Addressing Online Information Asymmetry

The increase in the popularity of online transactions, which has accelerated during the COVID-19 pandemic (Research and Markets 2020), has the potential to exacerbate the well-documented information asymmetry between consumers and traders (Bakos et al., 2014; Helleringer & Sibony, 2017; Mak, 2012). To an extent, this could be the result of the online environment providing different stimuli that may require different techniques of processing information than what consumers encounter in brick-and-mortar stores (Firth et al., 2019; Voinea et al., 2020). More importantly, however, the Internet removes a lot of barriers to the simultaneous provision of a multitude of information and facilitates various forms and techniques for its provision (Hogarth & Merry, 2011). Consequently, consumers may miss relevant information either due to their unfamiliarity with various online disclosure techniques or due to cognitive processing fatigue resulting from information overload (Hwang & Lin, 2016; Lee & Lee, 2004). This requires policymakers designing information obligations for online consumer transactions to focus not only on what information traders should provide to consumers but also on how this information is to be provided.

The European legislator has so far limited itself to obliging traders to provide online information to consumers in a transparent manner, for example, in such instruments of European consumer law applicable to online transactions as the Consumer Rights Directive 2011 (CRD) and the Unfair Contract Terms Directive 1993 (UCTD). National courts assess transparency of individual terms to a benchmark average targeted consumers (European Commission, 2019, p. 26; Luzak, 2020b). This paper presents findings of an international, interdisciplinary research projectFootnote 1 that documents the fallacies of the current working of the principle of transparency of consumer information and suggests several remedies.

The research results discussed in this paper collate several studies conducted in a complimentary manner. First, some of the results follow from a comparative, doctrinal inquiry into the application of the principle of transparency in rarely compared legal systems of Croatia, Germany, the Netherlands, Poland, and the UK. This part of the research allowed us to document the existing inconsistencies in the application of the principle of transparency in these countries, which represent different European legal cultures (Junuzović, 2018; Luzak & Junuzović, 2019). These inconsistencies have been further explored in a qualitative empirical study, which allowed us to obtain additional information on the current issues with the understanding and application of the principle of transparency in national laws (Luzak, 2020a; Seizov & Wulf, 2020; Wulf & Seizov, 2020a). Based on the findings from the comparative legal research and empirical legal research (Seizov et al., 2019), we have designed a set of recommendations on how to improve the transparent provision of online information to consumers. Finally, we tested these recommendations in a quantitative study (Wulf & Seizov, 2020a). Consequently, the research project consisted of a few studies with different methodologies, with results of each study feeding into the design of a subsequent one. This provided a unique opportunity to explore the principle of transparency in-depth and from various perspectives.

In the following sections, based on the results of our studies, we will first illustrate why it is necessary to further harmonise the assessment of the transparency of online information. This will be argued in two ways. First, by elaborating on the policy aims that the introduction of the principle of transparency was to serve and showing the weakness in the framework of European consumer law that hinders the attainment of these goals (“Towards a Harmonised Assessment of Online Transparency in European Consumer Law”). Second, by providing examples of inconsistencies in the application and interpretation of the supposedly harmonised principle of transparency in various Member States (“Examples of Divergences in the Transposition of the Principle of Transparency in National Law”). Our overview of the issues that result from the national transposition of the principle of transparency is by no means exhaustive, but rather serves to elucidate the need for a change. We then present our novel guidelines on how the principle of transparency could be made more effective (“Empirically Motivated Guidelines for Transparent Disclosures”), which we have empirically tested in a quantitative study this paper builds on. At the end of this paper, we present our policy recommendations (“Policy Recommendations”).

We would like to note from the outset that certain European consumer law provisions were adopted after our research project was concluded and that their potential impact on online information asymmetry was not examined by us. The Modernisation Directive 2019a, amending also the CRD, has introduced new information obligations and transparency requirements for contracts concluded through an online marketplace. As the provisions introduced by this directive were not yet implemented in the national legal systems of the Member States at the time this project was carried out, we have not included these provisions in our study. For the same reason, the references to transparency in Articles 12 (4 and 5) and 19 (1)(c) of the Digital Content Directive 2019b and Articles 3 (5) and 17 (2) Sale of Goods Directive 2019c are neither included in our research. Similarly, as the Digital Markets Act 2022a and the Digital Services Act 2022b were not yet adopted, these regulations have not been included in our study either. We would like to point out, however, that especially Article 25 of the Digital Services Act could influence transparency of contracts concluded through online platforms in the coming years, as it requires a provider of such platforms not to design, organize, or operate their online interfaces in an unfair manner.

A detailed discussion of the “average consumer” benchmark in EU law, to which we will refer repeatedly, is beyond the scope of this paper, as well. With sporadic reference to it stretching back to the 1970s, the concept was first formally defined—as a consumer “who is reasonably well-informed and reasonably observant and circumspect”—by the CJEU in Gut Springenheide (1998) and further developed in subsequent case-law. The standard first made its way into EU legislation in the Unfair Commercial Practices Directive 2005 (UCPD) and has since become a standard benchmark in EU consumer legislation (Duivenvoorde, 2015; Mak, 2011). The concept has been the subject of much theoretical debate (for a brief overview, see Purnhagen & Schebesta, 2020), with many questioning its adequacy, e.g., regarding its vagueness, potential over-optimism regarding actual consumers’ understanding of complicated contract terms, and the applicability of the benchmark to vulnerable consumer groups. Most recently, the structural, architectural power imbalance of the online environment raised questions as to need to recognise consumers’ inherent vulnerability in online transactions (Helberger et al., 2021). In practice, however, European enforcement agencies at the moment do not appear to regard the average consumer concept as overly problematic in application. This, at least, is the upshot of stakeholder interviews conducted by Wulf and Seizov (2020a), and it is mainly why the present paper refrains from investigating the suitability of the concept in more detail.

Towards a Harmonised Assessment of Online Transparency in European Consumer Law

Transparency and Its Aims

The principle of transparency in European consumer law accompanies the introduction of various mandatory information obligations for business-to-consumers transactions. Consequently, its purpose is closely linked to the objectives that information duties, one of the main protection instruments in European consumer law, aim to achieve. On the one hand, information obligations should help alleviate the information asymmetry between consumers and traders (Recital 5 CRD 2011). On the other hand, information obligations, when harmonised across the Member States, should create a level playing field for cross-border traders in the EU, ensuring a better functioning of the internal market (Helleringer & Sibony, 2017, pp. 618–619). Both these objectives can only be attained when traders provide consumers with transparent information.

Information asymmetry results from consumers being the weaker transactional parties, often having fewer resources than traders, with the latter also being repeat players on the market (Akerlof, 1970; Helleringer & Sibony, 2017, p. 620). When consumers do not possess relevant transactional information, they could make transactional choices that are suboptimal for them, and which could contribute to market inefficiencies. To counter this, the European legislator obliges traders to provide consumers with essential transactional information. However, as this information should enable consumers to make informed decisions, whether as to concluding a contract, terminating it, or exercising their other rights, consumers must not merely receive this information but also be able to access and understand it (Helleringer & Sibony, 2017, p. 620). For this reason, one of the objectives of the principle of transparency is to ensure that consumers receive information of a particular quality. The following section will further elaborate on the quality requirements that the European legislator has so far attached to the notion of information transparency.

