Special Rules for the Formation of Cross-Border B2C Electronic Contracts: Information Disclosure and Conflicts of Effect in Cooling-Off Provisions

Abstract

Cross-border business-to-consumer (B2C) electronic commerce compresses contract formation, performance and consumer redress into a single, highly automated platform interface, placing long-settled contract doctrine on declarations of intent, access to information and withdrawal under strain. This article takes the consumer cooling-off period, and the right of withdrawal that implements it, as its organizing object. Using doctrinal, comparative and economic analysis, it examines the interaction of pre-contractual disclosure, the commencement of the withdrawal period, the consequences of non-disclosure, immediate performance of digital content, choice-of-law clauses and cross-border jurisdiction. The article argues that the cooling-off period should not be read as an isolated post-formation remedy but as a continuous protective chain running from disclosure, through understanding and valid consent, the commencement of the period and its exercise, to cross-border enforceability. The decisive conflict in cross-border transactions is not primarily between the differing withdrawal periods of different jurisdictions but between platform party autonomy and mandatory consumer protection. The analysis is anchored in the most recent regulatory and judicial developments, including the mandatory electronic withdrawal function that applies in the European Union from 19 June 2026, the Court of Justice's 2026 rulings in Sky Österreich and Eisenberger Gerüstbau, the vacatur of the United States Federal Trade Commission's Negative Option Rule in 2025, and the removal of the European Union's EUR 150 customs de minimis threshold in July 2026. Drawing on the empirical economics of return policies, online trade costs and privacy regulation, the article proposes a functional withdrawal-protection framework in which institutional effectiveness is assessed by whether consumers can learn of, understand, operate and cross-border enforce the right, and it specifies the conditions under which such protection is economically defensible rather than merely formally available.

1. Introduction

Cross-border electronic commerce has changed more than the space in which transactions occur; it has changed the temporal and informational structure of contract formation itself. In a traditional face-to-face transaction, a consumer ordinarily forms a judgment incrementally, through inspection of the goods, explanation by sales staff and a written instrument. In a B2C online transaction, the contract is typically concluded instantaneously through a handful of clicks, checkboxes and automated confirmations, and what the consumer confronts is a body of standard terms, layered hyperlinks, pop-ups, default settings and a trader established in another jurisdiction. The central question is therefore not whether the consumer formally clicked to agree, but whether the decision rested on information of any real significance, and whether, having decided, the consumer retains a workable opportunity to correct course. Disclosure and withdrawal are accordingly two mutually dependent institutions of digital consumer protection rather than two separate ones. Research on disclosure in electronic commerce has established that informational transparency bears directly on consumer trust, judgment and transaction quality, while the complexity of platform interfaces can drive a wedge between formal disclosure and actual comprehension (Kolotylo-Kulkarni et al., 2021; Veltri et al., 2020; Zhou et al., 2018).

The cooling-off period, in comparative consumer law, is ordinarily expressed as a right of withdrawal from a distance contract: within a statutory window the consumer may undo the transactional commitment without giving reasons. The institution does not simply deny contractual force. It supplies an ex post corrective for informational asymmetry, the impossibility of physical inspection, impulsive decision-making and the behavioural influence of digital interfaces (Luzak, 2014; Rekaiti & Van den Bergh, 2000; Sparks et al., 2014). In a cross-border B2C setting, however, the existence of the right is not the same thing as its effective realization. A consumer may not know when the period begins; may be told that digital content is non-refundable once downloaded; may accept foreign law and a foreign forum in standard terms; and may find the withdrawal facility buried several menus deep.

The existing literature addresses disclosure in electronic commerce, withdrawal in distance contracts, digital content contracts, unfair terms and cross-border consumer protection, but it tends to take a single institution as its unit of analysis (Helberger et al., 2013; Loos & Luzak, 2016; Meskic et al., 2022; Steennot, 2013). Recent systematic reviews confirm the pattern: work on cross-border consumer protection and on cross-border consumer behavior has grown rapidly while remaining organized around discrete doctrinal or behavioural topics (Chen et al., 2023; Goetzinger & Spremić, 2025). This fragmentation obscures a fact that matters: disclosure, the withdrawal period, digital performance and conflict of laws are not independent rules but a joint mechanism that jointly determines whether a consumer can actually exit a contract. Normative conflicts of precisely this kind—between rules that are individually coherent but jointly incompatible when embedded in an automated transaction—have been identified as a structural feature of electronic contracting (Giannikis & Daskalopulu, 2011).

This article therefore advances a functional framework that treats cross-border B2C withdrawal protection as a chain: disclosure, understanding and valid consent, commencement, exercise, and cross-border enforceability. Three considerations make the present moment an unusually productive one for such an exercise. First, the European Union has moved from regulating the text of disclosure to regulating the operability of rights: from 19 June 2026 traders concluding distance contracts through an online interface must provide a dedicated electronic withdrawal function (Directive (EU) 2023/2673, inserting Article 11a into Directive 2011/83/EU). Second, the Court of Justice has in 2026 twice reshaped the boundaries of the right, narrowing the digital-content exception in Sky Österreich and, for the first time, recognizing an outer limit of abuse in Eisenberger Gerüstbau. Third, the comparative picture has become sharply divergent: the United Kingdom's subscription regime under the Digital Markets, Competition and Consumers Act 2024 awaits commencement, China's 2024 Implementing Regulation on consumer rights has taken effect, and in the United States the Federal Trade Commission's Negative Option Rule was vacated in its entirety in Custom Communications before it could be enforced. A framework built on the assumption of regulatory convergence would misdescribe this landscape.

The article also departs from the existing literature in a second respect. Legal scholarship on withdrawal rights has largely proceeded without engaging the empirical economics of the institutions it evaluates, even though those economics are now reasonably well developed: there are quantitative estimates of how return-policy leniency affects purchasing and returning, of how far distance and trust depress cross-border online trade, and of what compliance-heavy digital regulation costs firms and consumers. Section 7 brings that evidence to bear, and does so at a moment when cross-border B2C cost structures are being reset by the removal of the European Union's EUR 150 customs de minimis threshold. The purpose is not to reduce consumer protection to efficiency, but to identify which parts of the protective chain are economically robust and which impose costs disproportionate to the protection they deliver.

The remainder of the article proceeds as follows. Section 2 sets out the normative basis of the cooling-off period and the analytical framework. Sections 3 to 6 examine the four domestic links of the chain: disclosure; understanding and valid consent, including the digital-content problem; commencement and the consequences of non-disclosure; and exercise, where the shift from textual disclosure to interface operability is most visible. Section 7 addresses the cross-border link—choice of law, jurisdiction and actual compliance. Section 8 supplies the economic analysis. Section 9 discusses contributions, a research agenda and limitations, and Section 10 concludes.

2. The Cooling-Off Period and the Analytical Framework

2.1. From Contractual Bindingness to a Corrective Mechanism

Two justifications dominate the theoretical literature on withdrawal rights: informational disadvantage and behavioural correction. In a distance transaction the consumer cannot adequately encounter the goods or service before contracting and therefore needs an opportunity to reassess after taking delivery or learning more; at the same time, flash promotions, countdown timers, default options and interface stimuli can amplify impulsive decisions, so the law restores a space for reflection through a short, no-reasons exit (Karampatzos & Ilić, 2023; Luzak, 2014; Rekaiti & Van den Bergh, 2000). The first rationale concerns the quality of information; the second concerns the process of decision. Digital B2C contracts characteristically exhibit both defects at once, which is why treating the two rationales as alternatives understates the case for the institution in this setting.

European Union law institutionalizes the distance-contract withdrawal right as a structured protective mechanism. Consumers generally enjoy a fourteen-day period; the trigger differs between contracts for goods and contracts for services; and where the trader has failed to supply the required information about the right, the period is ordinarily extended (Directive 2011/83/EU). More importantly, the right is tied directly to the trader's pre-contractual information duties: the trader must explain that the right exists, its duration, its conditions and the procedure for exercising it. The cooling-off period is thus not a post-formation gratuity but a component of the rules of contract formation—a point that the case law has progressively sharpened, and which the empirical literature on whether consumers read standard terms at all makes difficult to dismiss as formalism (Bakos et al., 2014).

