Introduction
Finding ways to get consumers engaged in markets is a major current topic of debate, and nowhere more so than in the UK, where the government has recently published a Consumer Green Paper that focuses heavily on this issue (Department for Business, Energy and Industrial Strategy, 2018). But why is this such an important public policy issue and what can be done about it?
This article first examines why consumer engagement is so important for driving effective competition in markets; that is, competition that delivers real benefits for consumers. It then considers four key categories of intervention that can help drive consumer engagement, which in turn involve a variety of forms of disclosure. In doing so, it highlights the importance, when policy-makers are designing such interventions, of understanding how real consumers truly behave. It describes a number of recent examples from the UK, which has a rich experience of implementing such engagement measures. Finally, the article draws on both the successes and the failures of past interventions to identify some lessons for policy-makers when stepping into this area.
The prior economic literature in this area is limited. There is a wide and fast-growing theoretical literature examining the implications of consumer behavioural biases for competitive market outcomes (summarized in Heidhues & Böszegi, Reference Heidhues and Koszegi2018). Some of these papers consider the likely market effects of specific interventions. There is also a more nascent empirical economic literature examining real consumer decision-making behaviour and the impacts of particular engagement interventions. Some of these papers are cited below.
While this paper is related to that literature, its aim is rather different. It provides an overarching view of the role of engagement interventions in driving competition, with a particular focus on disclosure, and it derives policy implications.
Why is consumer engagement important for competition?
When government or regulators intervene to enhance consumer engagement, the focus is often on consumer protection. The underlying intuition here is that if consumers have the right tools to engage with the offers available on the market, then they will be better able to protect themselves and less at risk of exploitation by suppliers. Hence, when consumer law requires that suppliers provide information on their products, it is because this helps to ensure that consumers know what they are getting and are not defrauded. Likewise, when consumer law requires that suppliers do not conceal key aspects of a product offering within the small print of a contract, this is important for ensuring that consumers do not inadvertently take on more risk than they intended.
However, it is increasingly well understood that consumer engagement also has a crucial role to play in driving competition. Engagement interventions are thus important tools within competition policy as well. In the simplest economic models, with a single market price, competition will tend to be stronger the more that consumers engage in searching around for and switching to the best offering for them in the market. At an extreme, where consumer search is too limited, monopoly prices can arise even in the presence of multiple sellers (Diamond, Reference Diamond1971). This result has become known as the Diamond Paradox.
This finding shows – at the most basic level at least – that increasing consumer engagement in markets can help to drive more vigorous competition.Footnote 1 This will in turn be good for consumers in that it drives suppliers to lower their prices or otherwise improve their product offerings. While both search and switching are relevant, it is search costs that can have the more dramatic impact on restricting competition and increasing market power (Wilson, Reference Wilson2012).
But simply driving up engagement in the form of consumer search and switching is not always the only thing that matters for competitive outcomes. Two further concepts are relevant to this question: relative salience and relative consumer engagement.
Relative salience refers to a situation in which one dimension of a product is substantially more salient to consumers than other dimensions of the same product. Where this is the case, saliency bias will tend to result in consumers engaging far more with the most salient dimension and far less with other dimensions, and this can have important implications for competition (Bordalo et al., Reference Bordalo, Gennaioli and Shleifer2016). A key finding, which is intuitive, is that where consumers engage primarily on one dimension that is most salient to them, then competition will tend to be focused on this salient aspect. By contrast, suppliers will act in a more monopolistic manner on the less salient aspects of the product offering.
Where salience differs across various aspects of quality, this can lead to heavy competition on one quality dimension and more monopolistic behaviour in respect of others. However, in a number of markets, it will be price that is the most salient aspect of a product offering, or more specifically the upfront price. This relative salience of upfront prices over other aspects of the product offering can also be accentuated, such as through the use of price comparison websites (PCWs). These are often specifically designed to make upfront prices highly salient. This is good for the most part, but there is a downside risk. While we may expect PCWs to increase competition on upfront prices, this may be accompanied by weaker offerings in terms of quality or less salient aspects of the sales contract such as follow-on charges (Helfrich & Herweg, Reference Helfrich and Herweg2017). Moreover, the stronger the competition on the more salient aspects, the worse the offering may become on the less salient aspects.
This suggests that, as well as increasing engagement, it is important to think about the form such engagement takes and where possible to help consumers to make more holistic decisions. An example might be rules that require (or facilitate) enhanced salience of quality information on comparison websites. Alternatively, it may be justified to introduce direct regulation around some of the less salient aspects. Examples observed in practice include minimum rules around product safety or the small print in contracts (which it can be hard for a consumer to observe, let alone focus on).
Relative consumer engagement refers to a situation in which consumers differ in the extent to which they engage in a market. Where firms’ product offerings are uniform across consumers, this is less important. Competition will tend to increase with the extent of engagement of the marginal consumers, and the unengaged will receive the competitive benefit of that engagement.
By contrast, where firms’ can vary their product offerings across consumers, relative consumer engagement can become important for market outcomes. For example, in an environment where suppliers can engage in price discrimination, we might expect them to set lower prices for the more engaged consumers and higher prices for those consumers who are less engaged in terms of putting little or no effort into searching or switching (Shaffer & Zhang, Reference Shaffer and Zhang2000). In such contexts, enhancing engagement amongst the already engaged can have the effect of increasing the difference between the engagement levels of these two groups. As prices fall for the engaged, the unengaged may see less benefit or no benefit at all, or they may even see their prices rise.
The overall effect on average prices paid, and thus on consumer welfare, is ambiguous in such circumstances. As is well understood, price discrimination can itself either limit competition or enhance it, depending on circumstances. However, these sorts of price differentials could give rise to significant distributional concerns, with some consumers clearly not gaining benefits from competition. This may be politically important, if nothing else, and especially so where the relatively unengaged are also more vulnerable.