Consumers are not the only intended beneficiaries of the European harmonisation of information obligations. To facilitate cross-border trade, which is especially relevant for online transactions, the European legislator tried to assure, as much as possible, that traders need to comply with the same information duties in all Member States (Swinson, 2013, pp. 4–5). This would not only allow traders to limit the costs of providing their goods and services to consumers in other Member States but also ensure a fairer internal market, obliging traders to follow the same set of rules, and assisting with the comparability of their offers. Especially this last goal, making traders’ offers easily comparable, requires the adoption of the principle of information transparency. Here, transparency serves the purpose of increasing legal certainty for traders, as well as facilitating an assessment of their compliance with the information obligations. This will be further illustrated in the following section.

Uncertainty of the Current Transparency Toolbox

The European legislator sets the transparency requirements for consumer information only at a general, neutral level, not accounting for the individual consumers’ ability to understand the information (Koivisto, 2022, pp. 26–27), and varies them between different instruments of European consumer law. Articles 4(2) and 5 UCTD 1993 require traders to provide consumers with terms drafted in “plain, intelligible language.” Article 8(1) CRD 2011 reiterates these requirements for providing information in distance contracts, but Article 6(1) CRD 2011 introduces new obligations for traders: to disclose information in a “clear and comprehensible manner.” Moreover, Article 8(1) CRD 2011 states that if the information is provided on a durable medium, it should be “legible.” Furthermore, Article 8(2) CRD 2011 qualifies that certain information, related to consumers being placed under an obligation to pay by the conclusion of a contract, needs to be relayed to them in a “clear and prominent manner, and directly before the consumer places his order.” If that order results in an obligation to pay and is made by clicking on a button or a similar function, online traders need to label it “in an easily legible manner” and by use of an “unambiguous formulation.”

The above-mentioned provisions are just a few examples of the transparency toolbox that the European legislator has used so far. Whilst we could expect that the requirements as to the quality of the information that traders must provide to consumers may be numerous and detailed, only the first of these assumptions is true. An exception here is the mandatory model withdrawal form, which is quite detailed. We could also refer to the forthcoming changes of the precontractual information regarding financial services contracts concluded online. In this respect, the European Commission’s proposal to revise the CRD and to incorporate the provisions regarding distance marketing of financial services within the CRD (COM (2022) 204 final) may introduce rather detailed rules on the precontractual information, which consumers need to receive when they conclude such contracts. For example, the proposal introduces rules as to what information needs to reach consumers, when it must reach them (in principle at least one day before the contract is concluded to ensure that consumers are actually able to read the information), and how it must reach them (by indicating when information may be layered). Moreover, the proposal also introduces an obligation that when a trader provides the required “adequate explanations” by using online tools such as chat boxes, they must provide and explain to a consumer the key information regarding the proposed financial service contract. They also must make human intervention available if a consumer so requests. The proposal also contains a similar provision regarding transparency of information provided on online interfaces, as does the Digital Services Act 2022. Whilst we estimate that this proposal has a potential to further influence the EU transparency framework, this development in the transparency framework was not proposed in time to inform our research project. Additionally, despite the European legislator having recently started to elaborate in its guidelines on the more specific meaning of the transparency requirements, it still does not explain their relationship to one another (European Commission, 2019, part 3.3). Questions such as whether information provided in a plain language can still be unclear or whether unambiguous formulation guarantees comprehensibility remain unanswered (Luzak & Junuzović, 2019). Even the Court of Justice of the EU (CJEU), despite discussing transparency framework in some of its judgements, has not yet explained its various requirements separately, nor their links to each other (e.g., in cases such as Kásler (2014), paras 73–75; Matei (2015), paras 74–75; Andriciuc and Others (2017), paras 44–51).

Two main issues arise from the European legislator adopting only general transparency requirements, varying them between different provisions and legal instruments and not issuing further guidance on their interpretation and application. First, as the transparent information aims to enable consumers to make informed decisions, policymakers should properly instruct traders on how to ensure the high quality of the information they are providing to consumers (Weber, 2021, pp. 83–84). Broadly and vaguely set transparency requirements could miss that mark. Second, the current general transparency framework is bound to lead to legal uncertainty with both traders and national enforcement authorities, who are to ensure the compliance of market practices with the consumer protection framework. This has been confirmed in the qualitative part of our research project, where the interviews we conducted with national stakeholders illustrated divergences in the understanding of the notion of transparency and how it should be applied between various stakeholders within the same Member State, not only when compared with responses from other Member States (Seizov & Wulf, 2020; Wulf & Seizov, 2020a, 2020b).

It is especially relevant to ensure information transparency in the online environment. On the one hand, its absence may diminish the interest in the participation in e-commerce, as consumers could be discouraged by the lack of clarity and certainty of online transactions. Scholars have identified the lack of trust as one of the major deficiencies of e-commerce (Corbitt et al., 2003). On the other hand, transparency leads to accountability, which is crucial in the online environment, where the balance of power between the parties could be more skewed than offline (Select Committee on Communications House of Lords, 2019, p. 17). This could be the result of both the big tech companies dominating the online markets, as well as the increased dependency of consumers on online transactions (Levine, 2020). However, the algorithmic opacity of many online transactions imposes limits on what transparent information can achieve (Grochowski et al., 2021). The European General Data Protection Regulation 2016 (GDPR), which became effective in the Member States in May 2018, has established a number of requirements regarding transparency in the automated (algorithmic) processing of personal data, in particular in Articles 15 and 22 GDPR 2016. However, the effectiveness of these provisions is arguably hampered by vague legal standards and lax enforcement. For example, examining the online disclosures of 100 European companies and organizations, Wulf and Seizov (2022b) found that many of them failed to meet both the legal requirements and the expectations of most consumers. Previously, Wulf and Seizov (2020b) arrived at similar findings based on semi-structured stakeholder interviews. Both papers offer policy recommendations specifically with respect to improving algorithmic consumer transparency. Beyond this, our research project did not address either personalised information or information about algorithmic decision-making (Koivisto, 2020) and, therefore, we do not further elaborate on this topic in this paper.

In “Empirically Motivated Guidelines for Transparent Disclosures” of this paper, we propose a few solutions that could be applied by policymakers to promote further specification of the transparency requirements, which could help with increasing legal certainty. First, however, we will illustrate how the lack of a consistent and detailed transparency toolbox on the European level complicates the interpretation and application of the transparency requirements on the national level, in various Member States.