Cross-border B2C contracts are highly standardized and non-negotiable. Consumers ordinarily cannot alter the withdrawal clause, the applicable-law clause or the platform's process; their freedom consists in accepting or declining the contract as a whole rather than in genuine negotiation over terms. The withdrawal right accordingly supplies a limited corrective window for instantaneous, standardized contracting, while exceptions for digital content, customized goods and fully performed services protect the trader's legitimate reliance. The design question is where that boundary should fall, and the jurisprudence of 2020–2026 has been occupied almost entirely with fixing it.

2.2. A Five-Stage Protective Chain

The framework used here disaggregates withdrawal protection into five sequential stages, each a necessary condition for the next. Stage one is disclosure: before the consumer is bound, the trader must communicate the right and its conditions in a manner that is clear, accessible and capable of being retained. Stage two is understanding and valid consent, which matters most where digital content is to be supplied immediately, and which requires the separation of three distinct acts—consent to contract, request for early performance, and acknowledgment that the right will thereby be lost. Stage three is commencement: the trigger rule must correspond to the transaction type and to the point at which the consumer actually obtains an opportunity to evaluate. Stage four is exercise: the platform must furnish a low-friction and evidentially secure withdrawal channel. Stage five is cross-border enforceability: applicable-law and jurisdiction clauses must not deprive the consumer of mandatory protection.

The analytical payoff of the framework is that it makes failure diagnosable. A right that exists in substance but is undisclosed fails at stage one; a right that is disclosed but waived through a pre-ticked box fails at stage two; a right whose period is uniformly stated as running from purchase fails at stage three for goods contracts; a right that requires a telephone call to a foreign call center fails at stage four; and a right that can be vindicated only in a distant forum under unfamiliar law fails at stage five. Because the stages are sequential, the protective value of the chain is bounded by its weakest link rather than by the sum of its parts. This is the sense in which the framework is functional rather than doctrinal: it evaluates outcomes rather than the formal presence of rules.

The framework also clarifies what the phrase 'conflict of effect' should mean in this field. The conflict of practical importance is not between a fourteen-day period in one jurisdiction and a seven-day period in another. It is between the platform's private ordering—expressed through the combination of standard terms, interface architecture and forum selection—and the mandatory consumer protection of the consumer's own market. Because that combination operates jointly, doctrinal review of clauses one at a time will systematically underestimate its effect, and a functional assessment of the clause bundle together with the interface flow is required.

2.3. Scope, Sources and Method

This article is a doctrinal and comparative synthesis rather than an empirical study of platform terms. Its legal sources are the European Union consumer acquis and the conflict-of-laws instruments that frame it, together with the case law of the Court of Justice and selected national decisions from the United Kingdom and the United States; its comparative reference points are the United Kingdom's Digital Markets, Competition and Consumers Act 2024, the People's Republic of China's 2024 Implementing Regulation on the Protection of Consumer Rights and Interests, and the rulemaking record of the United States Federal Trade Commission. Its economic sources are peer-reviewed empirical studies of online trade costs, return policies and the effects of digital regulation.

European Union law is used as the principal reference point because its rule system is the most complete, not because it is representative. Two limits follow and should be stated at the outset. First, the international framework for electronic contracting deliberately leaves B2C transactions largely to domestic law: the UN Electronic Communications Convention excludes contracts concluded for personal, family or household purposes from its scope, so there is no general international instrument against which national withdrawal regimes can be measured. Second, several of the developments analyzed here are very recent—the electronic withdrawal function applies only from 19 June 2026, and the United Kingdom's subscription regime has not yet commenced—so the analysis concerns institutional design and its predictable effects rather than observed compliance outcomes.

3. Stage One: Disclosure as the Gateway to an Effective Right

3.1. The Dimensions of Digital Disclosure

Disclosure in cross-border electronic contracts should not be assessed by asking whether words were present on a page. Digital disclosure has dimensions of position, hierarchy, legibility, timing and retainability. A consumer may be technically able to reach a withdrawal policy through a hyperlink while never actually encountering that information before the transactional decision is made. Experimental work on platform transparency shows that the manner of presentation affects attention, comprehension and choice (Veltri et al., 2020), and survey evidence on perceived transparency in B2C commerce points the same way (Zhou et al., 2018). Legal 'disclosure' should therefore be tied to what the consumer can reasonably obtain and understand before being bound.

Information about the right of withdrawal must answer at least four questions: whether the right exists, how long it lasts, when the period begins and how it is exercised. If any of these is obscured, the right may lose practical meaning. Steennot (2013) shows that the protective function of the distance-contract withdrawal right is closely bound to the trader's information duties, and European Union law makes the linkage operative by extending the period where disclosure has not been given. Directive 2011/83/EU thus connects an information obligation directly to the temporal effect of a substantive right.

There is, however, a limit to what disclosure can achieve, and the literature is candid about it. The click-through evidence indicates that only a negligible fraction of consumers access standard terms at all, and those who do spend very little time on them (Bakos et al., 2014). The broader critique holds that mandated disclosure systematically fails to accomplish what is asked of it, because the cognitive and temporal demands it makes of recipients are unrealistic (Ben-Shahar & Schneider, 2014). This is not an argument for abandoning disclosure obligations, which retain an evidential and enforcement function even when unread. It is an argument for not treating disclosure as the terminus of protection—precisely the move that the functional framework resists, and precisely the move that Article 11a of Directive 2011/83/EU abandons by shifting the regulatory target from text to interface.

3.2. What Courts Have Made of Online Disclosure

Courts in several jurisdictions have converged on a notice-and-assent test for online terms while differing in how demanding they make it. In the United States, the Ninth Circuit held in Berman that a contract formed through a website is enforceable only where the site gives reasonably conspicuous notice of the terms and the consumer takes an action unambiguously manifesting assent to them; the webpages at issue satisfied neither limb. Chabolla applied that test to a 'sign-in wrap' flow in which the terms were hyperlinked but the operative buttons read only 'Continue' and 'Redeem Now', and held the consumer not bound—significantly for present purposes, in litigation about automatic-renewal charges. The doctrinal lesson is that the same interface features that defeat contract formation are the features that defeat effective disclosure of withdrawal rights.

English law reaches a comparable test by a different route. In Parker-Grennan the Court of Appeal asked whether the trader had done what was reasonably sufficient to bring the terms to the notice of the user, upheld the click-wrap process on the facts, and in doing so confirmed both that click-wrap can incorporate terms and that the sufficiency of notice is a substantive question rather than a formality. European Union law approaches the problem from the direction of the transaction button rather than the terms: Fuhrmann-2 required that the consumer be able to understand, from the wording on the ordering button alone, that clicking creates an obligation to pay, and Conny extended that requirement to situations where the payment obligation is conditional on later events. Read together, these decisions establish that the critical informational moment is the moment of the click, not the existence of a document somewhere behind it.

The convergence is analytically important because it shows courts operating, without using the vocabulary, on the first two stages of the protective chain. What they are testing is not the presence of terms but whether the interface delivered the information at the point of decision. That is a functional standard, and it aligns judicial review with the direction of legislative reform. It also indicates why transparency has proved difficult to pin down as a legal concept: it operates simultaneously as an informational, a procedural and a substantive requirement, and consumer law and data protection law have developed partly incompatible versions of it (Jabłonowska & Tagiuri, 2023).

3.3. Disclosure and the Allocation of Proof

Disclosure also raises a question of proof that the literature on transparency rarely addresses. Web pages can be changed. If a consumer must establish that a particular page failed to display withdrawal information correctly some months earlier, the evidential burden is close to impossible to discharge. The more defensible institutional arrangement requires the trader to retain the version of the terms actually displayed at the moment of contracting, together with timestamps and the consumer's confirmation record. The traceability of digital transactions ought to reduce, not increase, evidential difficulty.

This is one respect in which the functional framework generates a concrete and currently unmet design requirement. Directive 2011/83/EU requires confirmation of the contract on a durable medium, and Article 11a will require a withdrawal function, but neither imposes a retention obligation on the trader with respect to the interface state at the moment of formation. Where the trader is established outside the consumer's jurisdiction, the asymmetry compounds: the evidence is held abroad by the party with an interest in its unavailability. A retention-and-production duty, tied to the existing confirmation obligation, would be a low-cost addition relative to the enforcement gain, and it would give the notice-and-assent tests developed in Berman, Chabolla and Parker-Grennan something reliable to operate on.

