Navigating the new CMA consumer enforcement regime: an economic approach

Introduction

Following changes brought about by the Digital Markets, Competition and Consumers Act 2024 (“DMCC Act”), the Competition and Markets Authority (“CMA”) can now decide whether consumer protection laws have been infringed without having to take a business to court.^{1CMA, “The CMA’s approach to consumer protection” (April 2025), https://assets.publishing.service.gov.uk/media/67f37a0a22a7bab256d956de/CMA_Consumer_Approach_document.pdf} The CMA can also impose fines of up to 10% of a firm’s global turnover, and require redress for affected consumers without recourse to the courts.^{2CMA, “CMA to boost consumer and business confidence as new consumer protection regime comes into force” (7 April 2025), https://www.gov.uk/government/news/cma-to-boost-consumer-and-business-confidence-as-new-consumer-protection-regime-comes-into-force}

The law stipulates that certain commercial practices undertaken by “traders” are always deemed “unfair” (i.e., unlawful), including fake reviews and drip pricing.^{3See, for example, CMA, “Unfair Commercial practices” (18 November 2025), https://www.gov.uk/government/publications/unfair-commercial-practices-cma207/unfair-commercial-practices. A “trader” is a legal term in the UK and the EU. The UK definition is a person (“P”) acting for purposes relating to P’s business or a person acting in the name of, or on behalf of, P for purposes relating to P’s business.} However, to deem other practices unfair (such as “misleading actions and omissions”, “aggressive practices”, and “conduct that contravenes the requirements of professional diligence”), such practices must be likely to cause the “average consumer” to take a “transactional decision” that they would not have otherwise taken.^{4Op cit. Where the practice is targeted on a consumer group, the average consumer relates to the average member of the targeted group in question.}

While these terms are legal in nature (and apply both in the UK and the EU^{5The terms “average consumer” and “transactional decision” are found in Directive 2005/29/EC.}), establishing whether an impugned practice would be likely to materially distort consumer purchasing decisions is fundamentally a question of economic effect. As such, economic evidence is likely to form an important part of the answer in terms of compliance, defence against interventions, establishing the amount of redress or other remedial action, and financial penalties.

This article explains what factors may shift the dial when assessing these important issues. We explain that, while behavioural economics can be informative as regards what drives consumer decision making, its predictions are frequently ambiguous. Therefore, hard evidence on how a representative sample of consumers would behave in the specific contexts (and counterfactuals) under investigation will likely be key to the assessment. Drawing on these insights, we set out indicative steps, and evidence gathering techniques, that may be useful when firms self-assess their approach to information disclosure, or are subject to investigations into their B2C sales practices, whether under consumer or competition law.^{6Behavioural economics has featured in competition investigations. For example, the European Commission pointed to status quo bias (or default bias) as a reason why pre-installation of search engines would be valuable to search engine providers. See AT.40099 Google Android, paragraphs 812 and 851.}

Commercial practices and consumer welfare: the insights and limitations of behavioural economics

The DMCC Act distinguishes between two categories of commercial practice: those that are prohibited outright (object-style infringements), and those which require an assessment of effects.^{7There are 32 banned practices. Further, certain practices which are deemed, from a legal perspective, to amount to omission of material information from an invitation to purchase, or promotion of unfair commercial practices in a code of conduct are prohibited. CMA, “Unfair Commercial practices” (18 November 2025), https://www.gov.uk/government/publications/unfair-commercial-practices-cma207/unfair-commercial-practices#chapter1.}

This distinction can partly be grounded in insight from behavioural economics. Standard economic theory predicts that well-informed consumers make rational choices that maximise their own welfare. Behavioural economics qualifies the assumption of rationality. It recognises that consumers are subject to a range of cognitive limitations (often called “biases”) which mean that there are bounds on their rationality.^{8Some common examples are “default bias” (a tendency to stick to the default option), “loss aversion bias” (a tendency to prefer to avoid losses than acquireequivalent gains),and “information overload” (a situation where consumers are faced with so much information that it becomes difficult to make an informed decision, causing a tendency to rely on “rules of thumb” such as choosing the most salient option). See OECD, “Applying Behavioural Insights to Consumer and Competition Policy and Enforcement” (2023), page 4, https://one.oecd.org/document/DSTI/CP(2023)6/en/pdf6/en/pdf).} As such, consumers use heuristics that serve them well in familiar settings but may expose them to exploitation in others. In turn, firms that understand these cognitive patterns may (in pursuit of greater profit) design commercial practices that steer consumer choices away from decisions that would better serve their interests.^{9See Heidhues, P. and B. Köszegi “Handbook of Behavioral Economics: Applications and Foundations 1” (2018), for a detailed review.}