This can in turn result in a policy focus on reducing or eliminating the differentials. At a minimum, attempts may be made to increase the engagement of the relatively unengaged. In the most extreme situations, regulators may determine that more paternalistic interventions are justified. These could involve changes of choice architecture, such as a change of default that ensures that the least engaged are defaulted into a reasonably good option. Or they could even involve direct regulation of market outcomes, such as in the form of price caps. This issue is discussed further in the final section of this paper.
Interventions to enhance consumer engagement: some key categories
So, what sorts of interventions might we expect to enhance consumer engagement? In order to have an effective demand side that acts to help drive effective competition, consumers are required to engage with the market in three key ways (Bennett et al., Reference Bennett, Fletcher, Hurley and Ruck2010):
(1) They need to access information about the products (goods or services) available in the market.
(2) They need to assess that information, in terms of making comparisons across the various products and determining which best suits their preferences.
(3) They then need to act on that information, by purchasing (or, in some cases, switching to) this product.
These three ‘A's of consumer decision-making provide a useful framework for thinking about possible interventions. For example:
(1) Consumers can be helped to access information through placing simple disclosure obligations on suppliers, to ensure that key product information is in the public domain.
(2) Consumers can be helped to assess that information through comparison tools, which ensure that the format of any disclosure is standardized so as to be easily comparable across products or to require that disclosure is made to third parties who are able to facilitate consumer assessment.
(3) Consumers can be helped to act, following these assessments, by switching interventions, which make the purchasing process, or the process of switching, cheaper or easier.
However, before any of these choice enablers become relevant, consumers do have to engage with the market to at least some extent. In some markets, this lack of engagement can, in itself, constitute a significant block to competition. In identifying possible solutions to this, it is useful to consider two categories of consumer disengagement – conscious disengagement and unconscious disengagement – albeit noting that, in any particular situation, consumers may exhibit elements of both (this could be termed ‘semi-conscious’ disengagement).
The first category, conscious disengagement, occurs when consumers make a conscious decision not to engage, based on how much they expect the process to cost them, whether this be in terms of their time, effort or actual pecuniary costs. Such decisions could be entirely rational; searching and switching do incur genuine costs. However, they can potentially also be influenced by a variety of behavioural factors.
For example, if consumers expect the engagement process will be costly and time-consuming and that they will only gain over the longer term, then a consumer who is overly focused on the present (e.g., as a result of hyperbolic discounting) may be especially disinclined to engage – or alternatively, they may be especially inclined to leave the engagement to a later occasion. Likewise, a consumer with a strong status quo or endowment bias may be especially disinclined to consider switching to a new product or supplier, while a consumer with a strong default bias may be especially inclined to stick with the default product in the market, rather than engaging to find a better one. A consumer who struggles with numeracy or logic may be especially disinclined to engage with a process where those skills seem likely to be needed. Consumers who have never searched or switched before may also overestimate how difficult the process will in fact be and may be overly influenced by anecdotes from others who have had poor engagement experiences.
In this context, any intervention that helps consumers to access, assess and act on relevant market information, or otherwise reduces their expectations of the costs involved, can help to tip the balance towards getting a consumer to engage. As such, the types of intervention identified in the numbered list above may also act to enhance consumer engagement.
The second category, unconscious disengagement, occurs when consumers do not consciously decide not to engage; they simply do not do so. This category of disengagement is harder to address through simply facilitating consumer decision-making.
Again, behavioural factors may be relevant here, such as self-control. A working mother, who knows she is paying too much for her electricity, may have every intention of going home after work and looking online for an alternative supplier. But after cooking dinner, getting the kids to bed and tidying up the house, she could be forgiven for collapsing onto the sofa and forgetting about her electricity search plans. Saliency can also be relevant. If the need to engage is not both obvious and pressing, it may not command a consumer's limited attention, however beneficial it might be for the consumer.
In this context, interventions to increase consumer consciousness around engagement can be valuable. These can take the form of triggers, which enhance the saliency of the decision. They can also take the form of changes in choice architecture that either force a choice, as opposed to allowing a default option, or change the framing around the choice.
In the following sections, therefore, I will look at four overarching categories of engagement intervention:
(1) Pure disclosure: which is designed to enhance consumer access to relevant product information.
(2) Comparison tools: which are designed to enhance consumer assessment across different products.
(3) Switching interventions: which are designed to enhance the ability of consumers to act on their assessment.
(4) Pure engagement tools: which are designed to get consumers engaged in the market in the first place, one way or another, when other methods fail. This can be especially relevant for addressing unconscious consumer disengagement.
In doing so, I will provide a number of recent examples, primarily from the UK, where consumer engagement has been a particular policy focus for some time. In legal terms, these interventions can arise in various different ways, including through the application of consumer law, remedies within competition law, the use of regulatory tools, industry initiatives (typically after prompting by regulators) or sometimes government legislation.
It is worth emphasizing that, while the first category is termed ‘pure disclosure’, the other categories also involve disclosure of various forms. I also emphasize the importance, in developing these interventions, of really understanding consumer behaviour and the role that empirical evidence can play in informing such policy design. This evidence can be relatively formal, such as that derived from laboratory experiments or randomized controlled trials (RCTs; also known as field experiments). In many cases, though, the behavioural insights informing the interventions are primarily based on survey information, the ex post evaluation of past interventions or a broad understanding of the nature and direction of likely behavioural biases.
