Delivering Better Policies Through Behavioural Insights

Chapter 3. Protecting digital consumers

Abstract

This chapter explains how behavioural insights (BI) can help understand and address the impact of online advertising on consumers. It also proposes practical next steps for policymakers to improve consumer understanding of online disclosures and explores how experimental approaches can help tailor disclosures to consumers regarding personalised pricing.

Introduction

During the last decade, policymakers have substantially increased the use of behavioural insights (BI) in the design and delivery of consumer policy (OECD, 2017a). This has complemented more traditional approaches and improved consumer policymaking by making it a more evidence-based discipline and enhancing the design of consumer policy interventions (OECD, 2017a). Methods such as behavioural experiments and surveys have been helping policymakers better understand various policy questions and providing the evidence with which to help address those questions.

In parallel with the growing interest in BI and consumer policy, the OECD Committee on Consumer Policy (CCP) has incorporated behavioural approaches as part of its broader work on effective consumer policymaking. In 2005, the CCP’s first roundtable on economics for consumer policy recognised the potential for behavioural economics to offer a new source of insights for consumer policy, especially in the realm of information disclosure policies (OECD, 2006). In 2010, the CCP released the Consumer Policy Toolkit, which provided consumer authorities advice on how to define and respond to consumer problems, a clear six-step framework for policymaking and learnings from the fields of behavioural economics on consumer policymaking (OECD, 2010). Building on the toolkit, the CCP has released several publications focused on how governments and other public policy organisation can and have applied BI.

The CCP has found that the use of BI in consumer policy has mainly been in three areas: i) information disclosure and labelling; ii) regulation; and iii) consumer empowerment.

i) Information disclosure and labelling have been the most common, typically for price representation and in e-commerce (OECD, 2017a). Some consumer authorities have utilised BI to inform their understanding of deceptive and unfair commercial practices. For example, there have been several enforcement actions that relate to drip pricing in online markets – a practice that can trigger behavioural biases.

ii) Regulations have also been designed with a view to limiting the ability of businesses to take advantage of certain behavioural biases. For example, the latest EU Consumer Rights Directive (adopted in 2011) bans the use of pre-checked boxes for online sales, e.g. for express delivery options and travel insurance contracts when buying airline tickets (European Commission, 2014). This ban was informed by behavioural literature and its recognition of the power of default options (OECD, 2017a).

iii) In the area of consumer empowerment initiatives, some consumer authorities have sought to support consumers by providing them with tools to mitigate the effects of behavioural biases. For example, there have been cases where businesses have been required to provide consumers with a simplified version of a consumer contract with the aim of overcoming information overload. In other cases, consumers have been gaining access to their consumption data in machine (i.e. computer) readable formats.1 This has enabled intermediary services to provide actionable insights based on consumption patterns and transaction histories and support straightforward, effective comparison and decision-making in complex markets (and/or to automate these processes on behalf of consumers). Consumer education initiatives have also been informed by BI and designed with behavioural biases in mind (OECD, 2017a).

Despite the potential benefits, applying BI to certain aspects of consumer policy has not always been straightforward. This has especially been the case when the overarching aim is linked to an abstract or high-level objective. For example, the objective may be to empower consumers to make better decisions through improved online disclosures but it may be unclear whether a consumer should buy, not buy or shop around.

This contrasts with many of the broader policy areas in which BI have been applied, where policymakers (or indeed businesses) often have a clear behaviour that they want to “nudge” consumers or citizens towards (Thaler and Sunstein, 2008). Examples of this type of intervention include auto-enrolment for workplace pensions, along with policy interventions that aim to increase tax compliance (Hernandez et al., 2017), reduce energy use (Allcott, 2011), increase organ donation (Johnson and Goldstein, 2004), reduce litter (Kolodko, Read and Taj, 2016), or promote healthy eating (Shahnazari et al., 2016). In these scenarios, it is likely that policy interventions can be better-tailored and outcomes more easily measured than for, say, the improved online disclosure example given above.

Further, while many of the biases that have been uncovered by BI may be relevant to consumer issues broadly, this does not necessarily imply that wholesale changes to consumer policy are warranted. Instead, BI should be considered in the context of the specific policy question under investigation. In particular, findings from behavioural experiments cannot necessarily be generalised beyond the specific policy question that an experiment was designed to address (OECD, 2018).