Examples of Divergences in the Transposition of the Principle of Transparency in National Law

“Plain and Intelligible Language” of Standard Contract Terms

As indicated above, Article 5 UCTD 1993 requires contract terms in “plain, intelligible language.” The CJEU confirmed that this implies that consumers must have had a real opportunity of becoming acquainted with a contract term before the conclusion of the contract (e.g., in cases such as RWE Vertrieb (2013), paras 43–44; Constructora Principado (2014), para 25; Radlinger (2016), para 64; Gutiérrez Naranjo (2016), para 50; Andriciuc and Others (2017), paras 47–48; XZ v Ibercaja Banco SA (2020), para 47). The Court added that the notion of transparency requires that a consumer can appreciate the economic and legal consequences that result from a term. Transparency thus entails more than a term being “formally and grammatically intelligible” (e.g., mentioned in cases such as: Kásler (2014), paras 69–72; Matei (2015), paras 73–74; Van Hove (2015), para 33; Andriciuc and Others (2017), paras 44–48; OTP Bank and OTP Faktoring (2018), para 73; GT v HS (2019), paras 33 and 36; XZ v Ibercaja Banco SA (2020), paras 44–45). Moreover, the relationship between the UCTD’s transparency requirement and its unfairness test is unclear as the lack of transparency may but need not mean that the term is unfair (Amazon EU (2016), para 68; BNP Paribas (2021), paras 61–62). Similarly, the mere fact that a contractual term provides the consumer with incorrect and potentially misleading information that may constitute an unfair commercial practice does not automatically mean that the contractual term lacks transparency (BNP Paribas Personal Finance and others (2021), paras 76–77) or that it is unfair (Pereničová and Perenič (2012), para 44).

Apart from these general observations, the CJEU has not clarified the meaning of the transparency requirement. The European Commission, in its Guidance Document (European Commission, 2019), which is not binding on the courts, has tried to offer such guidance, but with little underpinning. According to the Commission, first, to determine whether consumers have had a real opportunity of becoming acquainted with a contract term before the conclusion of the contract, it must be established that they have had access to and were given the opportunity to read the terms beforehand. Where a contract term refers to an annex or another document, the same applies to these documents (European Commission, 2019, p. 26). Second, a court should look at the comprehensibility of individual terms for average targeted consumers “in light of the clarity of their wording and the specificity of the terminology used, as well as, where relevant, in conjunction with other contract terms” (European Commission, 2019, p. 26; Luzak, 2020b). In this respect, also the language in which the terms are drafted (i.e., in German, Greek or Bulgarian) and the familiarity of consumers with that language are to be considered. Thirdly, the presentation of contract terms is relevant. This includes the clarity of the visual presentation of the terms, including the font size used, whether traders structure the contract in a logical way, whether they give important stipulations due prominence, rather than hiding them amongst other provisions, and whether they place individual terms in a logical place within the set of terms (European Commission, 2019, p. 26).

The Guidance Document therefore suggests that in ascertaining whether the transparency requirement under the UCTD 1993 has been met, national courts should focus on:

  • Whether the trader has given consumers an actual possibility to read the relevant terms;

  • Whether the terms are comprehensible; and

  • How the terms are presented.

However, it provides little indication as to how courts should weigh or interpret these aspects of transparency. Moreover, neither the Court’s case-law nor the Commission’s guidelines indicate how these aspects relate to the notions of “plainness” and “intelligibility.” Do “plainness” and “intelligibility” constitute separate requirements or do they form one notion, without the constituting elements having individual meaning? Obviously, this lack of guidance may lead national legislators to introduce additional or more detailed requirements, national courts to interpret the transparency requirements differently, or national scholarship to develop its own categories for transparency, without much eye for a common European interpretation. To establish whether this is the case, we will briefly compare the transposition of Article 5 UCTD 1993 in Germany, Croatia, and the Netherlands.

In the German version of the UCTD 1993, the transparency requirement of “plain, intelligible language” requires traders to draft written texts “klar und verständlich.” Yet, a corresponding requirement is missing in the German Civil Code (BGB). Instead, § 307(1) BGB establishes that the mere fact that a court finds a term not to be “klar und verständlich” may be sufficient to declare the term unfair. The notions of plainness and intelligibility are not defined separately but rather treated as one notion of transparency. Curiously, German scholars often split the transparency principle into three separate requirements: completeness, certainty, and comprehensibility (Armbrüster, 2004; Präve, 2000; differently, e.g., Heinrichs, 1995). The requirement of completeness asks that all rights, obligations, and mandatory information are listed in full. The requirement of certainty focuses particularly on the clarity and comprehensibility of all legal and factual conditions for any amendment of standard contract terms and consequences thereof, in order to prevent unjustified discretion. Finally, the requirement of comprehensibility concerns the wording and the systematic arrangement of contract terms with the aim of ensuring that their scope, and their economic consequences, are clear to consumers.

In case-law, the focus seems to lie mostly on the criterion of comprehensibility. The German Supreme Court (BGH) indicated that the use of headings and paragraphs can clarify information that without them could have been too difficult to understand (NJW, 2018, 2193). The BGH ascertained that, even if a contract term was sufficiently transparent per se, this did not mean that it would be equally transparent when put into context with other contractual terms (NJW, 2016a, 1575; NJW, 2017, 1306). Thus, the BGH held that in order to ensure that a contract term was clear and comprehensible both in isolation and in the context of the contract, traders could rely on cross-referencing. However, traders need to use cross-references carefully, as the court also indicated that complicated chains of references could obscure the content of terms and therefore violate the transparency requirement (NJW, 2016b, 1646).

In the Croatian language version of the Directive, the transparency requirement is expressed by the formula “jasno i razumljivo” (Junuzović, 2018). Article 53 of the Consumer Protection Act 2014 (CPA) transposes this requirement. This provision obliges traders not only to draft terms using plain (jasan) and intelligible (razumljiv) words, but also to make them easily noticeable (lako uočljiv) (Junuzović, 2018). The Croatian legislator has not provided any further explanations as to the meaning of these three transparency requirements. They are interpreted separately and all three must be met for a standard contract term to be considered sufficiently transparent (Franak (2014), paras 50–56; Franak (2015), paras 33–35). Croatian courts have only briefly discussed the requirement of plainness so far. In the case Franak, the High Commercial Court (in appeal) and the Supreme Court (in cassation) held that the reviewed currency clauses were in plain language as they explicitly and unambiguously tied the value of the credit to the Swiss Franc (Franak (2014), para 52; Franak (2015), para 17). Moreover, the courts held that variable clauses, which allow a bank to unilaterally change the interest rates, were drafted in plain language as they expressly indicated that interest rates were subject to change at the bank’s discretion (Franak (2015), para 33). This line of reasoning was followed in subsequent disputes (Judgement P-123/2015 (2017); Judgement P-4459/15 (2016); Judgement P-4741/16 (2017); Judgement P-7372/15 (2017). Croatian courts thus appear to associate “plainness” with clarity of the term that is under review (Junuzović, 2018).