It should also be noted that, although these interventions are designed to change consumer behaviour, it is not usually feasible to place regulatory burdens on consumers themselves. As such, these interventions are typically imposed on suppliers and act to change the way in which suppliers interact with consumers, such as by requiring them to change what they disclose to consumers and how, with a view to this then facilitating or triggering consumer behaviour change.
Pure disclosure to enhance consumer engagement
The first category of engagement intervention listed above is ‘pure disclosure’. At one level, it is obvious that consumers can only make reasonable purchasing decisions across products if they have relevant information on these products. As such, simple disclosure interventions, which are designed to overcome a core asymmetric information problem between suppliers and their customers, have long been an element of consumer-focused regulation. A recent UK example has been a disclosure requirement on online platforms that resell secondary tickets for events to state clearly where the seat is and whether there are any restrictions on ticket resale (CMA, 2018a). This is relevant information for consumers, but it was not previously made available by these sites on a systematic basis.
Over recent years, however, there has been a growing recognition – partly reflecting developments in behavioural science – that simply providing information to consumers may not be enough to improve their decision-making. Indeed, many experiments have shown that they may potentially even make worse decisions if they feel overwhelmed by too much information; this is sometimes known as ‘information overload’. For example, the morass of boring-looking paperwork that suppliers are required to provide to consumers in some financial services markets may not help consumers in their decision-making, and may even discourage them from trying. Indeed, it has even been suggested that firms may therefore have an anti-competitive incentive to disclose too much (Persson, Reference Persson2018).
This has led to various alternative approaches to disclosure, which are more sensitive to behavioural factors. Three key categories are described below.
Disclosure requirements that go beyond simple information revelation to enhance consumer awareness and understanding of the information
It is well understood that the impact of disclosure on consumer decision-making can be greatly enhanced by presenting the relevant information in a more user-friendly and attractive way.
Improving the effectiveness of disclosure can potentially be achieved through principles-based interventions such as the UK Consumer Rights Act 2015. This requires that consumer contracts are transparent (plain, intelligible and legible where written) and prominent (brought to the consumer's attention in such a way that the average consumer would be aware of it). The reference to the ‘average consumer’ here is intended to reflect the need for suppliers to take account of real consumer behaviour.
However, we also observe regulatory requirements that are very specific about the way in which information should be presented based on behavioural evidence as to what works best in enhancing consumer attention and understanding. So, for example, when the UK Competition and Markets Authority (CMA) carried out an investigation into private motor insurance in 2014, it determined that consumers faced an acute lack of information when purchasing ‘no claims bonus’ protection as an add-on insurance product. As a remedy, the CMA required that consumers must be given, at the point of purchase, a tightly specified set of information to aid their decision-making (CMA, 2014a). This was required on the basis of qualitative consumer testing of different possible ways of providing relevant information (GfK, 2014).
Likewise, following its 2015 market study into cash savings accounts, the UK Financial Conduct Authority (FCA) required not only that interest rates on such accounts should be displayed prominently, but more specifically that for online banking this information should be no more than ‘one click away’ from the customer's account home page (FCA, 2015a). This was based on evidence that consumers typically will not bother clicking through a sequence of links to obtain the relevant information.
In some cases, it can be better to provide less information, not more. For example, the FCA has recently been reviewing its disclosure requirements for financial services firms and has concluded that several are ineffective, and some could actually harm consumer decision-making (FCA, 2015b).
Disclosure requirements that are specifically designed to make certain aspects of the offer more salient
As discussed above, consumers can easily become highly focused on one particular dimension of a purchasing choice, such as the upfront price of the core product, and give less attention to other important elements, such as the quality or prices of add-on products. Such relative salience can in turn lead to excessive competition on the salient dimension and too little competition on the others.
As an example, in 2018, the CMA intervened in the online car rental sector to enhance upfront transparency about key – but less salient – aspects of the product, such as prices of optional extras, key rental terms and requirements and any additional charges, such as fuel surcharges, young driver fees and out-of-hours pick-up charges (CMA, 2018b). This intervention, which requires online sites to show the total price payable for the full car rental package chosen, is expected to facilitate better consumer decision-making and thus enhance the effectiveness of competition in this sector.
Likewise, following its 2006 investigation into store cards, the UK Competition Commission (CC) mandated substantial additional provision of relevant information in consumers’ monthly statements. In doing so, it was explicit that the font size used “must not be less than the largest font size used for transaction and balance details,” and that the current annual percentage rate (APR) applicable to purchases must be shown in bold (CC, 2006a). This reflected the concern that consumers were overly focused on the purchasing power that store cards afforded them, while the costs involved were relatively less salient. The CC's ex post evaluation of this intervention 5 years later found that store card usage, average store card APRs and outstanding store card balances had all fallen since the start of the CC investigation (CC, 2011). Interestingly, it also noted that some of the reduction occurred during the life of the investigation, possibly due to the extensive public debate, including across the media, which accompanied this ongoing investigation. This shows the positive impact that greater consumer awareness can have on market engagement.
Interventions to prevent misleading or obfuscating disclosures
Suppliers are well aware of the difficulties that consumers can face in choosing between product offerings. Many responsible suppliers do their best to help consumers with this choice. However, it can also be in the selfish interests of less scrupulous suppliers to worsen consumer decision-making. This is clearly an attractive option for suppliers who are attempting to market weak offerings. However, the link with competition means that even suppliers with good products may have an interest in worsening consumer decision-making if they can thereby weaken market competition and so raise industry profits (Spiegler, Reference Spiegler2006).
Preventing suppliers from making misleading or obfuscating disclosures to consumers can potentially be achieved through general rules that focus on the key principles at stake. For example, the 2008 Consumer Protection from Unfair Trading Regulations prohibit any sales practice that involves misleading actions or omissions, where these are defined in terms of whether they materially impair an average consumer's ability to make an informed transactional decision and in doing so changes that decision (Office of Fair Trading & Department for Business, Enterprise and Regulatory Reform, 2008).