Applying BI to consumer policy may also raise new challenges. For instance, nudge-based behavioural interventions may attract criticism if they are perceived to be a form of manipulation. Consumer authorities that are new to BI may also face practical challenges when starting to work in this area, e.g. if they need to dedicate time and additional resources to it, including for capacity building (OECD, 2017a).

This chapter examines the application of BI to consumer policymaking, specifically in online advertising, disclosure agreements and personalised pricing. It brings together and showcases various strands of the CCP’s BI work, including a planned future experiment. The chapter first provides an example of how policymakers can utilise BI to analyse the consumer impact of online advertising. The discussion provides an overview of the CCP’s work in this area, which utilises the existing BI literature to supplement its analysis of the practice.

Second, the chapter examines one of the most common consumer policy interventions: online disclosure requirements. The analysis shows how BI can determine consumer reactions to online disclosures. It also discusses the CCP’s recommendations on designing more effective online disclosures that are based on an understanding of consumers’ behavioural biases.

Third, the chapter explores the behavioural biases that are relevant to personalised pricing and provides initial thoughts on an experimental approach to examine consumer reactions to personalised pricing.

Finally, the chapter provides concluding remarks on the role of BI and consumer policy.

Context and problem setting

Using behavioural insights to better understand the consumer experience: Online advertising and consumer biases

In its analysis of online advertising, the CCP has utilised the BI literature to examine consumer biases and its implications (OECD, 2019). This approach is an example of how policymakers can use existing research on BI as an analytical lens through which a range of market practices and their impact on consumers can be viewed. Consideration of the behavioural lens is valuable because the origins of a problem can sometimes lie in consumers’ behavioural biases or in firms’ behaviours – which, as this section highlights, can act in ways that exploit consumers’ behavioural biases.2

This section reproduces relevant aspects of the CCP’s analysis and highlights two key ways in which the supply-side of the online advertising ecosystem can manipulate consumers’ behavioural biases.

First, it shows how online advertising might present new ways to mislead consumers regarding the full costs of a product or service, or in respect of unexpected terms and conditions of a sale, in part due to behavioural biases such as the anchoring and endowment effects. Second, it looks at how the default or “status quo” bias might result in consumers disclosing and sharing more personal information than they would choose to, had they actively considered the choice. To provide context, a short overview of online advertising is provided below.

Online advertising: An overview

Advertising is always seeking to persuade, encourage or manipulate consumers into making purchases. It has long employed psychologists and other behavioural scientists in pursuit of these objectives, using them to “probe deep into consumers’ minds and build advertising campaigns based on what they found there” (Clay, 2002). Vance Packard’s seminal text The Hidden Persuaders (1957) attests to how the advertising industry was seeking and applying BI decades before governments and policymakers embraced the potential.

Online advertising is now the dominant form of advertising in many OECD countries and offers businesses the ability to reach consumers in ways that could only have been imagined previously. Online advertising has the potential to benefit consumers through more relevant and timely advertising and by funding a host of “free” online services. However, it also raises some new and complex challenges for consumers and consumer authorities (OECD, forthcoming).

As highlighted in Box ‎3.1, the OECD 2016 Recommendation on Consumer Protection in E-Commerce (OECD, 2016) includes provisions relating to advertising and marketing. In general, these provisions are intended to ensure that consumers understand when they are dealing with online advertising and that such advertising is not false or misleading. There is also a particular focus on consumer protection issues that can be challenging for consumers in the online context, such as pricing, digital content and endorsements (OECD, forthcoming).

Box ‎3.1. Selected OECD e-commerce recommendation principles on online advertising

• Advertising and marketing should be clearly identifiable as such (para 13).

• Advertising and marketing should identify the business on whose behalf the marketing or advertising is being conducted where failure to do so would be deceptive (para 14).

• Businesses should ensure that any advertising or marketing for goods or services are consistent with their actual characteristics, access and usage conditions (para 15).

• Businesses should ensure that advertised prices do not misrepresent or hide the total cost of a good or service (para 16).

• Endorsements used in advertising and marketing should be truthful, substantiated and reflect the opinions and actual experience of the endorsers. Any material connection between businesses and online endorsers, which might affect the weight or credibility that consumers give to an endorsement, should be clearly and conspicuously disclosed (para 17).