According to the Zagreb Commercial Court, the court of first instance in the case Franak (2013), for a variable clause to be intelligible, the contract had to specify clear and objective criteria on whose basis interest rates could change. However, the banks had either provided numerous such criteria without clarifying the relationship between them (Franak (2013), paras 152 and 155) or had not provided a single criterion. This means that consumers facing these clauses would be unclear about the risks they were exposed to. For this reason, the court held that the terms were not intelligible (Junuzović, 2018). Later in the procedure, the Zagreb High Commercial Court held that even though the currency clauses were formally and grammatically intelligible, as consumers were not provided with sufficient information concerning these clauses to comprehend their economic effects, they had to be found unintelligible (Judgement Pž-6632/17 (2018), paras 52–59).

Finally, regarding noticeability, Croatian courts have focussed on the placing of the contested terms and their style. For example, the Zagreb Commercial Court was of the opinion that for complex terms, like currency clauses, to be easily noticeable, they had to be highlighted in red ink, placed on the first page of the agreement, with a red hand pointing to them (Franak (2013), paras 176–177). Interestingly, this closely follows the well-known “red-hand rule” of Lord Denning in English contract law (J Spurling Ltd v Bradshaw (1956). Other courts deemed it was sufficient that traders displayed currency clauses on the first page of the credit contract (Franak (2014), para 52), did not hide them within it, but instead highlighted them, e.g., by means of headings (Judgement Pž-6632/17 (2018), paras 66–67). Furthermore, it was relevant that contract terms were not drafted in a small font size or were drafted in the same font size as the rest of the terms (Franak, 2014), para 52). No court has, however, explained what font could count as sufficiently noticeable, establishing neither what the normal font size should be nor when the font size would be considered too small (Junuzović, 2018).

In Dutch law, the words “in plain and intelligible language” (“in duidelijke en begrijpelijke taal”) form one concept of transparency, which does not need to be separated (Loos, 2018). The Dutch Supreme Court (Hoge Raad) clarified that transparency requires the party using standard contract terms to draft the consumer’s rights and obligations as clearly and understandably as possible. Moreover, it is not permitted to obscure the consumer’s rights and obligations by formulas that are unclear or difficult to understand (ABN Amro/Stichting SDB en Stichting Euribar (2019). Traders should thus avoid the use of legal jargon, in particular if it does not literally correspond to legal concepts but paraphrases them. Legal concepts, especially if they are not accompanied by an explanation in understandable language, are not easily comprehensible without obtaining legal advice (Loos, 2018, no 240a). Similarly, Dutch scholarship argues that where a set of standard contract terms consists of a disproportionately long text, traders might have breached the requirement of transparency. In such a case, the sheer length of the text prevents consumers from noticing its relevant provisions. Where a term has a fixed legal meaning, that meaning must be assigned to the term, but an exception must be made if that meaning differs from general usage (Loos, 2018, no 240a). References to legal provisions, a set of tariffs (Rechtbank Noord-Holland (27 May 2020), or a mathematical formula will be equally incomprehensible to consumers (Loos, 2018, no 240b). On the other hand, traders are not required to use oversimplified, childlike language (Loos, 2018).

Turning to the national transposition of Article 4(2) UCTD, it is important to note that in England, whereas Section 64 (1) of the Consumer Rights Act 2015 (CRA) excludes core terms from the unfairness test, Section 64 (2) adds that this takes place only if the term is transparent and prominent. Section 64 (2) CRA 2015 then provides that this is the case if the term “is brought to the consumer’s attention in such a way that an average consumer would be aware of the term.” No corresponding provision exists in the national legislations of the other legal systems included in our study.

In sum, despite a few similarities, we observe the following divergences in the transposing provisions of the UCTD 1993 transparency requirements under Croatian, Dutch, German, and English law. Where under Dutch and German law the words “duidelijk en begrijpelijk” and “klar und verständlich” have not been used to express different elements of the transparency requirements, this is to some extent the case with the words “jasno i razumljivo” in the Croatian version of the Directive. Whereas the requirement to use plain language seems to be equated in Croatia with clarity, the requirement of intelligibility is rather narrowly focused on the possibility for consumers to understand the economic consequences of a term. However, ultimately, Croatian courts also assess the intelligibility of a term by referring to clarity of its economic consequences (Junuzović, 2018). In England, the additional requirement of prominence is added when determining whether a core term is excluded from the unfairness test, albeit that the requirement appears to apply in addition to the requirement of transparency (and not as a part thereof).

In the Netherlands, general guidelines on how traders should understand the transparency requirement are largely missing. German scholarship has developed its own interpretation using the concepts of completeness, certainty, and comprehensibility on the basis of case-law, though neither courts nor legislation expressly mention these concepts. In practice, the German concepts of completeness and certainty seem to largely coincide with the Croatian interpretation of plainness, whereas the German concept of comprehensibility resembles the Croatian notion of intelligibility.

Remarkably, the Croatian legislator introduced an additional element of transparency by requiring standard terms to be easily noticeable (Junuzović, 2018), a requirement which does not follow from the UCTD 1993 and has not been expressly regulated under German or Dutch law. However, the German case-law shows that the German concept of comprehensibility encompasses the element of noticeability, and the Commission’s Guidance Document also suggests that when determining transparency, presentation is a key matter to assess.

Nevertheless, even though these three legal systems show some similarities in the assessment of various aspects of transparency under unfair terms legislation, and despite the fact that the concepts used in the Directive are the same in the Croatian, German, and Dutch language versions, there is no uniform approach to transparency of standard contract terms in these three legal regimes. The existing deviations could not easily be explained by the minimum harmonisation character of the UCTD 1993, as their introduction has not been supported by claims of enhancing consumer protection. This, therefore, suggests that the lack of guidance by the European legislator and the CJEU indeed leads to a different approach to transparency of standard contract terms in the Member States, with the Croatian legislator introducing the additional requirement of noticeability, and German scholarship coming up with new, more detailed, transparency categories. Moreover, as these examples show, without further guidance, the national courts will vary in their interpretation of the transparency requirements.

Clear and Comprehensible, or Plain and Intelligible Consumer Information: One or Two Notions?

Article 6(1) CRD 2011 asks online traders to provide the precontractual information “in a clear and comprehensible manner.” Article 8(1) CRD 2011 also requires the information to be given or made available “in plain and intelligible language” when a consumer concludes an online contract, thus using the same wording as Article 5 UCTD 1993. Interestingly, the German and Dutch language versions of the CRD 2011 use the same wording in both these provisions, which could suggest the interchangeability of these transparency requirements. By contrast, the Polish and Croatian versions follow the English language example, using different transparency notions. Therefore, we may ask whether the European legislator introduced to the CRD 2011 framework two different notions as transparency requirements, representing different aspects and aims of transparency, or whether there is only one such notion. As the CRD 2011 has a full harmonisation character, it should award consumers across the EU the same level of protection. We therefore examined the interpretation of these transparency requirements in various Member States and compared it to the transparency requirements in the UCTD.