Whilst the ‘average consumer’ is taken here to be reasonably well-informed, reasonably observant and circumspect, the intention is that the law should allow for real consumer behaviour. Indeed, it is explicitly stated that if a clearly identifiable group of consumers is particularly vulnerable to a trading practice (e.g., because of age, infirmity or credulity) in a way that a trader could reasonably be expected to foresee and if the practice is likely materially to distort decisions made only by that group, then the benchmark will be the average member of that group.
In some cases, more specific rules can also be valuable, which are designed to prevent identified instances of misleading disclosure. As an example, in 2012, the UK Office of Fair Trading (OFT) investigated the practice of ‘drip pricing’ in sales of airline tickets. This took the form of airlines setting low upfront ticket prices but then imposing high charges on customers at a late stage of the sales process for using a standard payment card. The outcome of this investigation was an outright ban on payment charges for debit cards (OFT, 2012). This intervention reflected evidence from laboratory experiments carried out on behalf of the OFT that found drip pricing to have significant negative consequences in terms of reducing consumer search across suppliers and raising overall prices paid (OFT, 2010). More recently, this ban has been extended across both credit and debit card payments and across all sectors through the second EU Payment Services Directive (2015).
Comparison tools to enhance consumer engagement
Even if consumers have plenty of relevant information, they may find it hard to make effective purchasing choices if this information is not easily comparable across products, given the computational complexity that can be required. Indeed, this may well deter them from even trying, resulting in low levels of search and weak competition.
As such, there has been substantial regulatory focus on interventions that are designed to help consumers make cross-product comparisons. Again, it is possible to identify a number of different policy approaches to this second category of interventions: tools to aid comparison and shopping around.
Disclosure requirements to enhance comparability of product offerings
An obvious first step in facilitating cross-product comparisons is to require information to be provided in a systematic way across products, such that it is more easily comparable.
Such requirements have been imposed in a number of UK markets. For example, following an OFT investigation into credit card comparisons, the industry agreed to adopt new summary boxes to present fees and charges in a systematic way. The tabular format used was consumer-tested and the intention was to both improve consumer understanding and to enhance search and switching.
Consumer testing was also used in the development of the new Key Information Document (KID) that must be provided to investors by suppliers offering packaged retail and insurance-based investment products (PRIIPs) under the recent EU PRIIPs regulation (European Commission, 2015). The information to be provided is tightly specified with the aim of helping consumers in both understanding and comparing PRIIPs. Likewise, the FCA employed consumer testing in designing a new summary box to be provided for UK cash savings accounts (FCA, 2015c). This newly required standardized disclosure is again intended to aid comparability and thus comparison across such accounts.
In these examples, the information provided was already available. The regulatory intervention simply required it to be set out in a systematic way. However, interventions can sometimes require the creation of new forms of information.
For example, an OFT study in 2002 found that it was difficult for consumers to compare printers on the basis of their likely total cost, including ink cartridge refills, as there was no information available to consumers that would allow them to estimate their likely usage of additional cartridges. The OFT therefore recommended that the industry develop a standardized method for calculating price per page for ink cartridges so that consumers could more effectively calculate and compare the full life cost of printers, including spending on ink (OFT, 2002).
This example raises another interesting point. Requiring suppliers to disclose comparable information can be useful even if consumers themselves do not directly use it, so long as there are third parties that utilize the information to provide enhanced recommendations to consumers. In the case of printers, the standardized ink cartridge costs play a key role in the recommendations made by retailers and consumer magazines and are thus likely to have improved consumer choices in this market (Which? Magazine, 2016). This is true even though consumers themselves still find personalized calculation hard.
Finally, in designing disclosures to facilitate comparison across products, regulators face a difficult trade-off in terms of how prescriptive to be about what information is provided and how it is provided, including how much to mandate that it be simplified. More precision is typically helpful for comparability across suppliers and across products, but it can potentially harm innovation if the disclosure does not make allowance for novel aspects of those products. Where disclosures are simplified, consumers are more likely to be able to make effective comparisons, but there is a risk that these may in fact deliver poor outcomes if the simplified information does not properly reflect what consumers most care about.
An example of the latter point is the legal requirement under the EU Consumer Credit Directive that suppliers of consumer credit include within their promotional material a representative APR of charge. This is a single figure that represents the total cost of the credit per year and is calculated on a clearly specific set of rules, harmonized at the EU level (European Commission, 2008). APRs are valuable for providing illustrative guidance on expected cost that is comparable across credit offerings. It could be argued that APRs would be useful, even if not strictly accurate, if they allow for effective like-with-like comparison. In many situations, this may well be true. However, experimental evidence suggests that APRs can encourage consumers to take out longer-term loans than they would choose if they were instead provided with information on total loan costs (Lunn et al., Reference Lunn, Bohacek and Rybicki2016). Likewise, experimental evidence finds that consumers of payday loans make better choices across offerings when shown the total amount payable than when shown APRs only (Bertrand & Morse, Reference Bertrand and Morse2011; FCA, 2015d).
Disclosure requirements to facilitate the development of third-party aggregators, such as PCWs
An alternative route to enhancing consumer decision-making is to develop third-party aggregators, such as PCWs or digital advice tools, or to facilitate their development.
Requiring suppliers to provide information in a comparable form can itself facilitate the development of such third-party aggregators. However, this may not be sufficient on its own. In some sectors, more may be required to ensure that such aggregators emerge.
One option could be for regulators themselves to create such comparison tools. However, they are in fact poorly placed to develop tools that, in order to be effective, really need to be attractive, easy to use and well publicized and to build on the latest technological developments. This is simply not what regulators are good at, nor do they have access to the requisite funding.