The law, in line with some of these insights, has identified some forms of behaviour that it considers are likely to be harmful (and rarely beneficial). For example, drip-pricing is presumed to be bad for consumers under the DMCC Act. Such conduct (whereby mandatory fees become visible only late in the purchase process) may in theory exploit a range of cognitive limitations, as set out in Box 1.

However, while there is no doubt that cognitive biases can exist, the degree of consumer irrationality is debated.^{10Some academics argue that behavioural economists are biased towards identifying biases that do not exist (or can be explained by rational behaviour). See, for example,Gigerenzer, G., “The Bias Bias in Behavioral Economics” (2018), Review of Behavioral Economics, https://pure.mpg.de/rest/items/item_3037697/component/file_3047156/content} Moreover, even where biases do exist, the challenge for policymakers is to identify whether they affect consumer behaviour systematically, i.e., which biases are displayed consistently such that they might offer scope for beneficial intervention by regulators or blanket bans on certain types of consumer-facing practices. The difficulty arises because consumers are heterogenous and so may react differently in very similar contexts.^{11See, for example, Gollwitzer, M. and Schwabe, J. “Context Dependency as a Predictor of Replicability” (2022), Review of General Psychology. The authors indicate that consumer heterogeneity can explain the difficulty of replicating outcomes in published psychological science articles, and that small changes in experimental settings can cause material changes in outcomes.} Indeed, cognitive biases can mean that well-meaning interventions (such as providing more disclosure) are ineffective^{12For example, when assessing the impact of disclosure remedies, it has been argued that bounded consumer attention can render many disclosures “useless” as consumers ignore them. See Loewenstein, G., Sunstein, C.R., and Golman, R., “Disclosure: Psychology changes everything” (2014), Annual Review of Economics, 6, pages 391–419.} or, worse, give rise to unintended consequences.^{13On the possibility of unintended consequences, see Sah, S., “The paradox of disclosure: shifting policies from revealing to resolving conflicts of interest”(2026), Behavioural Public Policy, page 3.}

As such, when assessing the impact of an impugned commercial practice, the question of how and when information is provided to customers during their purchasing journey and whether this is misleading or an act of omission is likely to be nuanced. In some circumstances, providing simple summary information may facilitate consumer decision-making, even if every detail is not set out.^{14For example, de Meza, D. E., Irlenbusch, B., and Reyniers, D. refer to a “pervasive finding… that cue competition occurs: more salient cues weaken the effects of less salient ones, and the presence of irrelevant cues causes subjects to make less use of relevant cues. Introducing additional accurate information may therefore lead to worse outcomes.” “Disclosure, Trust and Persuasion in Insurance Markets” (2010), IZA Discussion Paper No 5060, page 1, https://docs.iza.org/dp5060.pdf.} In other circumstances, this may mean that some consumers are harmed due to having insufficient information to understand the full offer. Even when there are concerns that information has not been provided in a wholly transparent fashion, the effect may vary significantly from case to case, and within a given case from consumer to consumer, depending on the characteristics of the product and its customers.^{15Heidhues, P. and Köszegi B. (2018) discuss in their review article how interventions may impact naïve and sophisticated consumers differently, sometimes with adverse consequences for overall welfare. For example, as consumers become more educated, this may reduce the scope for differentiation (and sometimes cross subsidisation) between consumer types. This can benefit one group, while harming the other group and (depending on the context) harm welfare overall.}  