• Businesses should take special care in advertising or marketing that is targeted to children, vulnerable or disadvantaged consumers, and others who may not have the capacity to fully understand the information with which they are presented (para 18).

Source: OECD (2016), OECD Recommendation of the Council on Consumer Protection in E-Commerce, pp. 11-12, http://dx.doi.org/10.1787/9789264255258-en.

While online advertising shares the same objectives as its analogue forebears, the means of achieving these are radically different. Developments in artificial intelligence (AI) and machine learning, coupled with the routine collection of massive amounts of personal data by online services, allow for the creation of highly detailed profiles about individual consumers, which in turn enables cost‑effective, precision-targeted (and retargeted) advertising at an unprecedented scale.

This hyper-personalised advertising at scale has been referred to as Online Behavioural Advertising (OBA), online profiling and behavioural targeting. Boerman et al. (2017) define OBA as “the practice of monitoring people’s online behavior and using the collected information to show people individually targeted advertisements”. The types of information that are being used in OBA include age, gender, location, education level, interests, online shopping behaviour and search history (Boerman, Kruikemeier and Zuiderveen Borgesius, 2017).

Complementary technologies track user interaction with online ads to determine the effectiveness of advertising campaigns; and to provide the infrastructure for advertising payments to be tied to specific user outcomes, such as “clicks”, webpage visits or purchases. In terms of form, the Internet allows for new ways in which to present text, images, video and audio, and provides for interactive and individually tailored advertising in ways that no prior medium could support.

Given the above, it is not surprising that the Internet has transformed the nature and form of advertising and, as a result, disrupted the advertising and marketing sectors. Recent growth in advertising revenues is being driven by double-digit growth in online advertising (Letang and Stillman, 2016). In the United States, spending on online advertising is expected to exceed spending on television advertising in 2017 (Schuuring et al., 2017). Advertising is Google’s primary revenue source, accounting for USD 79 billion in 2016 (Statista, 2018).

Benefits and risks for consumers

As the CCP’s analysis of online advertising highlights (OECD, 2019), it can provide both benefits and risk for consumers. Benefits include the potential for more targeted, relevant and timely ads that could see consumers benefit from reduced search costs, greater awareness of relevant products and identification of and access to better deals. Online advertising also funds a range of nominally free online services for consumers, including: search services (e.g. Google); social networking services (e.g. Facebook); and digital news outlets (e.g. HuffPost). These have become a part of the fabric of people’s digital experience. If such services were only available on a paid-for basis then some consumers might be worse off.

Risks include longstanding concerns around advertising’s potential to mislead and misrepresent, which can now assume a digital form, along with new concerns that are inherent to online advertising. These include (OECD, 2019):

• consumers may not be able to identify some forms of online advertising

• online advertising could reduce consumer trust online

• online advertising may prey on consumer biases and vulnerabilities

• the potential for misleading advertising online

• threats from “malvertising”

• threats associated with increased data collection.

Online advertising can take advantage of consumer biases and cause consumer detriment

The ability of online advertising to target consumers’ behavioural biases at scale and to potentially tailor the ads to a consumer’s specific vulnerabilities, means that consumer decision-making may be more prone to manipulation through online advertising than for other forms of advertising. Further, online advertising may present new ways in which to mislead consumers regarding the full costs of a product or service, or in respect of unexpected terms and conditions of a sale (OECD, 2019). Consumer authorities will have to remain vigilant to these potential threats. As outlined in Box ‎3.2, there are several behavioural biases that are broadly applicable to the sphere of consumer policy. The relevant behavioural biases specific to online advertising are examined below.

Box ‎3.2. Examples of behavioural biases in consumer policy

Anchoring: Consumers “anchor” decisions around information that they think is the most important. Consumers may fail to adjust their perception of the value of the offer sufficiently, even when additional information is provided to them since they cannot stray far from the anchor point.

Availability heuristic: This describes the tendency for consumers to make judgments about the likelihood of an event based on how easily they can recall a relevant example.

Choice/information overload: When faced with either complex products or a bewildering array of choices, consumers can sometimes ignore possible choices, walk away from markets, or choose not to choose. Consumers can also rely on relatively simple “rules of thumb” or “heuristics” to make decisions.

Confirmation bias: This is the tendency of individuals to seek or interpret evidence in ways consistent with their existing beliefs, expectations or a particular hypothesis.