As an alternative, regulators can mandate that industry develops such solutions. For example, following its 2014 investigation into private health care, the UK competition authority, the CMA, required the industry to set up an independent but industry-funded information organization that would make the information on fees and quality available via a website in a format that permits patients to search and compare results easily (CMA, 2014b).
Even mandating the creation of such a tool can be difficult, however, if this goes against the incentives of industry. It can be hard to specify precisely how such a tool should be designed, and yet – absent such precise specification – industry may have poor incentives to make a tool that works well. Precise specification is also difficult in the context of the dynamic nature of digital markets, and digital comparison tools in particular. In 2006, when the CC first required the development by industry of an independent PCW for the home collected credit market (CC, 2006b), it was focused on developing this tool for use within browsers on desktops, and perhaps laptops. The authority had no idea that consumers might one day spend substantially more time browsing the Internet on smartphones than on computers, let alone that they might make product comparisons through specialized apps.
A preferable option, where feasible, is for the regulator to facilitate the development of commercial comparison tools by requiring suppliers to disclose to such aggregators accurate and up-to-date information about their products. Ideally, this should be disclosed in a format that is stable over time in order to avoid the third-party aggregator having to continually redesign their algorithms so as to collect and present the relevant information.
For example, in the payday loan market, the CMA was concerned that it was difficult to search online across suppliers. In 2015, it therefore required suppliers to provide their data to at least one PCW that is both independent and FCA-regulated (CMA, 2015). A quick review of the current online situation shows this intervention to have been highly effective in facilitating the commercial development of PCW functionality covering payday loans.
In some markets, individual consumers effectively require bespoke personal comparisons based on their usage data. In such circumstances, disclosure requirements may need to go even further and require incumbent suppliers to make such data available.
As an example, the UK Open Banking initiative (itself a remedy arising from the CMA's 2016 Retail Banking investigation) requires banks to make personal transaction data available to authorized parties, with consumer consent. These data must be provided through standardized and open application programming interfaces in order to facilitate the development of more personalized price comparison services, and also new account information aggregation services (CMA, 2016a). These latter have a variety of potential advantages for competition. They should allow consumers to ‘multi-bank’ – that is, take banking services from separate suppliers – without forsaking the convenience of having a single banking interface, and they should also facilitate greater searching and switching across suppliers for individual banking services.
Interventions to ensure that third-party information collators are themselves effective in enhancing consumer decision-making
Whilst third-party aggregators have the potential to play a valuable role in enhancing consumer decision-making, it is important to recognize that they may not always carry out this function as well as they might.
This could be inadvertent. For example, a PCW might not be aware that the effect of enhancing the salience of prices may be for competition on quality to be reduced. Moreover, even if it wanted to make quality more salient, it might struggle to source comparable information on quality. In this situation, a regulatory requirement on suppliers to disclose underlying quality information may be a useful intervention.
The FCA is currently piloting such a disclosure requirement with the insurance industry, which requires suppliers to publish a series of comparable value measures. The data are not provided in an especially consumer-friendly format, but they can be utilized by third-party aggregators to provide consumers with enhanced, easy-to-use indicators of product quality. In 2018, the FCA published its early findings that this pilot disclosure had already incentivized some suppliers to enhance the quality and value for money of their products (FCA, 2018a).
However, constraining the effectiveness of consumer decision-making could potentially also be a deliberate strategy for third-party aggregators. They may well have a commercial incentive to do so when comparison tools are funded through commissions from suppliers, rather than directly by consumers. If a comparison tool can be designed to limit its effectiveness in driving effective competition between suppliers while remaining sufficiently good to attract consumers, then this may be a profitable strategy, so long as it is able to extract some of the additional rents through commissions.
While it may seem superficially attractive to require aggregators to change their funding model in order to avoid any risk of conflicts of interest with consumers, it can often be hard to identify a funding model that does this while enabling the aggregator to remain viable. As such, an alternative regulatory approach may be to place restrictions around how third-party aggregators themselves disclose information to consumers.
For example, in 2016 in the payday lending sector, the FCA imposed rules on PCWs that require that users should be able to search by amount and duration of loan, that search results must be displayed in ascending order of total amount payable and that the ranking must not be influenced by any commercial relationship (FCA, 2016a). These rules were informed by consumer testing in the form of behavioural laboratory experiments (FCA, 2015d).
In terms of commercial relationships, a less interventionist option would simply be to require PCWs to disclose their links and commissions and to set out clearly how their rankings are computed. However, some commentators are sceptical about the efficacy of this approach, given the responsibility it places on consumers to utilize this complex information effectively and to avoid being misled by the consequently biased rankings. This scepticism is partly based on evidence from the US Federal Trade Commission, which found that consumers were confused by additional information on commissions in an experimental setting relating to mortgage brokering, and they typically made worse choices as a result of receiving this information. Specifically, they seemed to infer (wrongly) that an apparently high commission to a broker must signal a poor-value product (Lacko & Pappalardo, Reference Lacko and Pappalardo2004).
As with many other issues discussed in this paper, finding ways to ensure that third-party aggregators act broadly in consumers’ best interests remains an important and complex area of policy – and indeed legal – debate.
Switching interventions to enhance consumer engagement
As discussed above, competition may also be weakened if consumers are disinclined to switch supplier. Consumers will be less likely to bother engaging in searching the market if they do not expect to switch, even if they find a better supplier. And suppliers’ incentives to offer good value for money will be reduced if they expect to keep their existing customers anyway and do not expect to win new ones. A third category of engagement intervention thus seeks to address such switching barriers.
Consumers may be deterred from switching on the basis of real switching costs. However, behavioural factors can also play an important role, and this has clear implications for optimal remedy design.