In short, because decision making is typically context- and consumerspecific, distinguishing between harmful versus benign or beneficial practices (and, more generally, identifying effective interventions) is likely to require highly case-specific empirical evidence.^{16For example, Loewenstein et al. state: “Sometimes public policies get ahead of the data justifying their implementation. In light of the complex economic and psychological mechanisms at play in the real world, one of the major themes of this paper is the difficulty of anticipating demand and supply side reactions to disclosure requirements. This is precisely why experimentation is so important. Ideally, new proposed disclosures should be tested on a limited scale, via randomized field experiments, before they are rolled out to the general public”. See Loewenstein, G., Sunstein, C.R., and Golman, R., “Disclosure: Psychology changes everything” (2014), Annual Review of Economics, 6, pages 391–419. See also section 5 of UKCN, “Helping people get a better deal: learning lessons about consumer facing remedies” (October 2018), https://assets.publishing.service.gov.uk/media/5bae5f86e5274a3e070ae944/UKCN_consumer_remedies_project_-_lessons_learned_report.pdf.}

An economic basis for operationalising the legislative test for effects

Outside of those practices which are presumed unlawful, the legislative framework gives the CMA discretion as to when and how it intervenes. In our view, economic evidence is likely to inform the application of that discretion and, as such, firms’ compliance assessment.

Specifically, as noted above, some practices are unlawful only if they are likely to cause the “average consumer” to take a sub-optimal “transactional decision” they would not have otherwise taken. The CMA states that the “average consumer” is defined by law as someone who is reasonably well-informed, reasonably observant, and reasonably circumspect. This, the CMA indicates, is a notional construct (not a statistical average) and there is no requirement to show evidence of actual consumers being affected by an unfair commercial practice.^{17See CMA, “Unfair Commercial practices” (18 November 2025), paragraph 2.35, https://assets.publishing.service.gov.uk/media/691b9bd821ef5aaa6543ee6f/Unfair_commercial_practices_CMA207_18_Nov_2025__2_.pdf.}

While these definitions are legal ones, economic analysis is well-placed to inform their use in practice. This is because consumers frequently exhibit heterogeneous characteristics and preferences. For example, some customers will place a high value on price, whilst others will favour quality. Some will expect to be long-term users of products whilst others will not. Equally, whilst some consumers will wish to understand the finer details of all terms and conditions when assessing a price, others may favour simplicity. Against this background, conduct which may materially impact the decisions of some types of customers may have little to no effect on others. This means that presumptions regarding what factors determine decision making may be unreliable or unrepresentative. To overcome this concern, empirical economic evidence is likely to be important and, in some cases, determinative.

Specifically, economic analysis can shed light on the important issue of materiality. For example, consider a hypothetical case of fees for exiting a contract early that are alleged to be insufficiently visible. In theory, more transparent disclosure of the fees in question would be useful for consumers who value highly the option of early termination. In practice, however, if the evidence indicated that the vast majority of consumers did not seek to exit their contracts early, this could suggest that exit fees are not an important consideration for most consumers. Indeed, for that reason, it might be beneficial to give limited prominence to such fees to avoid information overload. Further, if the size of the fees were small relative to other clearly disclosed amounts, then displaying the alleged hidden fees more prominently might not alter consumer behaviour.^{18See Competition and Markets Authority v Care UK Health & Social Holdings Ltd and Care UK Community Partnerships Ltd (High Court of Justice (Ch), 23 July 2021) – [2021] EWHC 2088 (Ch). Judge: Justice Bacon (discussed further below). The High Court did not consider that the average consumer would have changed their decision even if the fees in question were deemed to have been misleading and/or aggressive (which they were not). This was in part due to a focus on long-term affordability (of which the fees in question accounted for around 2%). Non-price factors (supported by survey evidence) were also deemed important to the average consumer, such as wanting the care home: to be located close to their family and/or friends; to have a good look and feel; to be clean and tidy; to have staff with a good attitude; and to have appropriate facilities (paragraphs 162-167).} Moreover, in some cases, omitted information may be material but the average consumer would be expected to find it elsewhere (e.g., from shopping around) meaning that disclosure of the information would not be expected to lead to a different decision being made.^{19See Secretary of State for Business, Innovation & Skills v PLT Anti-Marketing Ltd (Court of Appeal (Civ), 10 Feb 2015) – [2015] EWCA Civ 76. Judges: RichardsLJ, Ryder LJ, Briggs LJ. https://www.bailii.org/ew/cases/EWCA/Civ/2015/76.html}