Default and status quo effect: Presenting one choice as default option can induce consumers to choose that option. The power of default is related to the status quo effect, where consumers have a strong tendency to remain at the status quo since the disadvantages of departing from it loom larger than the advantages of doing so.

Endowment effect: Consumers often demand much more to give up an object than they would be willing to pay to acquire it. The value of a good for consumers increases when it becomes a part of a consumers’ endowment.

Fairness: Consumers are generally concerned that market transactions should be fair to other consumers and often concerned about the conditions of supply (e.g. labour condition, use of environmental resources). This means that consumers are concerned not only about their own interest.

Framing: Consumers are influenced not only by the content of the information provided by suppliers but also by how the information is presented. Presenting an option in a certain way may induce consumers to evaluate the choice from a particular reference point.

Hyperbolic discounting/myopia: Consumers’ discount rate tends to rise steeply the shorter the time period being considered. This means that consumers tend to treat the present as if it were more important than other time periods. This explains outcomes such as low retirement savings in the absence of compulsion.

Loss aversion: See endowment effect (above).

Overconfidence: Consumers tend to think that they are more likely to experience an outcome from some action that is better than the average expected outcome. For example, many drivers think that they are safer than the average person, and when consumers are told that 20% of customers will benefit from a particular product, they tend to expect that they will be the part of that 20%.

Priming effect: When consumers are repeatedly exposed to certain objects, for example, through publicity, certain attributes can play an undue role in consumer decisions. Priming can influence preferences by making certain dimensions salient that would otherwise have been considered as less important.

Social norms: Consumers are often guided by the values, actions and expectations of a particular society or group. For example, when people are made aware of what others are doing, it can reinforce individuals’ underlying motivations.

Time-inconsistency: While traditional economics assumes that consumers behave in a time-consistent way, i.e. that they are able to make decisions knowing their long-term interest and resist short-term actions that go against that, in reality, choices are not consistent across time periods. Consumers may face a conflict between short-term urges and long-term interests.

Sources: OECD (2017a), Use of Behavioural Insights in Consumer Policy, http://bit.ly/2Ic01fJ citing Kahneman, D., J.L. Knetsch and R.H. Thaler (1991), “Anomalies: The Endowment Effect, Loss Aversion, and Status Quo Bias”, Journal of Economic Perspectives, Vol. 5(1), pp 193-206; OECD (2006), “The roundtable on demand-side economics for consumer policy”, https://bit.ly/2Q7U0UK; OECD (2007), Roundtable on Economics for Consumer Policy: Summary Report, www.oecd.org/sti/consumer/39015963.pdf OECD (2010), Consumer Policy Toolkit, https://doi.org/10.1787/9789264079663-en; Office of Fair Trading (UK) (2012), Drip Pricing: UK Experience, https://bit.ly/2R1xNMs; McAuley, I. (2013), “Behavioural economics and public policy: Some insights”, International Journal of Behavioural Accounting and Finance, Vol.4(1), pp. 18-31; Oxera (2013), Behavioural Economics and its Impact on Competition Policy: A Practical Assessment with Illustrative Examples from Financial Services, www.oxera.com/Oxera/media/Oxera/downloads/reports/Behavioural-economics-and-its-impact-on-competition-policy.pdf?ext=.pdf; Shafir, E. (2008), “A behavioural perspective on consumer protection”, Competition and Consumer Law Journal, pp. 302-317; Behavioural Insights Team (UK) (2014), EAST: Four Simple Ways to Apply Behavioural Insights, www.behaviouralinsights.co.uk/publications/east-four-simple-ways-to-apply-behavioural-insights/.

Endowment, anchoring and framing effects

Practices such as drip pricing and bait pricing may mislead consumers, especially given these practices can take advantage of consumer biases (OECD, 2018). Drip pricing occurs where a company advertises its product at a certain (low) price but later adds on additional non-avoidable fees or surcharges. Drip pricing preys on the anchoring and endowment effects. It can result in consumer loss as consumers feel committed to the purchase decision and stick with it despite the price increasing during the transaction process. According to one online study, it can increase purchase intentions, price and value satisfaction and reduce search intentions (Xia and Monroe, 2004). Bait pricing occurs where a business advertises a product at a low price to attract consumers to their store/site but there is only a low volume of products on sale (that sell out) and consumers end up making a more expensive purchase once there. Like drip pricing, this also makes use of the endowment effect and may be detrimental to online consumers (Ellison and Fischer Ellison, 2009). For these reasons, a number of jurisdictions have taken enforcement action against, or have laws that prohibit, drip pricing and bait pricing (OECD, 2018).