Remedies to address real switching costs
Real switching costs may be pecuniary, such as fees incurred for exiting a contract, the need to pay duplicate fees from both suppliers during a transition period or a loss of warranty cover. Consumers can also face real but non-pecuniary switching costs. These could potentially relate to the time and effort associated with switching, such as having to phone up and spend time on ‘hold’ in order to switch supplier. They could also relate to consequential actions, such as the time spent contacting friends to give them one's new number if switching phone provider.
If there are real switching costs of this sort, an obvious regulatory intervention is to reduce them. The evidence suggests that reducing or eliminating such cost barriers can indeed be very effective in enhancing consumer switching.
Where such costs arise from contractual restrictions, simply requiring the removal of the restriction can be a powerful remedy. For example, in 2004, the OFT required car manufacturers to remove restrictions that existed within new car warranties that stated that use of an independent servicing provider would invalidate the warranty (OFT, 2004). This provided a clear disincentive to consumers switching away for their servicing. A 2006 ex post evaluation of this intervention found that freeing up consumers to use independent servicing providers saved consumers around £66–81 million per year (OFT, 2006).
Notably, the evaluation also found that consumer awareness of the change was limited and that the overall benefit of the intervention would have been significantly greater with enhanced consumer promotion. This finding is arguably relevant to many demand-side interventions; that is, they only work (or work best) if consumers are made aware of them.
In other situations, the intervention may need to go further than simply removing a contractual restriction. An example is the introduction of mobile (cell phone) number portability in the EU, which reduced the hassle cost of switching mobile operator by mandating that users can retain their number.Footnote 2 This intervention has been found to have decreased mobile prices by 7.9% on average (Cho et al., Reference Cho, Ferreira and Telang2016).
Remedies to address behavioural factors limiting switching
Even where the costs of switching are real, consumers may potentially overestimate them, and especially the downside risk of switching. For example, in response to a lack of competition in the UK retail banking market, which partly reflected limited switching, the major banks agreed to develop a reliable 7-day Current Account Switching Service (CASS). Despite this providing a good service, the 2016 CMA review of the market found poor consumer awareness of CASS and also low levels of confidence in the service, with consumers continuing to fear that switching bank would be time-consuming, burdensome and unreliable (CMA, 2016a). As well as requiring some small changes to the CASS scheme, the CMA required the banks to carry out a long-term promotional campaign to improve consumer awareness and understanding of it.
Behavioural factors can also create additional switching barriers of their own. For example, in 2011, the UK telecoms regulator Ofcom banned auto-renewing fixed-term contracts in telecoms (Ofcom, 2011). The primary problem with such contracts was not the automatic renewal itself, but the high termination fees that were then imposed within the new contract. In principle, there is no reason why consumers who wished to switch could not simply cancel during the window prior to auto-renewal. However, the empirical evidence developed for Ofcom found that customers on these contracts are far less likely to switch than other customers. This suggests that behavioural factors play a role, with the structure of the contract perhaps exploiting the difficulty consumers have in remembering to cancel within a particular time window (or their failure to understand the consequences of not doing so).
Also in telecoms, Ofcom has been concerned that consumers are deterred from switching mobile phone provider because, to do so, they have to contact their existing provider to terminate their agreement and to receive a ‘PAC code’ if they wish to retain their existing number. In the past, this has required a phone call. There is clearly a real cost of time associated with such calls, but Ofcom found that behavioural factors were if anything more relevant. Consumers dislike making such calls, not least because they do not like to face unwanted sales attempts from apparently disappointed salespersons, who are incentivized to try and retain customers. To ameliorate these concerns, in 2017, Ofcom mandated that providers introduce a system by which customers can cancel their agreement (and receive their PAC code) via text and Internet-based options. This should enable easy switching within one working day, without the need for any difficult conversations (Ofcom, 2017a).
Pure engagement tools
Facilitating better consumer decision-making through interventions of the sort described above may well be enough to trigger engagement amongst many consumers. Other customers, though, may need a further prod. This is particularly likely if they are not in fact making any sort of conscious decision not to engage, but rather their lack of engagement is unconscious.
Encouraging engagement amongst such consumers can be difficult. However, a number of interventions have been tried, with some success.
Disclosures that are designed to trigger consumer engagement
If consumers are provided with specific relevant information in a prominent, reader-friendly and timely way, then this can help to drive engagement. A recent example is the FCA's 2016 requirement on providers of general insurance to include last year's premium when sending out renewal letters (or an annualized version of the existing premium if there have been any mid-term adjustments) (FCA, 2016b).
This intervention was based on empirical evidence from RCTs that demonstrated that the inclusion of last year's premium led to increased switching, at least for those customers facing the largest price hikes (FCA, 2015e). Of course, consumers would already have had information about the previous year's premiums in their files somewhere, but reminding them of this information in the renewal letter, right next to the new premium, had a significant impact.
Disclosure remedies that are intended to aid decision-making when consumers are already using a product or service
Disclosure remedies are not always targeted at consumers’ initial purchasing decisions, but they can also be directed at improving their decision-making once they are using a product or service, without involving any change in supplier. This sort of disclosure remedy need not necessarily affect the intensity of competition in the market, but can nevertheless be important in ensuring that competition occurs on dimensions that deliver real benefits to consumers, rather than firms competing for customers in order to later extract as much out of them as possible.