Drawing on these considerations, Box 2 below sets out some potential practical steps for assessing an allegation of failure to disclose important information relevant to the purchase decision (or that might be used as part of a compliance check). First, we set out a representative consumer approach. We identify key steps in the decision-making process that a typical consumer would take, drawing inter alia on the CMA’s own “access, assess, act” economic framework for analysing consumer behaviour.^{20The three steps consider how consumers: access information (for example, on the prices and quality of alternative products); assess this information (for example, by comparing rival offers and making an informed choice between them); and act on the information (for example, by being able to switch supplier easily or to move to a better product from an existing supplier). See UKCN, “Helping people get a better deal: learning lessons about consumer facing remedies” (October 2018). See also, Fletcher, A., “Disclosure as a tool for enhancing consumer engagement and competition” (2019), Behavioural Public Policy, page 5.} We then highlight questions for assessing the materiality of any information omitted. Second, we present a robustness check where we start with the omitted information and identify characteristics consumers would need to have to be materially influenced by receiving that information – firms can then assess the importance of such consumers in their overall consumer base. Finally, we comment on consumer harm. Whilst we understand that the CMA does not formally need to demonstrate harm to consumers to determine that a commercial practice is unfair, consumer harm will likely inform the CMA enforcement priorities and the extent of any redress.^{21Such prioritisation would seem consistent with the CMA’s 4Ps framework (see, for example, the CMA’s “Annual Plan 2026 to 2027” (March 2026), page 12, https://assets.publishing.service.gov.uk/media/69bd1ce5ed813d0d8b690bf8/Annual_Plan_2026_to_2027.pdf}

We note that it would be straightforward to adapt the steps to account for misleading information (e.g., one can apply the framework to information inappropriately disclosed instead of information insufficiently disclosed). 

Compliance and defence: an evidential approach

In light of the above, economic evidence is likely to be important in assisting compliance with the DMCC Act, as well as rebutting claims that practices are unfair. Empirical analyses are likely also to factor into the assessment of redress and remedial actions. The type of economic evidence will depend on the data and context, and many techniques will already be familiar to competition practitioners accustomed to assessing consumer behaviour in mergers, dominance cases, market investigations and/or digital markets.

The key issues will include the need to identify (i) critical points in the decision-making process and (ii) whether information not received (or inappropriately provided) was likely to have been material enough to influence the ultimate decision.

Decision-making processes (and how they differ according to consumer types) can be established through survey techniques (as the High Court has confirmed^{22Cityfibre Ltd, R (On the Application Of) v The Advertising Standards Authority Ltd & Anor (High Court of Justice (Admin), 15 April 2019) - [2019] EWHC 950 Admin). Judge: Murray J. Here the High Court stated: “Although the ‘average consumer’ is a hypothetical person or legal construct, it is somewhat different from legal constructs used in legal tests in other contexts, such as ‘the reasonable person’, ‘the fair-minded observer’ or ‘ordinary decent people’, in that it is necessarily linked factually to a particular population of actual persons, namely, consumers at whom the relevant advertising is targeted. This is why the law permits the decision-maker applying the test to have regard, in appropriate circumstances, to a survey of consumer views or an expert’s report” (paragraph 107).}), conjoint analyses, or by monitoring user behaviour across websites.^{23Consumer surveys (if representative and structured correctly) can elicit whether consumers were aware of and understood a particular feature of a commercial practice and, if so, how this has impacted their consumer journey and decisions. Surveys and conjoint analyses can be used to assess how consumers would react to changes in the way information is presented, although care is required when interpreting “stated preferences” (i.e., consumers anticipating what they would do when responding to surveys, as opposed to recalling actual behaviour).}  In the latter case, for example, website session data can be employed to establish consumer engagement at key information provision and decision-making points.^{24Firms may use website sessions data to understand the extent to which consumers spend time viewing terms and conditions (T&Cs) before making a purchase. It may, for example, be possible to gauge whether consumers spend longer considering more complex T&Cs versus straightforward ones, or whether T&Cs are engaged with only momentarily in all cases.} 

Where the information omitted is likely to be material only for customers with certain characteristics, firms may establish whether such types are commonly found in their customer databases.

