II. Facts of the Case

The case concerned a decision taken by the Italian Competition Authority (AGCM) against the Bank Compass Banca SpA. When consumers signed the contract for a loan, they were simultaneously provided with a contract for insurance policies which included coverage of risks unrelated to the loan. Whereas consumers were not obliged to conclude such insurance policies, the AGCM argued that the commercial practice through the bias of “framing” conditioned consumers to conclude an insurance policy in addition to the loan and that the widespread perception of consumers was that loan and insurance constituted a single, inseparable offer. In order to ensure compliance with the Italian code on consumer law, the AGCM required, in particular, that Compass Banca should implement a seven-day cooling-off period between the signing of the loan and the insurance policy contract. Regarding the cooling-off period as disproportionate, Compass Banca proposed an alternative set of commitments which, however, were regarded as insufficient by the AGCM. On 27 November 2019, the AGCM thus published its decision where it held that the conduct by Compass Banca constituted an aggressive and thus unfair commercial practice in violation of provisions of the Italian consumer code that implements Articles 8 and 9(a) of the 2005 UCPD and it imposed a fine of 4.7 million euros on the bank.Footnote ¹⁵ To estimate the gravity of the violations, the AGCM highlighted how consumers asking for loans typically find themselves in a situation of need requiring rapid access to financial resources.

Compass Banca subsequently challenged the AGCM’s decision. First, in front of a Regional Administrative Court which dismissed the action in 2021. Then, it appealed the Regional Administrative Court’s decision to the Italian Council of State which referred five questions to the CJEU. This case note focuses exclusively on the first questionFootnote ¹⁶: Should the concept of the “average consumer” extend beyond the traditional concept of homo economicus and incorporate insights from the theory of bounded rationality and the increasing risk of cognitive influence in modern market dynamics which has shown how people often seemingly take irrational decisions, and which consequently would impose a need for a greater consumer protection?

III. Judgment by the CJEU

The CJEU begins its reasoning by reframing the question of the referring Italian court. The Italian Council of State had equated the concept of the average consumer, used as a benchmark in particular for the (un)fairness assessment of commercial practices in European consumer law, and the “homo economicus,” the perfectly rational economic actor in neoclassical economic theory. Should this concept be revised in light of findings on bounded rationality, which show that consumers are often exposed to cognitive influences – insights that may justify a higher level of consumer protection? Advocate General Emiliou structured his entire analysis around the idea that the average consumer is not a “homo economicus” and is sufficiently flexible to be perceived, in some situations, as an individual with bounded rationality.Footnote ¹⁷ In contrast to this, the CJEU in its reasoning, made no reference to the “homo economicus” a term which, as observed by the Advocate General, has never been mentioned either by the CJEU in its case law or by the 2005 UCPD.

Taking the “homo economicus” out of the equation, the Court’s analysis focused on whether the average consumer should be defined not only as a reasonably well-informed, observant, and circumspect individual, but also as one whose decision-making capacity is constrained by cognitive biases.Footnote ¹⁸ The Court accordingly begins by situating the average consumer within the architecture of the 2005 UCPD. The concept denotes a consumer who is “reasonably well-informed and reasonably observant and circumspect, taking into account social, cultural and linguistic factors.”Footnote ¹⁹ It is this benchmark of a “notional, typical consumer” which should allow courts in general to be able to assess on the same conditions when a commercial practice is unfair.

The average consumer is distinguished from the “very credulous or naïve” consumer. A commercial practice would not be considered unfair if it only materially distorts the economic behaviour of a consumer less circumspect than the average.Footnote ²⁰ Conversely, the Court demonstrates for the remainder of its analysis that the average consumer is neither constructed in its case law as a perfectly rational market actor. Referencing the 2013 Trento Sviluppo caseFootnote ²¹, it highlights on a general level that the objective of the UCPD is to achieve a high level of consumer protection.Footnote ²² This high level of protection is grounded in the acknowledgement that unfair commercial practices can harm consumers’ economic interests by materially distorting their behaviour.Footnote ²³ As noted by the Advocate General, taking protective measures to prevent their behaviour from being distorted would not be necessary if consumers always acted rationally.Footnote ²⁴ Put differently, why take measures to protect consumer autonomy in the market-place if consumers already act as perfectly rational actors?

The Court continues by exemplifying how the building blocks of the description of the average consumer would account for limitations. First, the average consumer is only “reasonably well-informed.” Consumers are presumed to be aware only of information that can reasonably be expected to be known, considering social, cultural and linguistic factors. Consequently, a commercial practice may be considered as unfair if traders omit material information.Footnote ²⁵ This would, in particular, be the case in technical sectors with major imbalances of information and expertise and where the average consumer in the absence of clear and sufficient information would be incapable of understanding all the details of an offer.Footnote ²⁶

