that the dealing requirement did not constitute an undesirable or unnecessary development. Borrowing the language of Lord Walker (OBG, at [266]), Lord Hamblen considered it a useful “control mechanism”, a requirement that “minimises the danger of there being indeterminate liability” (at [95]). Lord Hoffmann certainly made much of the need “to keep the tort within reasonable bounds” (OBG, at [135]). And as Lord Walker explained, this restraint was to be achieved by making actionable only losses that arise from “disruption caused, as between the third party and the claimant” (OBG, at [269]).
On close inspection, however, neither the House of Lords in OBG, nor the Supreme Court in Servier, seem to have alighted upon a control mechanism that is fit for purpose (in which case it may well constitute an undesirable development from which we should depart). Before explaining why this may be so, a preliminary point must be made, namely, that the “reasonable bounds” of the tort were not conceived in the same way in these two cases. In OBG, the apparent aim was to restrict those who could claim and the types of loss covered by the tort. This explains their Lordships’ insistence on “disruption caused” to the dealings between the claimant and the third party. In Servier, by contrast, the dealing requirement was seen as a means of preventing indeterminate liability.
Whether seen as a device to keep the tort confined to economic losses suffered by trading partners, or one designed to prevent liability on an inordinate scale, it seems doubtful that the dealing requirement can achieve what is intended. Suppose A uses unlawful means to prevent B providing refreshments for C's birthday party as per a contract between B and C. In such a case, C could show unlawful means of the required kind (intimidation) as well as the necessary dealing link with B. But C's loss would not be economic. C would sue for disappointment. Equally, if A intimidated a dentist, B, into treating C without adequate anaesthesia, C would sue for pain and suffering, not business losses. The dealing requirement, in other words, fails to confine the tort to economic losses as hoped in OBG.
Next, consider an example supplied by the claimants in Servier. Imagine that X, intending to cause loss to British Airways, dupes Heathrow's authorities into believing a bomb has been planted there. Scores of airlines – all of which have dealings with the airport – would have potential claims. The amounts involved would be huge, and very possibly indeterminate. Yet Lord Hamblen impliedly conceded that all the affected airlines could invoke the unlawful means tort in such circumstances (at [86]). So the dealing requirement also cannot guarantee the aversion of indeterminate liability.
Another hypothetical supplied by the claimants is also noteworthy. In it, the defendant deceives a doctor into amputating an unconscious patient's healthy limb by informing him that it is the patient's left leg that is due to be amputated, when in fact it is the right leg. The claimants argued it would be unjust to deny the patient a claim based on the unlawful means tort simply because the deceit would not affect the doctor's dealings with the patient. Lord Hamblen's answer (at [86]) was that there was no such injustice: the claimant, he said, “is likely to have a claim for malicious falsehood”. In so saying, His Lordship admitted that a non-economic loss might be recovered under the banner of another economic tort. What can of worms, then, might that open? If the decision in Servier was intended to restrict the scope of tortious liability in this area, it has arguably opened more doors to litigation than it closed.