For example, if a firm can establish (e.g., via regular customer surveys) that its customer base focuses largely on quality as opposed to price, then it might reasonably argue that a failure to disclose certain relatively small fees might not impact customer behaviour materially. Put differently, if the customer type for which disclosure might be important (here, fee-sensitive buyers) is rarely found amongst the firm’s customers, the failure to disclose is not likely to be material.

In a similar vein, firms may have good information on the extent to which their customer base splits into “informed” and “vulnerable” consumers (the latter being a priority consideration for the CMA^{25See, for example, the CMA’s “Annual Plan 2026 to 2027” (March 2026), page 11, Annual Plan 2026 to 2027 https://assets.publishing.service.gov.uk/media/69bd1ce5ed813d0d8b690bf8/Annual_Plan_2026_to_2027.pdf.} ). In such cases, for compliance purposes, firms may wish to verify that information is presented to vulnerable consumers in a way that engages them to take better decisions (or, at least, is unlikely to diminish their scope to take an informed decision). Another possibility (one that arose in the Care UK case before the High Court) is that vulnerable consumers are protected by informed consumers (e.g., family members) taking decisions on their behalf, although this will be a matter for evidence and is highly context specific.^{26See Care UK [2021] EWHC 2088 (Ch). The High Court defined the average consumer as a family member or other representative of the prospective carehome resident. This was based on a survey of over 100 representatives of care home residents, as well as 16 care home residents and five social workers. It was also consistent with the evidence of Care UK’s witnesses who indicated that it was extremely rare to deal with the residents themselves (paragraphs 72-72).}

Turning to tests of the impact of different ways of presenting information (and any subsequent consumer harm that may arise), it can be particularly informative to adopt randomised controlled trials (“RCTs”) and direct testing of choice architecture design on customer behaviour.^{27The idea behind RCTs is to compare identical groups in terms of consumer characteristics to assess how the behaviour of the treated group changes compared to the control group. For example, suppose the control group receives information in the way that is said to be fair, while the treatment group receives information in a way alleged to be unfair. If the outcomes for the treatment and control groups do not differ, it can be inferred that the alleged unfair practice did not impact consumer behaviour relative to the counterfactual of fairness.} For example, so-called A/B testing is a form of RCT used by firms in digital markets in the normal course of business. If designed carefully, these tests can be adapted for use in a compliance setting to audit choice architecture (e.g., to test for impacts on vulnerable customers). RCTs can also be used in an enforcement setting to assess the impact of alleged unfair practices or the effectiveness of proposed remedies or other policies (such as nudges to stimulate consumer engagement).^{28See, for example, FCA Occasional Paper No. 12, “Encouraging consumers to act at renewal Evidence from field trials in the home and motor insurance markets” (December 2015), https://www.fca.org.uk/publication/occasional-papers/occasional-paper-12.pdf}

Finally, if one consumer group is subject to an impugned practice but another not, then econometric techniques can (data permitting) isolate the impact of the impugned practice. Such analyses can therefore inform the extent to which harm arises and, if so, the amount of redress that may be due.^{29The FCA, for example, has recently estimated that failure to disclose certain commission arrangements between lenders and dealers has caused consumer harm in the motor finance sector. While the amount of harm has been strongly disputed by lenders, it is notable that econometric techniques have been used to estimate the extent to which APRs may have been inflated and that these estimates have informed the FCA’s approach to redress. See FCA, “FCA confirms motor finance redress scheme” (30 March 2026), https://www.fca.org.uk/news/statements/fca-confirms-motor-finance-redress-scheme.}

Conclusions

The DMCC Act permits the CMA to enforce consumer law directly and is expected to give rise to greater scrutiny of business practices. In turn, economic and empirical analysis of consumer behaviour will play an increasing and important role.

Drawing on the tools of behavioural economics, survey design and data analytics, economic analysis can help to answer questions about how consumers behave, what information they process and what (if any) decisions they would have made in different circumstances. As such, for firms under investigation, it provides a credible, evidence based means to review and respond to the regulator’s assessment. Further, for firms seeking to demonstrate compliance proactively, economic analysis can provide a structured framework for auditing commercial practices.