As an example, following pressure from both the OFT and UK government, the major UK banks agreed to offer consumers the option of sending text alerts if they were about to go into unarranged overdraft, which can in turn incur substantial fees, and allowing them a ‘grace period’ in which to sort out the situation (HM Treasury and Department of Business, Innovation and Skills, 2011). Taking advantage of a staggered rollout, the FCA was able to carry out an empirical ex post evaluation of this intervention. The text alerts were found to be powerful in reducing unarranged overdraft charges, with the impact found to be especially strong for consumers who also had mobile (cell phone) banking apps, for whom charges fell by 24% (FCA, 2015f).
However, while clearly beneficial, only around 3–8% of bank customers in fact signed up for such text alerts. On the basis of this, in 2016, the CMA required the banks to automatically enrol consumers into receiving text alerts by default (CMA, 2016a). Follow-up evidence collected by the FCA shows that the alerts have retained their positive impact across this wider customer base, albeit the alerts are found to be more effective for ‘rare’ overdraft users than for ‘heavy’ users (FCA, 2018b).
Changing the choice architecture facing consumers
Default bias is a key driver of consumer disengagement, and one that can potentially be exploited by suppliers to make sales. Where this is the case, consumer decision-making can be enhanced by interventions that change the choice architecture facing consumers, and in particular the default options.
Two recent examples from the EU are worth noting.
• Ban on ‘opt-out’ selling online. If a box involving an add-on sale is pre-ticked, consumers may be disinclined to untick it (‘opt out’) due to default bias, even if this means buying entirely unwanted products. This led the EU to instigate a ban on online sellers engaging in such opt-out selling for additional products or product attributes. The intention is that consumers should be required to make an active choice in respect of these, rather than falling into buying them from the supplier by default.Footnote 3 This remedy has the potential to enhance competition for these additional products, as well as generally improving consumer decision-making.
• Microsoft Internet Explorer competition remedy. This intervention was a rare example of a remedy recognizing behavioural biases under core EU antitrust law. The concern was that consumers automatically adopted Microsoft Internet Explorer, which came free with Microsoft Windows, even though there were alternative free browsers available. This default bias allowed Microsoft to essentially leverage its dominant position in operating systems into the browser market. The remedy required consumers to make an active choice of browser through a choice screen (known as the ballot box) that was carefully designed to be unbiased (European Commission, 2009). It seems to have been fairly effective, with Internet Explorer's market share dropping off far faster in the EU, where Microsoft was subject to the Commission's remedy, than in the USA over the period from March 2010 (when the remedy was implemented) to November 2014 (when the ballot box was dropped).Footnote 4
Another potential response to a lack of consumer engagement is collective switching on either an opt-in or opt-out basis (Deller et al., Reference Deller, Bernal, Hviid and Waddams Price2017). The UK energy regulator Ofgem has recently been trialling an opt-in variant in the UK energy market (Ofgem, 2018a).
Delaying the consumer decision to overcome point-of-sale effects
Finally, an important bias that is observed in real-life sales situations is that consumers are often reluctant to say no when dealing with salespeople face to face. This may be because they trust the advice of the salesperson or it may be that they do not want to upset them or deal with a difficult interpersonal situation, but the overall effect is that suppliers can sometimes face an important point-of-sale advantage when selling follow-on products, such as warranties or insurances relating to a primary product that a consumer has purchased. This can in turn lead to consumers purchasing additional products that they do not want or need, or alternatively not shopping around for the best deals on such products.
In such situations, interventions can potentially be used to separate the two purchasing decisions. A recent example is the FCA's 2015 intervention in relation to guaranteed asset protection (GAP) insurance (FCA, 2015g). This is typically sold alongside a car and covers the difference between the sale price of the car and its replacement value at the time of any insurance claim. When the FCA studied this market, it found that the gross profit margin on this form of insurance was around 4.5 times higher than for standard car insurance, and that standalone GAP insurance could be purchased for around half the price when sold instore as an add-on. Nevertheless, more than half of the purchasers of this product had not considered buying it prior to purchase, and less than a fifth had shopped around.
The FCA did not consider that instore warnings would be sufficient to overcome its concerns in this case. Instead, the FCA imposed a four-day deferral period before any add-on purchase could occur. The intention is that this should reduce the point-of-sale advantage, encourage consumers to make a more positive choice about the purchase and give them a chance to shop around before doing so.
Some lessons for policy-makers
As should be clear from the above discussion, regulators and competition authorities are increasingly turning to demand-side interventions to improve consumer engagement and decision-making and thus enhance competition. Many of these involve disclosure of different forms. In designing and implementing these interventions, a number of specific lessons can be drawn from the regulatory experience to date, many of which have been discussed above. There are, however, a number of more generic lessons that are worth highlighting.
First, in designing interventions of this sort, it is important to analyse carefully the barriers to decision-making that real consumers face so as to target these effectively. Where possible, consumer testing of remedies can also be hugely valuable. Effective interventions are not easy to design, as is demonstrated by experimental results showing that perfectly sensible-seeming interventions often have only limited impact. It can be hard to assess how consumers will react, even directionally. As a simple example, when the FCA carried out a RCT in 2013 to assess how best to design a letter to consumers offering redress, it found that women (but not men) were on average less likely to respond if the letter was signed by the CEO than if it was simply signed by the more generic ‘Customer Services Team’ (FCA, 2013). This finding was entirely unexpected and remains difficult to explain.
In carrying out such consumer testing, RCTs (also known as field experiments) are the gold standard, because real interventions can be tested on real consumers in real situations, without those consumers even being aware that they are part of an experiment. These typically require, however, the close involvement of the firms that hold the customer relationships. These firms may be uncomfortable about experimenting on their own customer base and also unenthusiastic about testing interventions that are designed to facilitate their consumers switching away from them. As such, firms may need to be mandated to engage with such processes if they are to yield valuable results.