4. Stage Two: Understanding, Valid Consent and the Digital Content Problem

4.1. The Enhanced-Consent Structure

Digital content is the most demanding setting for the cooling-off period. Tangible goods can be returned, but once a download, a stream, a software activation or immediate access has occurred, the trader may already have supplied an irreversible consumption value. If a consumer could use digital content in full and still withdraw unconditionally, opportunism would follow. Consumer law therefore permits the right to be lost through immediate performance, but on strict conditions. The question is not whether digital goods are categorically non-returnable; it is whether the consumer made an independent, express and informed choice before performance began.

European Union law imposes cumulative conditions for digital content not supplied on a tangible medium: the consumer must give prior express consent to performance beginning within the withdrawal period and must acknowledge awareness that the right will thereby be lost (Directive 2011/83/EU; Directive (EU) 2019/770). The structure demonstrates that beginning a download does not of itself establish waiver. An effective waiver requires that the request for immediate performance be separated from the acknowledgment of the consequence. Burying the loss of the right in general terms of service, in a pre-ticked box or in extended prose cannot satisfy this heightened consent standard, and the scholarship on digital content contracts anticipated exactly this difficulty when the regime was designed (Helberger et al., 2013).

The doctrinal reason for the heightened standard is that an ordinary click expresses an intention to enter a contract, whereas waiving a statutory withdrawal right requires a stronger degree of knowledge and confirmation, because the legal consequences differ: the former creates a contractual relationship, the latter extinguishes a protective right. Digital content therefore displays most clearly the inseparability of disclosure, consent and the rules on contractual effect. It is also the setting in which the interface can most cheaply manufacture apparent consent, which is why the empirical literature on manipulative design is directly relevant here rather than only at the stage of exercise (Mathur et al., 2019).

4.2. The Digital Content and Digital Service Boundary after Sky Österreich

The scope of the exception has now been substantially narrowed. In Sky Österreich the Court of Justice held that a streaming offering constitutes a digital service rather than digital content where the trader's offering is dynamic in nature and goes beyond the stable or continuous provision of specific content; the Court pointed to the updating of available content and to personalized recommendations based on user behavior as indicia of that dynamism. The consequence is that providers of such offerings cannot rely on the digital-content exception to the right of withdrawal at all, and the Court emphasized that the exception must be interpreted strictly.

The significance of this ruling for the framework advanced here is considerable, and it runs in a direction that the earlier literature did not anticipate. The exception was designed around a conception of digital content as a discrete, deliverable item whose consumption is irreversible. Contemporary platform offerings are not of that character: they are continuously updated, algorithmically personalized and consumed over time. Sky Österreich recognizes that the features which make an offering commercially attractive—personalization and dynamic curation—are the very features that remove it from the category in which withdrawal can be waived. Personalization, in other words, now carries a withdrawal-law cost. This is a genuinely novel form of conflict of effect, and it falls squarely at stage two of the chain: the more the interface adapts to the individual consumer, the less available the waiver of the right becomes.

The ruling also exposes a classification burden that cross-border traders are poorly placed to bear. Whether a given offering is dynamic enough to constitute a service is a fact-sensitive question that the Court expressly left to the referring court. A trader serving twenty-seven national markets through one interface must now either classify each offering and vary its withdrawal architecture accordingly, or default to treating everything as a service and forgo the exception entirely. The second course is cheaper, and for that reason the practical effect of Sky Österreich is likely to be a broad de facto extension of withdrawal rights over digital subscriptions—an outcome achieved by compliance economics rather than by legislative choice.

4.3. Compensation for Performance during the Withdrawal Period

Narrowing the exception raises the question of what the trader may recover for performance already rendered. Article 14(3) of Directive 2011/83/EU requires the consumer who has requested performance during the withdrawal period to pay an amount proportionate to what has been supplied, and the Court addressed the calculation of that amount in PE Digital, holding that the assessment must be made by reference to the total price agreed and, where that price is excessive, on the basis of the market value of what was supplied. Sky Österreich confirms that this compensatory route remains available where the withdrawal right survives: the provider may claim under Article 14(3) for services supplied up to the point of withdrawal.

The combination is doctrinally coherent and, in economic terms, superior to the exception it displaces. An exception operates as a cliff: consent to immediate performance extinguishes the right entirely, so the marginal cost to the consumer of a single minute's use is the whole value of the option to withdraw. A proportionate-compensation rule operates as a price: the consumer pays for what was consumed and retains the option for the remainder. Cliff rules invite the manufactured consent that the enhanced standard is meant to prevent, because the payoff to obtaining a click is discontinuous and large. Pricing rules do not. Viewed through the functional chain, the shift from Article 16(m) exception to Article 14(3) compensation converts a stage-two vulnerability into a stage-two pricing question, which is a considerable improvement in the robustness of the protection.

5. Stage Three: Commencement and the Consequences of Non-Disclosure

5.1. Differentiated Trigger Points

The commencement of the withdrawal period is the technical question most often underestimated in this field. In a contract for goods, the consumer can ordinarily complete an assessment of quality and suitability only after taking physical possession, so the trigger is linked to delivery; in a contract for services the period generally runs from conclusion of the contract. The divergent triggers reflect a functional difference in what the right is for, not a mere procedural arrangement. Where a cross-border platform states uniformly in its standard terms that the period runs for fourteen days 'from purchase', the effect across transaction types is a substantive curtailment of the consumer's rights, even though no individual clause states a period shorter than the statutory one. This is the clearest illustration of why clause-by-clause review is inadequate: the clause is facially compliant and substantively non-compliant at the same time.

Eisenberger Gerüstbau has now added a further complication at this stage. The Court held that an addendum concluded exclusively through means of distance communication may itself constitute a distance contract even where the main contract does not. A variation agreed by email or in-app message can therefore carry its own withdrawal period and its own disclosure obligations, independently of the transaction from which it arises. For platform businesses whose contractual relationships are continuously modified—upgrades, add-ons, tier changes, renewals—the practical consequence is that commencement is not a single event in the life of a relationship but a recurring one. The framework accommodates this readily, because it treats each stage as a test applied to a transaction rather than to a relationship; a doctrinal account organized around the moment of contract formation does not.

5.2. Extension as a Sanction, and the Outer Limit of Abuse

Extension of the period for breach of the information duty expresses consumer law's limit on formal autonomy. The logic is that a trader may not profit from its own informational default. Where the consumer does not know that the right exists, requiring its exercise within the ordinary period lacks justification. Non-disclosure therefore does more than generate administrative liability or damages; it changes the moment at which the contract enters a stable, non-revocable state. This linkage between the information duty and temporal effect is the structural feature of the cooling-off regime most deserving of attention in cross-border electronic contracting, because it is the mechanism by which a stage-one failure is converted into a stage-three remedy without the consumer having to litigate the adequacy of disclosure as such.

Until 2026 the extended period had no clearly articulated outer limit. Eisenberger Gerüstbau supplies one: although consumers benefit from a broad right of withdrawal, its exercise may in exceptional circumstances be abusive—as where the consumer withdraws at the end of an extended period after the services have already been fully performed, with the intention of avoiding payment. The ruling should be read narrowly, and the Court framed it in terms of exceptional circumstances rather than as a general proportionality review of withdrawal decisions. But it completes the structure in a way the earlier literature had left open. The sanction for non-disclosure is not unbounded; it is bounded by the purpose of the right, which is to restore an opportunity for evaluation rather than to supply a costless option over fully performed services.

For cross-border traders the problem extends into conflict of laws. A platform may design a uniform withdrawal period by reference to the law of its place of establishment while directing its activities at markets with higher protective standards. A uniform period in the platform's terms cannot of itself displace the mandatory protection potentially applicable at the consumer's habitual residence, which is the subject of Section 7. What Eisenberger Gerüstbau adds is that the reverse inference is also unavailable: the consumer cannot convert an extended period into an indefinite entitlement regardless of performance.