The BI literature also indicates that anchoring and framing effects may also be relevant with regards to the identification of online advertising. They could have an impact in relation to native3 and user-generated4 advertising when commercial messages largely look the same as the other content on a site. This could mean that consumers might not understand that such content is advertising and may give the information it conveys greater weight than they otherwise would have given it, due in part to anchoring or framing effects. Anchoring could lead consumers to make mistakes in valuing an offer or in comparing offers (OECD, forthcoming).

To the extent that a business has personal information about a consumer, this may also give them the ability to anchor or frame an advertisement in a way that highlights the characteristics of the product or service that the consumer values, while downplaying other characteristics. This type of conduct could be harmful if it results in consumers being misled or deceived. This could be through misrepresenting the true (financial) cost of a good or service, or by failing to highlight unexpected terms and conditions (for example, where the behaviour amounts to a subscription trap).

Use of social norms and persuasion profiling

Advertisements can use social norms to encourage sales. Social norms can be effective since consumers are often guided by the values, actions and expectations of whatever society or group they consider themselves to be a part of.

Some commentators have raised concerns about online advertisers using “persuasion profiling” to take advantage of the social norms that resonate best with a particular consumer. Such persuasion profiling could be used to target a consumer in real time, given that access to personal information enables a business to know about the consumer’s habits, current location and general vulnerabilities (Calo, 2014). Such targeting could take advantage of time-inconsistency biases, where consumers pursue short-term urges at the expense of their long-term interests. If this form of targeting is used to mislead consumers, then there could be potential for consumer harm (OECD, forthcoming).

Default and status quo bias

Default and status quo biases may lead consumers to disclose and share more personal information than they would otherwise choose to, given the tendency for individuals to go along with whatever the default (or status quo) choice or setting is, even when this may not be in their best interest. Default privacy settings that lead to a high level of disclosure and sharing could hence result in consumers disclosing and sharing more personal information than they would choose to, had they actively considered the choice (Calo, 2014). Conversely, default privacy settings that are more protective of consumers may be an effective way to improve consumer privacy online (OECD, forthcoming).

Using behavioural insights to further understand the impact of online advertising

The OECD’s analysis notes that several of the potential risks associated with online advertising could be better understood by undertaking behavioural experiments to determine the extent of the problem, and possibly to test potential solutions (OECD, forthcoming). In particular, it notes that behavioural experiments could be undertaken to:

• Test consumer identification of online advertising (building on the work already undertaken in Korean, Norway and the United States).

• Test consumer understanding of privacy statements to better understand when such statements might result in consumer harm by misleading consumers.

• Test how consumers react to receiving personalised ads or prices if businesses are required to communicate how those ads or prices have been personalised to the consumer.

• To test the effectiveness of measures that target behavioural biases (for example, “scarcity cues”).

Where issues can potentially be addressed through improved disclosure, the CCP’s recent work on improving online disclosures with BI is relevant (see more below).

Improving interventions with behavioural insights: Online disclosures

Online disclosure requirements are one of the policy interventions most commonly implemented by consumer authorities. The analysis presented in this section is based on the CCP report Improving Online Disclosures with Behavioural Insights, which provides a more detailed discussion (OECD, 2018).

The objective of the report was to use a BI lens to assess how consumers react to online disclosures. Based on this assessment, the CCP also made high-level recommendations on enhancing the design of online information disclosures in ways that could help ensure consumers were better-empowered online. This approach provided a demonstration of how BI could be used to evaluate a policy intervention.5 As such, it is in keeping with Step 6 of the Consumer Policy Toolkit (OECD, 2010), which sets out the need for a policy review process to evaluate the effectiveness of a policy.

This section reproduces key parts of the CCP’s analysis from its recent report as well as its key recommendations. To provide context, it starts with a short overview of online information disclosures.