In English law, the UCTD 1993 is transposed through the Consumer Rights Act 2015. Section 68 CRA 2015 indicates that contract terms are to be drafted in plain and intelligible language, thus using the same wording as the UCTD 1993. The CRD 2011 is transposed through the Consumer Contracts (Information, Cancellation, and Additional Charges) Regulations 2013 (CCRs). Section 13(1)(a) CCRs 2013 requires information to be given in a clear and comprehensible manner. Yet, no definition of the requirements of clarity and comprehensibility has been introduced, likely due to the fact that the transposition of the CRD 2011 was based on the “copy out” technique in England and the introduction of such definitions would require further elaboration on these concepts by the English legislator (Giliker, 2015, pp. 10–11). Moreover, Section 13(1)(a) CCRs 2013 transposes both Articles 6(1) and 8(1) CRD 2011 and does not distinguish between the transparency notions used in them. This could suggest that the English legislator was of the opinion that the two notions coincide, despite the European legislator having used different terminology in the English language version of the CRD 2011.

In Croatia, Article 57(1) CPA 2014—transposing Article 6(1) CRD 2011—states that pre-contractual and contractual information must be provided in a clear (jasan) and comprehensible (razumljiv) manner. Article 66(1) CPA 2014, transposing Article 8(1) CRD 2011, obliges traders to provide information in a simple (jednostavan) and intelligible (razumljiv) language. Interestingly, as mentioned above, the Croatian legislator introduced the trader’s obligation to use “plain and intelligible” language from the UCTD 1993 as “jasan” and “razumljiv.” Although there is, therefore, some consistency between the transparency requirements applicable in the Croatian law on the basis of Articles 52, 53, 57(1), 66(1) CPA 2014 to testing standard terms’ unfairness and providing consumers with precontractual information, the Croatian and English language versions of the CRD 2011 seem less compatible. The Croatian language version of the CRD 2011 namely introduces the requirement of providing the information in a simple (jednostavan) language, instead of the requirement of “plain” language in the English language version of Article 8(1) CRD 2011. The use of the different transparency requirements, plain versus simple language, could suggest that they aim to achieve different goals, unless both plainness and simplicity are perceived as embodying elements of clarity.

The German language version of the CRD 2011 makes use of one notion to express transparency in both Articles 6(1) and 8(1) CRD 2011—traders must provide the information in a “klar” and “verständlich” manner, adapted to the means of distance communication they use. This notion is literally transposed into German law in Article 312d paragraph 1 BGB and Article 246a § 1 and § 4 paragraphs 1 and 3 EGBGB, thus using the same wording as under the German language version of Article 5 UCTD 1993 and its transposition in Article 307(1) BGB. Unfortunately, it does not look as if the German legislator offered any further explanations of the meaning and scope of these transparency requirements in the provisions transposing the CRD 2011 (Bundestag -Drucksache 17/12637, p.75). This did however occur in the guidelines on cost traps in e-commerce, which concern the CRD 2011 provisions and had been implemented in German law early (Bundestag—Drucksache 17/7745, p. 11). Here, the German legislator explained that the requirements of clarity and comprehensibility ask traders to limit the pre-contractual information to the mandatory information prescribed by law and to make it clearly stand out from the rest of the text, as well as from other design elements on the trader’s website. Traders should further draft it in a manner that is easy to grasp for consumers, and clearly and easily recognisable considering its font style, size, and colour. Furthermore, the legislator set out that comprehensibility requires that information is formulated in a clear and unambiguous language and without any confusing or distracting additions (Bundestag—Drucksache 17/7745, p. 11). Despite these detailed guidelines, it remains uncertain what the precise relationship is between the notions of “klar” and “verständlich” and whether their meaning is to be understood in the same manner in the transposition of the CRD 2011 and the UCTD 1993.

Similarly, the Dutch language version of the UCTD 1993 uses the concept that terms must be drafted in “duidelijke en begrijpelijke taal.” This wording is literally transposed in Article 6:238(2) Dutch Civil Code (Burgerlijk Wetboek, BW). The Dutch language version of the CRD also uses the concept of “duidelijk en begrijpelijk,” for both the general requirement to provide precontractual information in a clear and comprehensible manner and the specific requirement under Article 8(1) CRD 2011. This was transposed ad verbatim in Articles 6:230 l, 230 m(1), and 230v(1) BW. Apparently, the Dutch legislator, too, considered the notions used in various provisions of the CRD 2011 and Article 5 UCTD 1993 to be one and the same (Loos, ).

In the Polish language version of Article 6(1) CRD 2011, the words “clear and comprehensible” translate as “jasny i zrozumiały sposób,” which corresponds to the English wording (Luzak, ). Similarly, in Article 8(1) CRD 2011, the words “plain and intelligible language” translate into “w prostym i zrozumiałym języku.” There is, therefore, a slight difference between the two transparency notions expressed in the Polish language version of the CRD 2011. The first notion is transposed in Article 12(1) Act on Consumer Rights 2014 (ACR) through the words “w sposób jasny i zrozumiały,” meaning “in a clear and comprehensible way.” The second notion is transposed in Article 14(1) ACR 2014, which provides that traders issue information to consumers expressed in “plain language” (“prostym językiem”). This latter provision omits, therefore, the requirement of intelligibility and introduces a different transparency requirement compared to clarity and comprehensibility. However, in the Polish scholarship, it is argued that traders need to provide the information in plain language for it to be clear and comprehensible in accordance with Article 12(1) ACR 2014 (Mikłaszewicz, 2018, Article 14 point 4, Article 17 point 5, and Article 18 point 4). Such a correlation has not yet been drawn on the European level of interpreting these provisions. It should also be mentioned here that the Polish transposition of Article 5 UCTD 1993 in Article 3851 Polish Civil Code (KC) refers to plain and intelligible language as “jednoznaczny” (which actually means “unambiguous”) and “zrozumiały.” It introduces, therefore, a different transparency framework under the UCTD 1993 regime and it remains unclear whether the goals that these transparency requirements aim to achieve differ (Luzak, 2020a).

This short introduction to the transposition measures of the transparency requirements in these five Member States shows that there is no consistency in whether national legislators adopt one or more frameworks of transparency requirements. This is, only to an extent, the result of the lack of consistency in the different language versions of the CRD 2011, as the lack of guidance by the European legislator on the meaning of these terms and their correlation aggravates the issue. Indeed, we are not sure how the CJEU would currently answer the question as to whether there is one or two notions of transparency requirements in the CRD 2011. Nevertheless, we would argue that, despite its use of imprecise terminology, the European legislator intended to impose an obligation on traders to provide information that is transparent both in content and form, and that keeping this goal in mind could lead to a consistent and coherent interpretation and application of both the UCTD 1993 and the CRD 2011 (Luzak, 2020a).