Second, whether or not such ex ante consumer testing is done, it is important that regulators revisit markets to carry out ex post evaluation of the effectiveness of interventions. This is partly because it is hard to assess likely supplier reactions through consumer testing. It is partly also because suppliers are more likely to act in a way that is supportive of the desired outcomes if they are aware that there will be future evaluation and an associated risk of further intervention if there has been insufficient improvement.
Third, the tone of voice used by regulators in this area is important. They can easily slip into a tone that suggests demand-side problems in markets are somehow consumers’ ‘fault’ or that consumers are being stupid or irrational when they fail to spend their spare time seeking out the best deal in a particular market. It is, of course, entirely rational for consumers to be quite picky about how they use their precious and limited time and effort. Moreover, to the extent that there are behavioural factors at play in weakening consumer decision-making, these are often the types of behavioural biases and psychological traits that we all exhibit as human beings. Just because someone does not choose to spend their evening online seeking the optimal energy tariff does not mean they are behaving stupidly.
This matters. If regulators appear to ‘blame’ consumers for their poor decision-making, then this could in fact have the negative consequence of alienating consumers and further discouraging market engagement. This could in turn further weaken or distort competition. It is better to describe these interventions simply as tools to enhance engagement and consumer decision-making, essentially by making this quicker and easier.
Fourth, engagement interventions of the sort discussed here may not be a panacea. They can certainly help to improve consumer engagement and decision-making, and in doing so enhance competition, but they are unlikely to solve all of the issues in the market, as it can be very hard to create tools that fully engage all consumers or generate perfect decision-making.
This point is of particular relevance within markets that are characterized by a mix of more engaged customers and less engaged customers. As discussed above, where we observe such differences in relative consumer engagement we may find that proposed interventions are readily able to enhance engagement – and thus competition – for the already more engaged customers, but that raising engagement levels amongst the unengaged can be rather more difficult. As discussed above, where such markets are also characterized by price discrimination, increased engagement by the already engaged may have no beneficial impact on the less engaged, and may even result in their facing higher prices (once supplier reactions are taken into account).
In assessing the overall impact of any demand-side intervention, it is therefore important to think about distributional consequences across consumers. Considerations of fairness and equality may play a particular role when the less engaged consumers are also, on average, more vulnerable.
Fifth, where demand-side interventions are imperfect, and especially where this has significant distributional consequences, regulators face a difficult choice. Do they rely on such measures, and accept the imperfect market outcomes, or do they reach for ‘outcome control’ measures, such as price regulation, which are more interventionist?
In recent years, UK regulators (and the UK government) have shown an increasing willingness to consider such a policy response, with price regulation recently being introduced into: the payday lending market (FCA, 2014); that part of the telephony market where customers take a landline only (that is, with no broadband) (Ofcom, 2017b); that part of the energy market where customers pay by pre-payment meters (CMA, 2016b); and now across the wider energy market (Ofgem, 2018b).
Such price regulation has a very different basis to the price regulation that was prevalent in many countries following the privatization of public utilities, which reflected concerns about market power due to supply-side concentration. Here, the concerns are about market power due to demand-side issues, which in turn reflect behavioural factors. In each UK case cited above, the driver of the market problem is either weak engagement by consumers – typically, a subset of more vulnerable customers – or (in the case of payday lending) poor consumer decision-making in the presence of significant behavioural biases. There is an intention, across these cases, to relax or remove the price regulation at a later stage once a way has been found to increase consumer engagement. Extensive testing of possible engagement interventions is currently underway.
Price regulation – in the form of imposing price caps - can clearly be beneficial in terms of reducing the highest prices paid by consumers in a situation where competition for the less engaged customers is ineffective. However, it should be noted that such price regulation runs the risk of weakening competition for engaged customers. The more that regulation ensures that consumers get a good deal, irrespective of their engagement efforts, the lower will be the incentives of consumers to engage, and this will in turn tend to reduce competition. This effect could potentially even be sufficient to mean that industry profits rise – and total consumer outlays increase – when price regulation is introduced.
As such, there is an ongoing debate within the UK about how to deal with markets characterized by price discrimination, alongside substantial differences in relative customer engagement. This debate has been precipitated by concerns about the UK energy market and is focusing on whether there are smarter customer engagement interventions available to alleviate the problems in these markets, or indeed whether there are smarter forms of outcome control regulation that might not have the downsides of imposing price caps.
Conclusion
In brief conclusion, consumer-facing interventions that are designed to enhance consumer engagement are becoming increasingly prevalent and topical. In 2018, they were the subject of an Organisation for Economic Co-operation and Development (OECD) roundtable discussion and also a conference held by the UK Competition Network.Footnote 5
This article discusses some key categories of such interventions, some important policy issues and some key lessons for policy-makers, drawing on recent experience from the UK. In particular, it emphasizes the importance of consumer testing to ensuring that such interventions are effective.
However, the article also highlights that while existing consumer engagement interventions can be valuable to enhancing competition, they may not be a panacea. This is especially a problem in markets characterized by substantial differences in relative customer engagement combined with price discrimination. More work is needed to develop both better customer engagement interventions and better approaches to outcome regulation in markets where it is unrealistic to raise engagement levels further.
Acknowledgements
This work was supported by the Economic and Social Research Council (grant number ES/P008976/1). Amelia Fletcher also acknowledges 2016 financial support from UK consumer organisation Which? for her early work in this area (see Fletcher, Reference Fletcher2016). She thanks James Edgar, Stefan Hunt, Morten Hviid, Adam Land, Bruce Lyons and Lionel Page for useful comments. Amelia is a non-executive director on the Boards of the UK Financial Conduct Authority, Competition and Markets Authority and Payment Systems Regulator, and a decision-maker on enforcement cases at Ofgem. All views are, however, her own and do not necessarily reflect the views of any organization she is associated with.