6. Stage Four: From Disclosure to Operability

6.1. The Asymmetry of Entry and Exit

Digital consumer protection is moving from the provision of information to the operability of rights. Traditional rules tend to assume that once the trader has disclosed the right, the consumer can locate and exercise it; the platform environment shows that the operating cost of a right matters as much as its existence. Complex account menus, concealed support entry points, cancellation processes available only by telephone, and the asymmetric design in which contracting requires one click while withdrawing requires many steps all materially weaken cooling-off protection.

This asymmetry is not incidental but measurable and, on the evidence, deliberate. A large-scale crawl of shopping websites documented the prevalence of manipulative interface practices and classified them by mechanism (Mathur et al., 2019); experimental work established that such practices substantially change consumer decisions, with aggressive variants producing very large effects on acceptance of unwanted terms and mild variants producing measurable ones (Luguri & Strahilevitz, 2021). Procedural fairness in digital contracts therefore cannot be measured by the content of terms alone; it must take in the interface architecture of entry, confirmation, retention and exit. Where a consumer can complete a purchase in seconds but cannot withdraw at comparable cost, the platform has reallocated exit costs at the technical level, and the reallocation is invisible to any review confined to the text of the contract. Article 25 of the Digital Services Act now addresses interface design directly, and its arrival marks the point at which European Union law began to treat the interface, rather than the document, as the regulated object (Cauffman & Goanta, 2021).

6.2. The Electronic Withdrawal Function and the Comparative Divergence of 2024–2026

The most consequential development is the mandatory electronic withdrawal function. Directive (EU) 2023/2673 inserts Article 11a into Directive 2011/83/EU, requiring traders who conclude distance contracts with consumers through an online interface to provide a dedicated withdrawal function labeled 'withdraw from contract here' or in unambiguous equivalent wording, continuously available throughout the withdrawal period, prominently displayed and easily accessible. The obligations apply from 19 June 2026. The animating principle is explicitly one of symmetry: withdrawal must not be more burdensome than conclusion, so that a contract concluded with a click can be withdrawn with a click. This is the functional standard adopted as positive law, and it converts stage four from an evaluative criterion into an enforceable obligation.

Automatic renewal is the setting in which the operability problem bites hardest, and here the comparative picture has diverged sharply. In European Union law, Sofatutor holds that where a subscription begins with a free period and renews automatically unless terminated, the withdrawal right is in principle guaranteed only once, provided the trader informed the consumer in a clear, comprehensible and explicit manner that the contract would become chargeable after the free period; only where that information was not given does a second right arise. The right is thus rationed in exchange for a disclosure standard—a trade-off that only holds if the disclosure standard is enforced, which returns the analysis to stage one.

The United Kingdom has legislated a more prescriptive regime and then deferred it. Sections 264 to 266 of the Digital Markets, Competition and Consumers Act 2024 confer cooling-off rights on subscription contracts, requiring straightforward exit, exit online where the consumer signed up online, and a fourteen-day cooling-off period after a trial or a contract of twelve months or more renews. Chapter 4 of Part 4, on alternative dispute resolution, was commenced in April 2026, but the subscription regime itself has not been brought into force and is now expected in 2027. The United States moved in the opposite direction: the Federal Trade Commission's Negative Option Rule, which would have imposed comparable click-to-cancel requirements, was vacated in its entirety in Custom Communications days before enforcement was to begin, on the ground that the Commission had omitted the preliminary regulatory analysis required for rules with an annual economic effect of at least USD 100 million; a fresh rulemaking was initiated in January 2026. China has taken a third route, addressing automatic renewal through a general attention-drawing duty on operators rather than a prescribed exit mechanism, under the Implementing Regulation on the Protection of Consumer Rights and Interests in force since 1 July 2024.

Three points follow for the framework. First, stage-four protection is where regulatory divergence is now greatest, which is precisely the stage at which a single global interface must be configured; the cross-border trader cannot satisfy all four regimes with one design, and will tend to converge on the most demanding, which at present is the European Union's. Second, the vacatur in Custom Communications is instructive on the mechanism rather than the merits: the rule fell because of the cost of compliance it would have imposed, assessed procedurally, not because click-to-cancel was held to be substantively unjustified. Stage-four obligations are, of all the links in the chain, the most exposed to cost-based challenge, which makes the economic evidence in Section 8 directly relevant to their defensibility. Third, the European Union's own next step is already in view: the Digital Fairness Act, announced in the Consumer Agenda 2030 and expected as a legislative proposal in late 2026, is directed at manipulative interface design, addictive design and unfair personalization, and would extend regulation further into the territory identified here as stage four.

7. Stage Five: Cross-Border Enforceability

7.1. Choice of Law: Article 6 Rome I as an Objective Rule

Cross-border B2C electronic contracts routinely contain applicable-law and forum-selection clauses. For the trader, uniformly selecting the law of its headquarters reduces compliance cost; for the consumer, such clauses can create the mistaken impression that rights may be asserted only under foreign law. Article 6 of Rome I provides that where the trader pursues commercial activities in, or directs them to, the country of the consumer's habitual residence, the consumer contract is specially protected: even where the parties have chosen another law, that choice cannot deprive the consumer of the protection of provisions that cannot be derogated from by agreement under the law that would otherwise apply.

Club La Costa clarified the character of that protection in a way that matters for the framework. The Court treated the law designated by Article 6 as an objective conflict-of-laws rule rather than as a defense available only if the consumer invokes it. The distinction is not merely technical. If mandatory protection operated as a pleaded defense, stage five would depend on the consumer knowing that the protection existed and asserting it—reintroducing at the enforcement stage exactly the informational dependency that stages one and two are meant to resolve. Treating Article 6 as objective removes that dependency, at least in principle, and makes the withdrawal right of the consumer's home market applicable irrespective of the consumer's legal sophistication.

Two qualifications should be recorded. First, Article 6 protects only against derogation; it does not displace the chosen law in its entirety, so a court must still identify which home-state rules are non-derogable, and the answer for withdrawal rights, while clear in European Union law, is not uniform elsewhere. Second, the protection is conditional on the trader having directed activities at the consumer's state, which is a question of fact that a platform can influence through the design of its market-facing interfaces. Stage-five protection is therefore not automatic; it is contingent on a characterization that the trader partly controls.

7.2. Jurisdiction, Targeted Activity and Click-Wrapped Forum Clauses

Jurisdiction follows a comparable logic. Brussels I bis establishes special rules for qualifying consumer contracts, permitting the consumer to sue in the courts of their own domicile and restricting the effect of pre-dispute forum agreements, and it likewise turns on whether the trader directed activities to the consumer's state. Club La Costa also addressed the determination of a company's domicile for these purposes, confirming that the analysis attaches to the legal person rather than to corporate-group arrangements. A cross-border platform accordingly cannot, through a foreign-court clause in its terms of service, transfer all procedural cost to the consumer. A substantive withdrawal right that can be realized only through distant, expensive litigation in an unfamiliar language is not effective protection, whatever its formal status.

The treatment of click-wrapped forum clauses illustrates how uneven this protection is in practice. In El Majdoub the Court held that accepting general terms containing a jurisdiction clause by click-wrapping satisfies the formal requirement of a durable record, and that the record need not actually have been printed or saved—it need only be possible to produce one. The ruling is commonly cited for the proposition that click-wrap suffices to conclude a forum agreement, but its scope is confined to business-to-business relations and to the formal validity of the clause. It says nothing about whether such a clause would bind a consumer, where the special regime and the unfair-terms control under Directive 93/13/EEC apply. The distinction is worth stating explicitly because the elision of it in practice is one route by which consumers are led to believe that a click-accepted forum clause is binding on them when it is not.

7.3. The Clause Bundle and the Evidence on Actual Compliance

The core 'conflict of effect' is thus not merely that two states prescribe different withdrawal periods. It is the conflict between platform private ordering and mandatory consumer protection. Choice-of-law clauses, jurisdiction clauses, immediate-performance clauses and refund restrictions can operate together within a single set of standard terms, and the doctrinal literature on unfair terms in contracts with online service providers has long argued that the enforcement response to such clause bundles is inadequate to the scale of the problem (Loos & Luzak, 2016). Cross-border B2C contracts should therefore be subjected to functional review asking whether the combination of clauses substantially reduces the consumer's ability to learn of, exercise or enforce the right—rather than assessing each clause in isolation, which is how such bundles survive.