The importance of online information disclosure: An overview

Access to good information is essential if consumers are to make decisions in their best interests when shopping online. For this reason, information disclosure requirements have been a key policy tool for empowering online consumers across the OECD. The importance of online disclosure is reflected in the OECD 2016 Recommendation on Consumer Protection in E-commerce (OECD, 2016), which dedicates an entire section to this subject.

Online information disclosures provide information about the seller, the goods and services on offer and the transaction itself, including information about payment methods, privacy policies and available dispute resolution and redress options (OECD, 2016). Businesses may make such disclosures through advertising and marketing (ranging from display banners to embedded “native advertising” in online blogs, social media sites and news sites), contractual terms and conditions, and legally‑required notices. This information can be conveyed in different ways, including through pop-ups, links, text, images, audio and video. Businesses can provide information at different times during the customer journey, including through pre-transaction advertising and marketing, and during the course of the transaction, including during the payment process.

While online disclosures may be relatively new, economists have long focused on the role of information in correcting market failures that harm consumers. In particular, information economics recognised that, if left to the market, consumers may not always have enough or the right type of information to make informed decisions. This is especially relevant where there are “information asymmetries” (i.e. when sellers know more about the features and quality of their products or services than consumers). Further, the information economics literature recognises that there are costs involved for businesses in providing information and for consumers in searching for and understanding information. It, therefore, posits that disclosures that make product pricing and features more transparent could reduce search costs, potentially improving consumer outcomes. For these reasons, most OECD countries have legislation or guidelines that ensure that consumers have access to clear, accurate and easily accessible information when shopping online (and off) (OECD, 2018).

The online shopping experience can differ significantly from “bricks and mortar” retail. For example, while consumers may not be able to touch and feel products when shopping online, they usually have access to a wider variety of information concerning those products and may be able to sample digital content. While there may be more information online, however, consumer attention remains a scarce resource both online and off. Further, what works for a recipient of a printed disclosure may not work when transferred to another recipient’s “screen of choice”, be it a computer monitor, tablet or mobile device (Benartzi and Lehrer, 2017). Different delivery channels require businesses to revise both the format and content of information.

Conventional approaches to online disclosure have often assumed that well-informed consumers will reliably make decisions that are in their best interests. However, as highlighted below, consumers can be influenced by behavioural biases that might limit the effectiveness of some forms of online disclosures. Further, BI can highlight why some of the disclosure tactics employed by certain businesses can be effective in eliciting consumer behaviour that is not always in a consumer’s best interest.

As the following highlights, while information disclosure is likely to remain a key tool for empowering consumers, findings from BI indicate that a rethink is required about the usefulness of certain forms of information disclosure.

Consumers’ behavioural biases relevant to online disclosures

The CCP’s analysis identified several behavioural biases that are relevant to consumers’ interactions with online disclosures. The following highlights how consumer detriment can arise when these biases affect whether and to what extent consumers engage with and comprehend disclosures online. Further, in instances where a business fails to adequately disclose either the true price of a product or service or the terms on which the offer is available, these biases can expose consumers to practices that are deceptive, misleading, fraudulent or unfair.

Information overload

Numerous studies have found that consumers are particularly prone to information overload when shopping online (Benartzi and Lehrer, 2017; Office of Fair Trading (UK), 2007). One way in which information overload can manifest itself is in few consumers reading online terms and conditions in full, if at all (see Box ‎3.3 below).

Further, businesses can potentially take advantage of information overload by making their products, services or prices more complex than required. Bar-Gill (2012) has raised concerns about this in the credit card, mortgage and mobile phone markets.

Anchoring and framing effects

With regards to framing, several studies have shown that the timing, context, layout and form of information disclosures can influence consumers’ ability to comprehend them (Federal Trade Commission (US), 2016).

Anchoring can mean that consumers do not value the entire offer properly, even when additional information is provided. This can lead to sub-optimal choices and consumer detriment. One common anchor point, especially when consumers are also facing information overload, is price.

Reference pricing (which compares a sale price to a pre-sale or competitor’s price) can use framing and anchoring effects to inflate the perceived value of an offer. If reference prices are misleading, this has the potential to cause consumer detriment and to distort market outcomes (Office of Fair Trading (UK), 2010). While traditional economic theory suggests this should not have any impact, behavioural studies show that reference prices influence consumers’ assessment of value (Ahmetoglu et al., 2010). Drip pricing and bait pricing can also prey on this behavioural bias.