The Button Requirement

Article 4 CRD 2011 clarifies the Directive’s full harmonisation character. This implies that Member States may not introduce or maintain national provisions diverging from the CRD 2011 provisions, including provisions on transparency. However, it is not always clear-cut whether a national provision that at first glance seems to add a transparency requirement indeed diverges from the CRD 2011 provisions and thus is not allowed under that Directive. In two of the examined Member States, questions arose as to the compatibility of the national provisions transposing the CRD 2011 with the so-called button requirement of Article 8(2) CRD 2011. According to this provision, where traders offer the possibility to conclude by electronic means a distance contract which places consumers under an obligation to pay, traders shall make consumers aware “in a clear and prominent manner” of certain information, e.g., the main characteristics of the goods or services or the final price including all taxes and additional costs. In order to do so, the trader may place a button or a similar function on its website that “shall be labelled in an easily legible manner only with the words ‘order with obligation to pay’ or a corresponding unambiguous formulation indicating that placing the order entails an obligation to pay the trader.” In England, Poland, and the Netherlands, the wording of this CRD 2011 provision has been copied into national law. In the two other legal systems included in our study, however, national legislators introduced either a different wording, which would be allowed as the CRD 2011 is a directive, or an additional requirement, which would be in breach of the full harmonisation nature of the Directive.

In Croatia, Article 67(3) CPA 2014 requires the button to be labelled in an “easily visible manner” (lako uočljiv) for the resulting contract to be binding on the consumer. Whether the requirement of an “easily visible manner” is the same as “easily legible manner” (lako čitljivo), which the Croatian version of the CRD 2011 refers to, seems debatable.

In Germany, the possible derogation from Article 8(2) CRD 2011 is of a rather different nature. Article 312j(2) BGB requires that the information mentioned must be provided in a “clear” (klar), “comprehensible” (verständlich), and “prominent manner” (in hervorgehobener Weise). The German legislator thus seems to have added the requirement that the information has to be comprehensible. However, as Article 6(1) CRD 2011 generally requires that all information—and therefore probably also the information that must be provided under Article 8(2) CRD 2011—is to be provided “in a clear and comprehensible manner,” it could be argued that mentioning the requirement of comprehensibility here is only a reminder of that general requirement that must be fulfilled as well under Article 8(2) CRD 2011.

This uncertainty is further amplified by the CJEU declaring that if the button is labelled with a different wording than “order with obligation to pay” it is for the national court to ascertain whether the wording used is “in everyday language and in the mind of the average consumer (…), necessarily and systematically associated with the creation of an obligation to pay” (Fuhrmann-2 (2022), para 33). Leaving this matter to the discretion of national courts with such an open formula almost invites diverging interpretations, even at the national level. In our view, the imprecise formula provided by the CRD 2011 and its interpretation by the CJEU is not workable for traders or courts. For instance, does the wording “place your order” meet the CJEU’s criteria? In the Netherlands, some courts hold that it does (Rechtbank Noord-Nederland (Assen) (13 September 2022); Rechtbank Noord-Nederland (Leeuwarden) (13 September 2022), whereas most courts hold that it does not (Rechtbank Noord-Holland (Haarlem) (4 May 2022); Rechtbank Gelderland (Arnhem) (22 June 2022); Rechtbank Overijssel (Zwolle) (12 July 2022); Rechtbank Rotterdam (4 August 2022); Rechtbank Amsterdam (19 August 2022). An average consumer should understand that if they place an order, this will entail an obligation to pay, and everyday language will tell us the same. However, average consumers may not understand that the order is placed already when they click on this particular button, as they may have learned by experience that additional costs are often added later on in the ordering process.

These examples highlight again that even though it seems clear what transparency requirements follow from Article 8(2) CRD 2011, the interpretation and application of the transparency requirements on the Member State level may differ without further European guidelines.

Empirically Motivated Guidelines for Transparent Disclosures

The first ideas for designing European guidelines for more transparent disclosures came from the multidisciplinary literature review conducted within this research project (Seizov et al., 2019). The review indicated that communication science and document design offer actionable guidance on how to organize long texts with clear hierarchies (Waller, 2017). Understanding of the information could be improved by breaking texts down into meaningful chunks (Johnson & Mayer, 2012), and readers’ attention could be better retained with headlines as well as visual and iconic elements (Holsanova, 2012). Critical linguistics provides insights into the social functions of language (Kress & Hodge, 1979) and advises against the use of modal constructions and hazy temporal adverbs (Pollach, 2005). Eye tracking reveals the different reading paths that experienced and inexperienced subjects take through a complex informative text (Bucher & Niemann, 2012), suggesting that the view of the “average consumer” as a reasonably well-informed, observant, and circumspect individual may set the bar too high, and therefore the information designed with such a reader in mind may be too challenging. In addition, the way consumers interact with disclosures in real life may differ significantly from what policymakers envisage (Ben-Shahar & Schneider, 2014, p. 228). This means that, in practice, consumers may not use disclosures as a pre-contractual tool for making an informed decision—an observation that scores of previous empirical studies have confirmed (Bakos et al., 2014; Ben-Shahar & Chilton, 2016; Helleringer & Sibony, 2017). The previous research findings on transparency, stemming from different disciplines, rarely interlinked, informed the design of our qualitative study.

How to Improve Transparency?

Our qualitative legal research aimed to restart the debate on transparent disclosure with a series of stakeholder interviews conducted in Croatia, Germany, Poland, the Netherlands, and the UK. In conversations with judges, lawyers, consumer protection authorities and organizations, representatives of state institutions, and traders, we explored the transparency characteristics of information obligations, as well as the challenges and opportunities they face, particularly in the context of online trade. We then analysed the interview transcripts to pinpoint shared ideas on how transparency could be improved and divided these ideas into three categories of measures: textual, contextual, and technical (Luzak, 2020a; Seizov & Wulf, 2020; Wulf & Seizov, 2020a, 2020b).

Textual Measures for Improving Transparency

One of the leading findings from our interview sample was that the language of disclosures should be kept accessible. The vocabulary choices should reflect the proficiency levels of the average country resident and, in general, aim lower than the legally defined “average consumer” standard. A text at the B1 language comprehension level of the Common European Framework of Reference for Languages would ensure the greatest level of consumer understanding, and any deviations would diminish transparency. Sentences should be kept short wherever possible. Compound sentences should be broken up to promote comprehension. Furthermore, the texts should employ the active voice and avoid modal phrasing wherever possible. This recommendation also finds support in the relevant research literature, mostly coming from the critical linguistics and persuasive communication perspectives (Benoliel & Becher, 2019; Pollach, 2005; Sigel, 2019).

Textual improvements to the transparency of disclosures concern not only the content but also the layout of the notices. Information obligations should prescribe a clear structure of headings and subheadings. Clarifying the hierarchy of a text in that way is essential to both commanding consumers’ attention and transmitting information to them effectively, according to studies in information design, and a majority of respondents agree with that empirically validated postulate. Additionally, eye-tracking research has shown that interrelated topics should be clustered together to support understanding, as discussed in previous research on information design (Holsanova et al, 2009; Johnson & Mayer, 2012). Another way to improve understanding is to eliminate discrepancies between headlines and text and to avoid misleading asterisked claims. Wherever possible, text length should be reduced, but not at the expense of obscuring or removing relevant contractual details or essential legal terminology. As a middle-ground solution to the information length challenge, many respondents found visual and other non-textual formats as well as one-pagers to be viable transparency enhancers.

Contextual Measures for Improving Transparency

Online disclosure is at play in a variety of contracting contexts, to which our respondents drew particular attention. The characteristics of the product or service being purchased, the business sector in which the transaction takes place, and some defining characteristics of the consumer may all affect the decision to proceed with the conclusion and execution of a contract.