The empirical evidence indicates that the gap between formal and effective protection at this stage is wide. A study coding one hundred sets of online terms of service found extensive non-compliance with European Union consumer law among foreign service providers, including clauses that European Union law renders unenforceable (Pałka et al., 2025). That finding is important for the framework in a specific way: it shows that stage-five protection frequently fails not because the conflict rules are defective but because unenforceable terms remain in circulation, shaping consumer beliefs and deterring claims even though a court would not uphold them. Unenforceability is a remedy available to the consumer who litigates; it is not a remedy for the consumer who reads the clause and concludes that nothing can be done. Comparative scholarship on transnational consumer protection has made the parallel point that private-law entitlements and public enforcement capacity have diverged in the digital single market (Meskic et al., 2022; Targański, 2025).

8. Economic Analysis: What the Protective Chain Costs and What It Buys

8.1. The Efficiency Case for a Mandatory Cooling-Off Right

The law-and-economics literature on cooling-off periods identifies a narrow but real efficiency case. Mandatory withdrawal rights can be justified where informational asymmetry is severe, where inspection before contracting is impossible, and where the seller's superior information about product fit makes private contracting over return rights unreliable; outside those conditions, the mandatory character of the right imposes costs without corresponding informational gain (Rekaiti & Van den Bergh, 2000). More recent analysis reaches a similar conclusion by a different route, treating the right as a device that corrects a decision-making defect rather than an informational one, and noting that its welfare effects depend on how far consumers actually use it (Karampatzos & Ilić, 2023). Distance selling satisfies the informational conditions almost by definition, which is why the institution has proved durable.

The broader digital-economics framework explains why the analysis cannot stop at the substantive right. Digitization lowers search, replication, transportation, tracking and verification costs, and the economic consequences of a regulatory intervention in digital markets depend on which of these costs it alters (Goldfarb & Tucker, 2019). A withdrawal right operates principally on verification cost: it lets the consumer verify fit after contracting rather than before. Interpreted this way, the five stages of the protective chain are not equally valuable. Stages one to three reduce the consumer's verification cost; stage four reduces the transaction cost of acting on what verification reveals; stage five determines whether either reduction survives the crossing of a border. A regime that invests heavily in disclosure while leaving exit costly has improved information without improving the ability to act on it, which is the specific inefficiency that Article 11a addresses.

8.2. Evidence from the Economics of Return Policies

The most directly relevant empirical evidence comes from the economics of return policies, which are the commercial analogue of statutory withdrawal rights. A meta-analytic review of the effect of return-policy leniency found that leniency increases purchasing and also increases returning, with the purchase effect generally the larger of the two, and that the components of leniency differ in their effects: leniency in time and effort tends to raise purchasing without proportionately raising returns, whereas leniency in money and scope raises both (Janakiraman et al., 2016). The distinction maps onto the framework with unusual precision. Stage-four reforms are reforms to effort leniency—they make exercising the right easier without widening its substantive scope—and are therefore the category for which the evidence predicts a favorable ratio of demand gain to return cost.

This is the strongest economic argument available for the electronic withdrawal function, and it is worth stating plainly because the regulatory debate has not made it. Article 11a reduces the effort cost of exit without altering the fourteen-day period or the categories of excluded contracts. On the return-policy evidence, that is the configuration most likely to expand transactions by more than it expands withdrawals. The countervailing consideration is that returns are genuinely costly to handle and that the cost falls unevenly across channels and product categories, so that firms with weak reverse logistics bear disproportionate burdens (Ofek et al., 2011). The distributional implication is that a uniform stage-four obligation is not uniform in incidence: it is cheapest for large platforms with established returns infrastructure and dearest for small cross-border sellers, which is a competition consequence that the European Union's impact assessments have tended to treat as secondary.

8.3. Distance, Trust and Cross-Border Online Trade

The premise of the entire inquiry is that cross-border B2C trade is substantial enough for its contractual rules to matter, and that it remains suppressed by frictions that law can partly address. Both propositions have empirical support. Online marketplaces demonstrably reduce the effect of distance on trade: analysis of eBay transaction data found the distance elasticity of online trade to be substantially smaller than that of offline trade, with the reduction attributable largely to lower information and search frictions (Lendle et al., 2016). Earlier work had already established that online trade does not escape gravity altogether but is less bound by it, and that the residual distance effect reflects information costs rather than transport costs alone (Blum & Goldfarb, 2006; Hortaçsu et al., 2009).

Trust and legal uncertainty are identifiable components of the residual friction. Analysis of cross-border online trade within the European Union found that language, trust and the perceived reliability of dispute resolution were significant impediments alongside distance and logistics, and that the effects were larger for cross-border than for domestic online transactions (Gomez-Herrera et al., 2014). Systematic review of consumer behavior in cross-border electronic commerce confirms perceived risk and uncertainty over redress as recurrent determinants of purchase intention (Chen et al., 2023). Consumer-facing scholarship on the decision to rescind a purchase points in the same direction: the availability of a credible exit affects the willingness to commit in the first place (Sparks et al., 2014). These findings supply the demand-side rationale for stage-five protection. If consumers discount cross-border offers because they doubt that rights will be enforceable, then credible enforceability is not only a protective measure but a trade-facilitating one, and the benefit accrues to compliant traders as well as to consumers.

8.4. The Cost of Compliance-Heavy Digital Regulation: Evidence from the GDPR

Against this must be set the evidence on what compliance-heavy digital regulation actually costs, and here the General Data Protection Regulation offers the closest available natural experiment. The findings are consistent across independent studies and are not favorable to an assumption that protective digital regulation is low-cost. Evaluation of European web traffic and e-commerce outcomes found measurable reductions in recorded website use and in e-commerce activity following the Regulation's application, with effects concentrated in outcomes dependent on third-party data (Goldberg et al., 2024). Vendor-level analysis found that the web-technology market became more concentrated, with the relative position of the largest vendor improving as smaller vendors lost share (Johnson et al., 2023). Cross-border analysis found substantial regulatory spillovers, including reductions in the volume of data collected and the withdrawal of some non-European Union firms from European markets (Peukert et al., 2022). Venture financing for European technology firms declined relative to comparable United States firms in the period following implementation (Jia et al., 2021).

Three cautions are needed before transposing these findings. First, the Regulation is a far more demanding compliance instrument than Article 11a: it imposes continuing organisational obligations, whereas a withdrawal function is a bounded interface change. Second, the identified costs fell disproportionately on data-dependent intermediation rather than on direct consumer-facing contracting. Third, the studies measure costs rather than net welfare, and none purports to establish that the Regulation was welfare-reducing overall. The transposable lesson is narrower and concerns incidence rather than magnitude: compliance-heavy digital rules tend to advantage large incumbents over small entrants, and to induce some foreign firms to exit rather than comply. Both effects are directly relevant to stage five, because the consumers least protected in practice are those transacting with traders who have chosen non-compliance over either compliance or exit—the population that Pałka et al. (2025) document.

The Custom Communications vacatur should be read in this light. The Negative Option Rule fell on the ground that the Federal Trade Commission had not performed the preliminary regulatory analysis required where a rule's annual economic effect reaches USD 100 million. Whatever one makes of the procedural holding, the episode establishes that stage-four obligations are the most exposed of the five to cost-based legal challenge, because their compliance cost is concentrated, quantifiable and borne by identifiable firms, whereas their benefits are diffuse. Regulators pursuing operability requirements therefore have a strong instrumental reason to quantify benefits using evidence of the kind discussed in Section 8.2, and the absence of such quantification is a vulnerability rather than a detail.

8.5. The Friction Shock of 2026

The cost structure of cross-border B2C commerce is being reset as this article is written, and the interaction with withdrawal rights has not been examined. Following political agreement among Member States in November 2025, the European Union removed the EUR 150 customs duty relief threshold with effect from 1 July 2026, applying an interim flat duty per item pending full classification-based treatment when the Customs Data Hub becomes operational, with an e-commerce handling fee to follow. The measure is the most significant change to low-value import treatment since the customs union was established.