Second, the average consumer is characterised as being only “reasonably observant and circumspect.” According to the Court, it is this building block which would account for the presence of cognitive biases in the behaviour of the average consumer,Footnote ²⁷ and which thus implicitly acknowledges the limitations of the information paradigm in European consumer law. The Court exemplifies this by a few references to past case law: In Lloyd Schuhfabrik Meyer the Court acknowledged that the consumers’ level of attention is likely to vary according to the category of goods and services in question and that the risk of confusion of trademarks in the mind of the average consumer can thus differ.Footnote ²⁸ Furthermore, even if traders fulfil all their information obligations and theoretically eliminate the information asymmetry between the parties, empowering the consumer to make an informed decision, the disproportionate emphasis on certain elements of the information could still be deemed misleading.Footnote ²⁹

For the Court it is thus clear that cognitive biases constitute one of many constraints which may impair the decision-making capacity of the average consumer. However, the influence alone of cognitive biases on the decision-making capacity of a consumer would not be sufficient to declare a commercial practice as being unfair. Instead, it must be “duly established” that “in the particular circumstances of a specific situation” the commercial in practice in question affects the average consumer to such an extent as to materially distort his or her behaviour.Footnote ³⁰ Taking into account the case law of the CJEU, national courts should use their own faculty of judgment to determine the reaction of the average consumers in a given case.Footnote ³¹

IV. Comment

In Compass Banca the Court did not challenge or propose to amend the concept of the average consumer. It left no doubt that the notion remains sufficiently flexible to account for cognitive biases, as the consumer is not perceived as a perfectly rational market actor, but a reasonably well-informed, observant and circumspect one. The novelty, however, lies in the explicit recognition that consumer behaviour may be materially distorted by cognitive biases, among other constraints, requiring national courts and enforcement authorities to take such factors into account. To a certain extent it constitutes a continuation of its previous case law such as Teekanne, where it deviated from a strict interpretation of the average consumer and opted instead for a more realistic analysis of how consumers react to and can be misled by commercial practices.Footnote ³²

Furthermore, past decisions such as Lloyd Schuhfabrik Meyer or Teekanne concerned specific areas of internal market law (trademark and food law)Footnote ³³. In these cases, the Court acknowledged how the risk of confusion of trademarks may vary according to the category of goods and services in question or how a list of ingredients on products might not suffice to prevent that consumers are misled by images on packaging – however, it did not refer to how consumer behaviour can be impaired in general by biases. In Compass Banca, it made this step precisely by generally acknowledging that the definition of the average consumer does not exclude the possibility of taking cognitive biases into consideration when assessing the (un)fairness of commercial practices. A very cautious adaptation of the (un)fairness benchmark in European consumer law.

This slight adaptation is also less surprising when compared to how other recent regulations that regulate the digital single market recognise the exploitation of consumer bias as a problem and thus address practices of dark patterns, i.e., manipulative techniques used to “deceive users by nudging them into decisions (…) or to unreasonably bias the decision-making of the users (…) to subvert or impair their autonomy.”Footnote ³⁴

At the same time, however, the wording of the Court suggests that the threshold for declaring a commercial practice as unfair on the basis of cognitive biases seems rather high: it is not simply the risk of a cognitive bias occurring that would necessarily have the effect of materially distorting the behaviour of the average consumer – but national courts have to ascertain that it is duly established that in the particular circumstances of a specific situation consumer behaviour is materially distorted. Furthermore, the CJEU did not address the referring court’s remark how the risk of cognitive influence in modern markets would impose a need for greater consumer protection – a reference which the Court could have combined with a reference to Article 38 of the Charter of Fundamental Rights of the EU if it had wanted to underline the constitutional importance of consumer protection in the European legal order.

The Court did therefore not question whether the notion of the average consumer might require, for instance, a more vulnerability-attentive interpretation that regards vulnerability as a universal condition rather than an exceptional status confined to specfic group of consumers.Footnote ³⁵ For some scholars, such a shift would be particularly important in digital markets, where choice architectures are designed to infer, amplify or even create vulnerabilities, and where the current framework of the 2005 UCPD, with its benchmark of the average consumer, remains ill-equipped to respond effectively.Footnote ³⁶ It thus still holds true that the court is “unreceptive to a broad reading of what may mislead” and that “[s]ignificant levels of consumer confusion must be shown before the Directive’s control is activated.”Footnote ³⁷ Nevertheless, with Compass Banca the Court has cautiously broadened the spectrum of factors that have to be considered when assessing whether a commercial practice distorts the average consumer’s behaviour.

The judgment may prove particularly relevant for practices in the digital sphere where consumer law intersects with data protection law and the more recent digital platform lawsFootnote ³⁸ and where the exploitation of consumer biases through manipulative techniques are especially widespread.Footnote ³⁹ However, whether a common understanding will emerge across Member States as to when it is duly established that such biases materially distort consumer behaviour remains to be seen. In the absence of clearer guidance and given the broad discretion left to national courts, diverging interpretations and resulting legal fragmentation regarding (un)fairness and the average consumer benchmark are a likely outcome.Footnote ⁴⁰ Further guidance may emerge through a future Digital Fairness Act which should protect consumers against specific vulnerabilities in online environments.Footnote ⁴¹ In light of the recent de-regulatory turn in EU digital lawFootnote ⁴², however, it remains to be seen whether the Commission will move towards a more paternalistic understanding of the vulnerable consumer, or whether the (slightly updated) image of the average consumer is ultimately here to stay as the normative leitbild of the European consumer law framework.