The breadth and depth of disclosure should adequately reflect the complexity and impact of the business sector in which the online trader operates. In particular, the purchasing of inherently complex and/or costly products and services requires greater input of relevant and accessible information. In essence, this is a call for differentiated disclosure that matches the complexity of the transaction and perhaps also accounts for the defining characteristics of consumers (Luzak, 2021). If policymakers follow this suggestion from the stakeholders, this could make pertinent contract information more adequate and effective, as evidenced by experiments in eye-tracking research. Furthermore, sector-specific disclosure requirements can help businesses “promote meaningful performance appraisal” (Evangelinos et al., 2016, p. 386) in the realm of customer communication management, making this a win–win strategy.

Finally, it may also matter for consumer understanding whether a consumer reads a disclosure in the pre- or post-contract conclusion context (Wulf & Seizov, 2020a). The post-contract conclusion context is the situation where a consumer has a dispute with a business. Whilst this setting is not the primary target of disclosure legislation, it is a more realistic instance of the actual use of legal information online. Here the consumer has a real incentive to obtain information about their rights and obligations. Thus, the full potential of transparency-enhancing measures may only be realised in this context. This insight is not only relevant for businesses that disclose legal information online, but it may also inform future disclosure policymaking.

Technical Measures for Improving Transparency

Finally, there are several technical steps that can enhance the transparency of online information. Responsive web design should be tailored to retain clear text structures even on mobile screens, which stakeholders note consumers increasingly use to conclude contracts. Responsive web design has been a major topic in the wider information design and usability literature (Cosgrove, 2018; Groth & Haslwanter, 2016; Hussain & Mkpojiogu, 2015). It may pose practical design challenges to businesses that wish to maintain their visual identity across numerous screen size configurations (Kang & Satterfield, 2019), but it has undeniable transparency-enhancing effects well worth pursuing.

Font sizes should remain legible by setting hard limits on how responsive they are to various screen sizes. Font types, though also a matter of website aesthetics, should not be too outlandish. The scientific literature similarly berates the occasionally careless use of font sizes and types, including transparency-hindering and generally off-putting practices such as relying on a “merry motley of sizes, colours, spacings, fonts” (Ben-Shahar & Schneider, 2014, p. 42). In addition, “a threshold font size of 10–12 point may positively affect the readability of the user interface and therefore decrease think time” (Grover et al., 2017, p. 4). Furthermore, a consistent and transparent usage of font types and sizes has been shown to improve readers’ subjective confidence in the truthfulness of the presented online information (Cho & Weiss, 2017), a factor that is likely to contribute to consumers’ willingness to read and ability to understand information notices. Finally, all vital contractual information must either be presented upfront or be no more than “one click away.” General information such as Terms of Use, Privacy Policy, and Contact should be clearly accessible from any page on the trader’s website. We have summarised the full transparency guidelines in Fig. 1.

Fig. 1
figure 1

Guidelines for transparent online information: summary

Full size image

Testing the Guidelines for Transparent Disclosure

The expert opinions and advice briefly summarised above provide new ideas for behavioural research into disclosure transparency. A host of previous quantitative research adopted an ex ante perspective that tested consumer attention and understanding at the point in the contract conclusion process when they are naturally lowest, i.e., in the rush of completing a transaction. Whether it came to privacy policies (Ben-Shahar & Chilton, 2016), student loan agreements (Darolia & Harper, 2018), online software purchases (Bakos et al., 2014), or product information and health notices (Kersbergen & Field, 2017), the findings were pessimistic: Consumers did not pay attention to the vital information before their very eyes. Instead of dismissing disclosures altogether, however, we should take all these findings with a grain of salt. Apart from expecting consumers to be most attentive when everything points in the opposite direction, most previous behavioural research engaged in minimal disclosure manipulation (i.e., ad hoc textual edits) that was often not consulted with communication experts.

We conducted a behavioural experiment to test empirically motivated, multimodal (i.e., pertaining to text, visuals, and document design) disclosure optimisation measures both in a pre- and a post-purchase online shopping scenario (Wulf & Seizov, 2022a). The optimisations were based on the multidisciplinary, multi-stakeholder findings briefly summarised above, as we focused the experiment on employing the above-mentioned strategies for improvements of disclosures: textual, contextual, and technical. N = 835 British participants were randomly assigned to either a pre- or a post-purchase scenario (contextual measures) and then shown one of four disclosure variants (scenarios): A—a densely written text-only disclosure (“non-transparent disclosure’), B—a linguistically optimised and neatly structured text-only disclosure (“textually optimised disclosure”), C—a visually formatted one-pager that presented the main contractual stipulations in graphic form (“visually optimised disclosure”), or D—a combination of C and then B (“visually and textually/multimodally optimised disclosure”).Footnote 2 Scenarios B through D employed a variety of textual and some technical measures for improving the transparency of disclosures. Whilst the textual measures relied on linguistic optimisation of the disclosure text and applying visual cues to represent its content, technical measures required us to pay attention to font types and sizes, as well as the accessibility of information. With the latter in mind, we allowed the participants to spend as much time as they wished on the disclosure. Subsequently, the participants were asked three specific questions about compensation, return policy, and product characteristics in order to check how much of the disclosure they had read, understood, and remembered. Going back to the disclosure to locate the necessary information was not allowed. We summarise the eight experimental conditions in Fig. 2.

Fig. 2
figure 2

The eight conditions used in the behavioural experiment (source: Wulf & Seizov, 2022a)

Full size image

Unfortunately, in our empirical study, we did not have a chance to test all our technical guidelines. To do so would have required advanced web design and programming capabilities. For example, not least in light of the feedback we received during the stakeholder interviews, it does not seem right that traders may use the same type of disclosure, regardless of the medium through which consumers are accessing the (pre-) contractual information and concluding a contract. Therefore, we recommend further attention being given to responsive web design and layering of information, e.g., when displayed on a PC screen versus a smartphone. Access through a smaller screen should probably necessitate additional guidelines on disclosure drafting and presentation.

Since the chief criticism against disclosure in its current form is that it takes too long to read and obstructs consumer understanding, in our empirical analysis we focus on reading time and understanding, as measured by the share of correct responses to those three questions on the disclosure content (see Fig. 3). As confirmed by multivariate regression analysis (Wulf & Seizov, 2022a), our textually or visually optimised disclosures (Scenarios B and C) significantly reduced disclosure reading time in comparison to the non-transparent disclosure (Scenario A), both before and after the purchase. The multimodally optimised disclosure (Scenario D), on the other hand, received similar average attention time as the non-transparent one. This is not surprising, considering that in terms of pages (words and illustration combined), it was nearly twice as long. As expected, pre-purchase reading times were only about half as long as post-purchase reading times. This finding supports the expectation of the relevance of the context in which disclosures are approached, with consumers having a higher motivation to read and giving more attention to disclosures after completing a transaction and in case of a problem.