The economic literature indicates that the welfare stakes are considerable and unevenly distributed. Analysis of de minimis imports finds that low-value shipments are disproportionately consumed by lower-income households, so that raising the effective tariff on such shipments is regressive in incidence (Fajgelbaum & Khandelwal, 2024). The interaction with the protective chain is direct and, so far as can be determined, unremarked. First, per-item duties and handling fees raise the cost of the reverse logistics on which withdrawal depends: a consumer exercising a right of withdrawal on a low-value cross-border item may face a refund that is small relative to the duties and fees already incurred, which weakens stage four economically even where Article 11a makes it available procedurally. Second, the two measures take effect within eleven days of one another—19 June and 1 July 2026—so that the same cross-border sellers absorb an interface-compliance cost and a tariff cost simultaneously. Third, if duty and fee treatment is not disclosed in the total price presented before contracting, the consumer's stage-one information is incomplete in precisely the respect that determines whether withdrawal is worth exercising.

That third point connects to a regulatory development on the other side of the Atlantic. The Federal Trade Commission's Rule on Unfair or Deceptive Fees, effective 12 May 2025, requires that the total price inclusive of all mandatory charges be disclosed at least as prominently as any other price, although its scope is confined to live-event ticketing and short-term lodging rather than extending to goods commerce generally. The instrument is nonetheless instructive as a model: it regulates the completeness of the price signal at the point of decision rather than the permissibility of the underlying charges. Applied to cross-border B2C goods after the removal of the de minimis threshold, an equivalent total-landed-cost disclosure duty would address a stage-one gap that the European Union acquis does not currently close, and would do so without altering any substantive withdrawal entitlement.

8.6. Implications: Withdrawal Rights as a Priced Option

Bringing the evidence together suggests that the withdrawal right is best conceived, economically, as an option that the consumer holds and that someone pays for. The option has a value to the consumer that rises with informational asymmetry and with the cost of pre-contractual inspection, and a cost to the trader that rises with the probability of exercise and the expense of handling what is returned. Three consequences follow for institutional design.

First, the efficient way to strengthen the institution is to reduce the cost of exercising the option rather than to widen its scope, because the return-policy evidence indicates that effort leniency expands demand more than it expands returns while money and scope leniency expand both (Janakiraman et al., 2016). Article 11a is, on this analysis, well-targeted; proposals to lengthen the fourteen-day period are not. Second, exceptions structured as cliffs are economically inferior to exceptions structured as prices, for the reason given in Section 4.3: a cliff makes the marginal return to obtaining a consent click discontinuous and large, which is what generates the manipulative design that stage-two rules then have to police. The drift of the case law from the Article 16(m) exception towards Article 14(3) proportionate compensation is therefore welfare-improving as well as doctrinally coherent. Third, stage-five enforceability should be understood as a trade-facilitating investment rather than a pure protective cost, since the frictions it addresses are the same information and trust frictions that the trade literature identifies as suppressing cross-border online transactions (Gomez-Herrera et al., 2014; Lendle et al., 2016).

The analysis also identifies where the economic case is weakest. A uniform operability obligation has non-uniform incidence, falling hardest on small cross-border sellers without established reverse logistics (Ofek et al., 2011), and the GDPR evidence suggests that such obligations can consolidate markets around large incumbents and induce marginal foreign traders to exit rather than comply (Johnson et al., 2023; Peukert et al., 2022). A regime that raises compliance cost while leaving enforcement against non-compliant foreign traders weak risks the worst combination: compliant traders bear the cost, non-compliant traders retain the market, and consumers transacting with the latter gain nothing. Stage four and stage five must therefore advance together, and the case for the withdrawal function is strongest where cross-border enforcement capacity is strongest.

9. Discussion

9.1. Contribution and the Functional Withdrawal-Protection Framework

The principal contribution of this article is to relocate the cooling-off period from the category of isolated ex post remedy into the process of electronic contract formation. Traditional analysis treats information duties as a pre-formation question, the withdrawal right as a post-formation question, and applicable law and jurisdiction as questions arising only once a dispute has crystallized. Cross-border platform transactions show that these three phases are technically connected by one interface and one set of standard terms. What the consumer sees when ordering, what is clicked, how the loss of rights is communicated, whether a withdrawal facility can later be found, and which law governs a dispute together determine the real strength of the contractual commitment.

The five-stage framework consolidates this into an assessable structure. Stage one is disclosure before the consumer is bound, in a form that is clear, accessible and retainable. Stage two is understanding and valid consent, with a strict separation between agreement to contract, request for early performance and acknowledgment of the loss of the right. Stage three is commencement, matched to transaction type and to the point at which evaluation becomes possible, and now recurring whenever a distance addendum is agreed. Stage four is exercise through a low-friction, evidentially secure channel. Stage five is cross-border enforceability, under which applicable-law and jurisdiction clauses cannot remove mandatory protection. Because the stages are sequential, protection is bounded by the weakest link, and the framework therefore predicts that investment in the strongest link yields no protective return.

Read against the 2023–2026 developments, the framework also yields a diagnosis of where each system now stands. European Union law has moved decisively on stages two and four—narrowing the digital-content exception, converting it towards proportionate compensation, and making the withdrawal function mandatory—while leaving the evidential dimension of stage one unaddressed and stage-five enforcement dependent on private litigation. The United Kingdom has legislated for stage four and not commenced it. The United States has had its stage-four rule vacated on cost-procedure grounds while regulating price completeness at stage one in a narrow sector. China has addressed stage one for automatic renewal through a general attention duty without prescribing a stage-four mechanism. No system is strong across all five stages, and the pattern of weakness differs by system rather than by degree of consumer protection generally.

Three implications for institutional design follow. For traders, compliance should not end with stating the right in the terms: it should extend to retaining verifiable disclosure records, identifying clearly the conditions under which the right is lost, and providing an exit path commensurate with the complexity of the entry path. For regulators, the objects of attention should be whether interfaces create unnecessary friction, whether choice-of-law clauses mislead consumers about the protection available to them, and whether platforms use cross-border structure to lower the probability of effective redress—the last being the point at which stage-four and stage-five obligations must be advanced together rather than separately. For judicial review, the shift should be from the validity of individual clauses to the aggregate effect of the clause bundle and the interface flow, an approach for which the notice-and-assent reasoning in Berman, Chabolla and Parker-Grennan already provides usable materials.

9.2. Research Agenda

The framework generates a research agenda with four strands. The first is empirical measurement of stage-four compliance after 19 June 2026. Article 11a supplies an unusually clean natural experiment: a mandatory, observable, date-certain interface change applying to traders serving European Union consumers but not to the same traders' non-European Union interfaces. The design permits estimation of the effect of exit-cost reduction on both purchasing and withdrawal rates, which is the quantity the return-policy literature predicts but has not measured for a statutory right.

The second is the classification problem created by Sky Österreich. Determining which digital offerings are dynamic enough to constitute services is, at present, a question answered contract by contract. Coding a representative sample of subscription offerings against the Court's indicia would establish how much of the digital subscription market has in fact lost the benefit of the exception, and whether traders are reclassifying or simply abandoning the waiver.

The third is behavioural. Whether disclosure improves comprehension remains contested, and the literature on mandated disclosure is skeptical (Ben-Shahar & Schneider, 2014). The specific untested question is whether the separation of consent, early-performance request and rights-loss acknowledgment required at stage two actually produces differential comprehension, or whether it merely adds a third click. Experimental work here would bear directly on whether the enhanced-consent structure is doing the work assigned to it.

The fourth concerns stage-five penetration. Pałka et al. (2025) establish widespread circulation of terms that European Union law renders unenforceable; what is not known is how far such terms alter consumer behavior. Measuring the deterrent effect of unenforceable clauses—as distinct from their legal validity—would quantify the gap between formal and effective protection, and would indicate whether the appropriate response is enhanced private enforcement, public enforcement, or a disclosure duty directed at the enforceability of the terms themselves.

9.3. Limitations

Four limitations should be stated. First, the article is a doctrinal and comparative synthesis; it does not code platform terms at scale, and its claims about platform practice rest on the empirical work of others. Second, European Union law serves as the principal reference point because its rule system is the most complete, which risks presenting as a general structure what may be a regional one; the treatment of no-reasons withdrawal, digital content and conflict of laws is not uniform across jurisdictions, and the international framework expressly leaves B2C contracting to domestic law.

Third, several of the developments analyzed are too recent for their effects to be observed. Article 11a applies from 19 June 2026, the removal of the de minimis threshold from 1 July 2026, and the United Kingdom's subscription regime is not yet in force; the analysis of these measures is therefore about institutional design and predictable incidence, not about realized outcomes. Fourth, the economic evidence marshaled in Section 8 is transposed from adjacent settings—commercial return policies, online trade costs and data protection compliance—rather than drawn from studies of withdrawal rights themselves, which do not yet exist at comparable quality. The direction of the inferences is well supported; their magnitude is not, and they should be treated as identifying the sign of an effect rather than its size.

10. Conclusion

The cooling-off problem in cross-border B2C electronic contracts is not the narrow question whether consumers have fourteen days in which to change their minds. It is a question of contractual effect shaped jointly by information, consent, time, interface and conflict of laws. The right of withdrawal acquires substantive meaning only where the consumer obtains comprehensible information about it before contracting, makes an independent and express confirmation when waiving it, can identify when the period begins, can complete a withdrawal at reasonable cost, and does not lose mandatory protection through choice-of-law or jurisdiction clauses in a cross-border dispute.

The developments of 2023 to 2026 have moved the law of the European Union decisively towards this functional conception. Sky Österreich has narrowed the digital-content exception by reference to the very features—dynamism and personalization—that characterize contemporary platform offerings; Eisenberger Gerüstbau has extended the notion of the distance contract to addenda while marking out an outer limit of abuse; and Article 11a has converted the operability of the right from an academic criterion into a legal obligation from 19 June 2026. The comparative record is more equivocal: the United Kingdom has legislated and waited, the United States has seen its equivalent rule vacated on cost-procedure grounds, and China has relied on a general attention duty. Convergence should not be assumed.

Cross-border digital consumer protection should accordingly move from the formal conferral of rights to the functional realizability of rights. The framework proposed here—disclosure, understanding and valid consent, commencement, exercise, and cross-border enforceability—links consumer contract law, digital platform governance and conflict of laws into a single protective structure whose strength is set by its weakest element. The economic evidence indicates that this is not merely a protective program: reducing the cost of exercising withdrawal rights is the intervention most likely to expand transactions by more than it expands returns, and credible cross-border enforceability addresses the same information and trust frictions that suppress cross-border online trade. The central task for future rulemaking is therefore not to add further disclosure text, but to ensure that consumers in digital markets can see, understand, use and retain their rights across borders—and to advance operability and enforceability together, since neither delivers protection on its own.

References

Bakos, Y., Marotta-Wurgler, F., & Trossen, D. R. (2014). Does anyone read the fine print? Consumer attention to standard-form contracts. The Journal of Legal Studies, 43(1), 1–35. https://doi.org/10.1086/674424

Ben-Shahar, O., & Schneider, C. E. (2014). More than you wanted to know: The failure of mandated disclosure. Princeton University Press. https://doi.org/10.23943/princeton/9780691161709.001.0001

Blum, B. S., & Goldfarb, A. (2006). Does the internet defy the law of gravity? Journal of International Economics, 70(2), 384–405. https://doi.org/10.1016/j.jinteco.2005.10.002

Cauffman, C., & Goanta, C. (2021). A new order: The Digital Services Act and consumer protection. European Journal of Risk Regulation, 12(4), 758–774. https://doi.org/10.1017/err.2021.8

Chen, J., Lan, Y., & Chang, Y. (2023). Consumer behaviour in cross‐border e‐commerce: Systematic literature review and future research agenda. International Journal of Consumer Studies, 47(6), 2609–2669. https://doi.org/10.1111/ijcs.12969

Fajgelbaum, P., & Khandelwal, A. (2024). The value of de minimis imports. NBER Working Paper No. 32607. https://doi.org/10.3386/w32607

Giannikis, G. K., & Daskalopulu, A. (2011). Normative conflicts in electronic contracts. Electronic Commerce Research and Applications, 10(2), 247–267. https://doi.org/10.1016/j.elerap.2010.09.005

Goetzinger, P., & Spremić, M. (2025). Cross-border e-commerce and its implications for consumer protection (2019–2024): A systematic literature review. European Journal of Economics, 5(1), 1–7. https://doi.org/10.33422/eje.v5i1.972

Goldberg, S. G., Johnson, G. A., & Shriver, S. K. (2024). Regulating privacy online: An economic evaluation of the GDPR. American Economic Journal: Economic Policy, 16(1), 325–358. https://doi.org/10.1257/pol.20210309

Goldfarb, A., & Tucker, C. (2019). Digital economics. Journal of Economic Literature, 57(1), 3–43. https://doi.org/10.1257/jel.20171452

Gomez-Herrera, E., Martens, B., & Turlea, G. (2014). The drivers and impediments for cross-border e-commerce in the EU. Information Economics and Policy, 28, 83–96. https://doi.org/10.1016/j.infoecopol.2014.05.002

Helberger, N., Loos, M. B. M., Guibault, L., Mak, C., & Pessers, L. (2013). Digital content contracts for consumers. Journal of Consumer Policy, 36(1), 37–57. https://doi.org/10.1007/s10603-012-9201-1

Hortaçsu, A., Martínez-Jerez, F. A., & Douglas, J. (2009). The geography of trade in online transactions: Evidence from eBay and MercadoLibre. American Economic Journal: Microeconomics, 1(1), 53–74. https://doi.org/10.1257/mic.1.1.53

Jabłonowska, A., & Tagiuri, G. (2023). Rescuing transparency in the digital economy: In search of a common notion in EU consumer and data protection law. Yearbook of European Law, 42, 347–387. https://doi.org/10.1093/yel/yead005

Janakiraman, N., Syrdal, H. A., & Freling, R. (2016). The effect of return policy leniency on consumer purchase and return decisions: A meta-analytic review. Journal of Retailing, 92(2), 226–235. https://doi.org/10.1016/j.jretai.2015.11.002

Jia, J., Jin, G. Z., & Wagman, L. (2021). The short-run effects of the General Data Protection Regulation on technology venture investment. Marketing Science, 40(4), 661–684. https://doi.org/10.1287/mksc.2020.1271

Johnson, G. A., Shriver, S. K., & Goldberg, S. G. (2023). Privacy and market concentration: Intended and unintended consequences of the GDPR. Management Science, 69(10), 5695–5721. https://doi.org/10.1287/mnsc.2023.4709

Karampatzos, A., & Ilić, N. (2023). Law and economics of the withdrawal right in EU consumer law. Review of Law & Economics, 19(3), 435–452. https://doi.org/10.1515/rle-2022-0076

Kolotylo-Kulkarni, M., Xia, W., & Dhillon, G. (2021). Information disclosure in e-commerce: A systematic review and agenda for future research. Journal of Business Research, 126, 221–238. https://doi.org/10.1016/j.jbusres.2020.12.006

Lendle, A., Olarreaga, M., Schropp, S., & Vézina, P. L. (2016). There goes gravity: eBay and the death of distance. The Economic Journal, 126(591), 406–441. https://doi.org/10.1111/ecoj.12286

Loos, M., & Luzak, J. (2016). Wanted: A bigger stick. On unfair terms in consumer contracts with online service providers. Journal of Consumer Policy, 39(1), 63–90. https://doi.org/10.1007/s10603-015-9303-7

Luguri, J., & Strahilevitz, L. J. (2021). Shining a light on dark patterns. Journal of Legal Analysis, 13(1), 43–109. https://doi.org/10.1093/jla/laaa006

Luzak, J. A. (2014). To withdraw or not to withdraw? Evaluation of the mandatory right of withdrawal in consumer distance selling contracts taking into account its behavioural effects on consumers. Journal of Consumer Policy, 37(1), 91–111. https://doi.org/10.1007/s10603-013-9249-6

Mathur, A., Acar, G., Friedman, M. J., Lucherini, E., Mayer, J., Chetty, M., & Narayanan, A. (2019). Dark patterns at scale: Findings from a crawl of 11K shopping websites. Proceedings of the ACM on Human-Computer Interaction, 3(CSCW), 1–32. https://doi.org/10.1145/3359183