Fig. 3
figure 3

The four disclosure variants in a pre- and post-purchase scenario. The left-hand panel shows average reading times in seconds; the right-hand panel shows consumer understanding, measured as the average percentage of correct answers to three knowledge questions (source: Wulf & Seizov, 2022a).

Full size image

Turning to disclosure understanding, here too the empirically motivated optimisations we implemented were clearly successful. In both the pre- and post-purchase scenarios, there is a stable trend of increasing consumer understanding as we move from variant A all the way to variant D. Compared to the non-transparent disclosure (Scenario A), the textually and visually optimised disclosures (Scenarios B and C) deliver greater understanding at a lesser time investment. The multimodally optimised disclosure (Scenario D), in turn, requires the same time investment as the non-transparent one (Scenario A) but brings significantly more understanding; in other words, the consumer’s time is well spent. The post-purchase Scenarios C (visually optimised) and D (multimodally optimised) showed the highest frequency of correct answers, both at 74 percent. Respondents who viewed the multimodally optimised disclosure in the pre-purchase scenario and the visually optimised disclosure in the post-purchase scenario were most likely to answer all three knowledge questions correctly.

There are a few immediate take-aways from the findings above. First, when used alone, text is the least efficient communication medium for consumer information online. In the non-transparent textual disclosure (Scenario A), both before and after the purchase, participants took a long time to learn the least. In addition, textual optimisation alone can only go so far in reducing the information burden and informing consumers better: moving from textually (Scenario B) to visually enhanced (Scenario C) information presentation dramatically improved both disclosure reading time and consumer understanding. Combining the two modes (Scenario D) increased reading time in comparison to the two mono-modal optimised scenarios, likely due to the sheer volume of shared information, but it also significantly improved understanding, particularly with regard to participants who read the information before the purchase.

Furthermore, the overlap in consumer understanding between the visual and multimodal post-purchase scenarios (Scenarios C2 and D2, both at 74% understanding) takes us back to the question of matching disclosure complexity with product complexity, which we have discussed above. The behavioural experiment involved the purchase of custom-designed drinking glasses, an everyday product that is unlikely to require lengthy consumer information notices. Thus, investing more time and attention into reading a full-blown multimodal information notice to find the solution to a transaction problem does not bring gains in understanding—the same essential information can be presented transparently in a sparing visual form. Therefore, for simple products, a well-designed visual one-pager may be sufficient (i.e., Scenario C), whilst more complex transaction may require a thorough multimodal display (i.e., Scenario D), as it best promotes understanding, especially in the pre-purchase stage.

Finally, the behavioural experiment confirmed that the reading context plays a vital role in consumer attention and knowledge acquisition. Compared to post-purchase scenarios, consumers dedicated significantly less time to reading the Terms & Conditions and invariably performed worse on the knowledge test in pre-purchase scenarios. Changing the disclosure-reading context from pre- to post-purchase resulted in consumer attention increases between 41 and 162 percent as well as in consumer understanding gains between 14 and 21 percent. This finding adds further credence to the debate on differentiated disclosure. For instance, a short, simplified, visually rich disclosure may perform best in the low-attention pre-purchase setting, whilst a thorough, more complex multimodal disclosure is more appropriate to the high-motivation, less time-sensitive post-purchase scenario.

In sum, the findings from the sample of 835 British participants confirmed that transparency in online disclosure is definitely worth pursuing and could be optimised. Increasing the transparency of a disclosure text reduced the reading time and improved participants’ understanding and retention of the contract terms. Therefore, it is premature to declare information obligations a lost cause. We lay out our proposals to policymakers in the following paragraph.

Policy Recommendations

The European Commission recently started to apply legal information design strategies to improve consumer disclosures, e.g., in the Commission Implementing Regulation 2019, and this paper further evidences the need for such approach. First, on the basis of doctrinal and comparative legal research, we illustrated why it is necessary to further harmonise the guidance on the principle of transparency, indicating differences in national laws and practice in applying this principle. Second, we documented the stakeholders’ interest in policymakers providing additional guidance on how to draft transparent consumer disclosures and how to enforce compliance with the principle of transparency. Third, we proposed the adoption of specific, novel transparency guidelines, which have been empirically tested. The guidelines that we recommend envisage introducing all of the above-mentioned types of measures: textual, contextual, and technical. Whilst the legislative transparency requirements could remain general, e.g., plain and intelligible language, traders, consumers, and enforcement authorities all require more legal certainty as to what amounts to compliance with these requirements. Our research project suggests the way forward in this respect.

The empirical study we conducted showed that textual adjustments to disclosures improved both reading time and consumer understanding of the disclosed information. Therefore, our first recommendation is for the European Commission and other policymakers to indicate that the transparent provision of information necessitates traders drafting disclosures:

  • In short sentences,

  • In active voice,

  • In accessible language, and

  • In a coherent layout.

These textual guidelines could be made more specific by employing technical measures. For example, policymakers could declare that accessible language requires drafting disclosures at the B1 language comprehension level, that short sentences would not exceed 3 lines of text at a font size of 10–12, or that a coherent layout requires providing a table of contents for disclosures. The more detailed the guidelines, the more legal certainty traders and consumers can expect. That being said, traders should retain some flexibility to tailor their disclosures in a way that matches their brand design, e.g., by choosing the colour, exact size, and type of the disclosure font.

Our empirical study further showed that visual cues improve the reading time and understanding of disclosures, thus policymakers should further consider improving the transparency framework by prescribing their use. The visual cues include providing consumers with a one-pager, using thematic icons and minimal text, which are textual strategies for improving the transparency of disclosures. Our recommendations go further as we see the necessity of also following contextual guidelines, which refer to the type of contract being concluded. If policymakers draft disclosures for simple everyday transactions, e.g., consumer sales contracts, we would recommend prescribing the use of visual cues. However, when a transaction is more complex, e.g., consumer credit, the recommendation should be to use a multimodal disclosure that combines visual cues and transparent textual information. This would increase the chances of consumers understanding these complex products and services, even if it does not decrease the reading time of disclosures.

Finally, our research emphasized that consumers are more motivated to access and attempt to understand disclosures in a post-contractual context. Shifting the debate on information obligations to highlight their post-contractual function could help to alleviate the current information apathy. A visual one-pager could serve as a model for such brief and accessible pre-contractual disclosures of the key contract terms. The full contract terms, which consumers are unlikely to read at this stage of the transaction, should continue to be brought to their attention, but without an expectation that they will read them. Instead, consumers must be given the option either to access the full terms immediately and/or to save them for later reference if questions or concerns arise. This proposal would not fundamentally alter the current disclosure policy. It would retain the core information obligations but compartmentalise them for greater effectiveness and the benefit of consumers (Wulf & Seizov, 2022a, 2022b). It would reduce the information overload at the precontractual stage by limiting the information obligations at that stage to the core information (Weber, 2021, p. 84). This would limit the burden and costs for traders at that stage whilst at the same time improving consumers’ knowledge of their core rights and obligations. They would also retain access to the information they only need at a later stage.