Meskic, Z., Albakjaji, M., Omerovic, E., & Alhussein, H. (2022). Transnational consumer protection in e-commerce. International Journal of Service Science, Management, Engineering, and Technology, 13(1), 1–15. https://doi.org/10.4018/ijssmet.299972

Ofek, E., Katona, Z., & Sarvary, M. (2011). “Bricks and clicks”: The impact of product returns on the strategies of multichannel retailers. Marketing Science, 30(1), 42–60. https://doi.org/10.1287/mksc.1100.0588

Pałka, P., Porębski, A., Pałosz, R., & Wiśniewska, K. (2025). Do foreign service providers simply ignore the EU consumer law? An empirical study of 100 online terms of service. Journal of Consumer Policy, 48(3), 233–268. https://doi.org/10.1007/s10603-024-09580-z

Peukert, C., Bechtold, S., Batikas, M., & Kretschmer, T. (2022). Regulatory spillovers and data governance: Evidence from the GDPR. Marketing Science, 41(4), 746–768. https://doi.org/10.1287/mksc.2021.1339

Rekaiti, P., & Van den Bergh, R. (2000). Cooling-off periods in the consumer laws of the EC Member States. A comparative law and economics approach. Journal of Consumer Policy, 23(4), 371–408. https://doi.org/10.1023/a:1007203426046

Sparks, B. A., Bradley, G. L., Jennings, G. R., & Johnston, N. R. (2014). Cooling off and backing out: Understanding consumer decisions to rescind a product purchase. Journal of Business Research, 67, 2903–2910. https://doi.org/10.1016/j.jbusres.2012.06.003

Steennot, R. (2013). The right of withdrawal under the Consumer Rights Directive as a tool to protect consumers concluding a distance contract. Computer Law & Security Review, 29(2), 105–119. https://doi.org/10.1016/j.clsr.2013.01.005

Targański, B. (2025). The legal aspects of consumer protection in cross-border e-commerce. In A. Grudecka & K. Kozłowski (Eds.), Digital internationalisation of firms (pp. 223–241). Routledge. https://doi.org/10.4324/9781003617396-14

Veltri, G. A., Lupiáñez-Villanueva, F., Folkvord, F., Theben, A., & Gaskell, G. (2020). The impact of online platform transparency of information on consumers’ choices. Behavioural Public Policy, 7(1), 55–82. https://doi.org/10.1017/bpp.2020.11

Zhou, L., Wang, W., Xu, J., Liu, T., & Gu, J. (2018). Perceived information transparency in B2C e-commerce: An empirical investigation. Information & Management, 55(7), 912–927. https://doi.org/10.1016/j.im.2018.04.005

Legislation, Case Law and Official Documents

Council Directive 93/13/EEC of 5 April 1993 on unfair terms in consumer contracts. http://data.europa.eu/eli/dir/1993/13/oj

Regulation (EC) No 593/2008 of the European Parliament and of the Council of 17 June 2008 on the law applicable to contractual obligations (Rome I). http://data.europa.eu/eli/reg/2008/593/oj

Directive 2011/83/EU of the European Parliament and of the Council of 25 October 2011 on consumer rights (Consumer Rights Directive). http://data.europa.eu/eli/dir/2011/83/oj

Regulation (EU) No 1215/2012 of the European Parliament and of the Council of 12 December 2012 on jurisdiction and the recognition and enforcement of judgments in civil and commercial matters (Brussels I bis). http://data.europa.eu/eli/reg/2012/1215/oj

Directive (EU) 2019/770 of the European Parliament and of the Council of 20 May 2019 on certain aspects concerning contracts for the supply of digital content and digital services. http://data.europa.eu/eli/dir/2019/770/oj

Regulation (EU) 2022/2065 of the European Parliament and of the Council of 19 October 2022 on a Single Market for Digital Services (Digital Services Act). http://data.europa.eu/eli/reg/2022/2065/oj

Directive (EU) 2023/2673 of the European Parliament and of the Council of 22 November 2023 amending Directive 2011/83/EU as regards financial services contracts concluded at a distance and repealing Directive 2002/65/EC. http://data.europa.eu/eli/dir/2023/2673/oj

Digital Markets, Competition and Consumers Act 2024 (c. 13) (United Kingdom), ss. 253–275 (subscription contracts), commenced in part by SI 2026/284. https://www.legislation.gov.uk/ukpga/2024/13

Regulation on the Implementation of the Law of the People's Republic of China on the Protection of Consumer Rights and Interests (State Council, promulgated 19 March 2024, in force 1 July 2024). https://english.www.gov.cn/policies/latestreleases/202403/19/content_WS65f974b2c6d0868f4e8e53d1.html

Federal Trade Commission, Rule on Unfair or Deceptive Fees, 16 C.F.R. pt. 464 (effective 12 May 2025) (United States). https://www.ecfr.gov/current/title-16/chapter-I/subchapter-D/part-464

Federal Trade Commission, Negative Option Rule (“Click-to-Cancel”), 16 C.F.R. pt. 425, 90 Fed. Reg. (2024) (United States) (vacated by Custom Communications, Inc. v. FTC, 2025). https://www.ftc.gov/legal-library/browse/rules/negative-option-rule

United Nations Convention on the Use of Electronic Communications in International Contracts (2005), Art. 2(1)(a) (excluding contracts concluded for personal, family or household purposes). https://uncitral.un.org/en/texts/ecommerce/conventions/electronic_communications

European Commission. (2025, July). Digital Fairness Act: Call for evidence and public consultation. https://www.europarl.europa.eu/legislative-train/theme-protecting-our-democracy-upholding-our-values/file-digital-fairness-act

European Commission, Directorate-General for Taxation and Customs Union. (2025, November 13). E-commerce: EUR 150 customs duty exemption threshold to be removed as of 2026. https://taxation-customs.ec.europa.eu/news/e-commerce-150-eur-customs-duty-exemption-threshold-be-removed-2026-2025-11-13_en

Case C-234/25, Sky Österreich Fernsehen GmbH, Judgment of the Court of 9 July 2026 (streaming as a digital service; strict reading of the Art. 16(m) exception).

Case C-564/24, Eisenberger Gerüstbau GmbH v JK, Judgment of the Court of 5 March 2026 (distance contract by addendum; abuse of the right of withdrawal).

Case C-400/22, Conny GmbH, Judgment of the Court of 30 May 2024 (Art. 8(2) order-button labeling applies to conditional payment obligations).

Case C-821/21, NM v Club La Costa (UK) plc and Others, Judgment of the Court of 14 September 2023 (Art. 6 Rome I as an objective conflicts rule; Arts. 18(1) and 63 Brussels I bis).

Case C-565/22, Verein für Konsumenteninformation v Sofatutor GmbH, Judgment of the Court of 5 October 2023 (single withdrawal right in free-trial contracts that renew automatically).

Case C-249/21, Fuhrmann-2-GmbH v B., Judgment of the Court of 7 April 2022 (order button must be unambiguous on its own wording).

Case C-641/19, EU v PE Digital GmbH, Judgment of the Court of 8 October 2020 (amount payable for services supplied before withdrawal under Art. 14(3)).

Case C-322/14, Jaouad El Majdoub v CarsOnTheWeb.Deutschland GmbH, Judgment of the Court of 21 May 2015 (click-wrapping and the durable-record requirement for jurisdiction agreements).

Parker-Grennan v Camelot UK Lotteries Ltd [2024] EWCA Civ 185 (Court of Appeal of England and Wales) (incorporation of online standard terms; reasonable sufficiency of notice).

Berman v. Freedom Financial Network, LLC, 30 F.4th 849 (9th Cir. 2022) (reasonably conspicuous notice and unambiguous manifestation of assent).

Chabolla v. ClassPass, Inc., 129 F.4th 1147 (9th Cir. 2025) (sign-in wrap held unenforceable; automatic-renewal claims).

Custom Communications, Inc. v. Federal Trade Commission, No. 24-3137, 2025 WL 1873489 (8th Cir. 8 July 2025) (Negative Option Rule vacated for want of a preliminary regulatory analysis).