Box ‎3.3. Do consumers read online terms and conditions?

There is growing evidence and acceptance that most consumers do not read online terms and conditions (T&Cs) in full when making online purchases. Estimates of readership vary depending on the presentation of T&Cs, the product or service they relate to and the way readership is measured (OECD, 2018).

For example, research undertaken for the European Commission found that while between 90% and 95% of consumers accept online terms and conditions, very few read these in full. Readership varied depending on how the terms and conditions were presented. Where consumers had to click through to terms and conditions, only 9.4% opened them, whereas 77.9% of consumers said that they at least scanned terms and conditions that could be scrolled through (Elshout et al., 2016).

The 2017 Ipsos Global Trends survey found that across the 23 countries featured, 64% of respondents agreed with the statement that: “I often don’t bother fully reading terms and conditions on a website before accepting them” (Ipsos, 2017). However, self-reporting by consumers may be prone to overstating the actual figure. Server-side surveys indicate that barely 1% of consumers actually read terms and conditions (Ipsos, 2014).

The type of product also influences readership, with higher reported readership rates for mortgages (73%) and car rentals (72%), and lower rates for transactions on peer platforms (17% read them carefully) (Stark and Choplin, 2009; OECD, 2017b).

Readership of End User License Agreements (EULAs) appears to be even lower, with only 0.2% of consumers accessing EULAs (Bakos et al., 2014), a situation that the requirement of obtaining consumer consent does little to improve (Marotta-Wurgler, 2012). It has been calculated that the median time spent on software EULAs is 6 seconds, with at least 70% of users spending less than 12 seconds on the license page (Sauro, 2011).

A documentary film concerning online terms and conditions asserts it would take an average of one month per year for consumers to read the terms they are presented with online (Hyrax Films, 2013).

Sources: OECD (2018), “Improving online disclosures with behavioural insights”, https://doi.org/10.1787/39026ff4-en (accessed on 10 August 2018); Elshout, M. et al. (2016), European Commission (2016), Study on Consumers’ Attitudes Towards Terms and Conditions (T&Cs) Final Report, http://bit.ly/2FrybKl (accessed 4 November 2018); Ipsos (2014), Global Trends 2014 - Navigating the New, http://bit.ly/2FEMmiA (accessed 4 November 2018); Stark, D.P. and J.M. Choplin (2009), “A license to deceive: Enforcing contractual myths despite consumer psychological realities”, https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1340166 (accessed 4 November 2018); Bakos, Y., F. Marotta-Wurgler and D.R. Trossen (2014), “Does anyone read the fine print? Consumer attention to standard-form contracts”, http://dx.doi.org/10.1086/674424 (accessed 4 November 2018); Marotta-Wurgler, F. (2012), “Does contract disclosure matter?”, https://bit.ly/2p0ZVyw (accessed 4 November 2018); OECD (2017b), Trust in Peer Platform Markets - Consumer Survey Findings, https://doi.org/10.1787/1a893b58-en (accessed 4 November 2018); Sauro, J. (2011), Do Users Read License Agreements?, https://measuringu.com/eula/ (accessed 4 November 2018); Hyrax Films (2013), Terms & Conditions May Apply (Film), http://tacma.net/ (accessed 4 November 2018).

The endowment effect and loss aversion

As noted below, drip pricing preys on the endowment effect (Office of Fair Trading (UK), 2010). Once consumers have decided to make a particular online purchase (especially one involving considerable search time and effort), that decision effectively becomes part of their endowment. That is, in their mind, they have already acquired the good or service in question. If the cost of the purchase then increases beyond the price disclosed in the advert due to drip pricing, loss aversion may make it more difficult for consumers to step away from the purchase (OFT, 2012). This could result in consumers making purchases that are not in their best interests; that is, consumers may not shop around enough and may make purchases at higher prices than they would otherwise. Drip pricing can also make it much more difficult for consumers to understand and compare final prices (Greenleaf et al., 2016).

Bait pricing can also take advantage of the endowment effect and be detrimental to consumers (Ellison and Fischer Ellison, 2009). It occurs when a business advertises a product at a low price to attract consumers to its website but fails to disclose that only a limited quantity of the advertised product is available. If that limited quantity has sold out by the time an interested consumer reaches the website, the endowment effect may lead the consumer to purchase a more expensive alternative from the business rather than seeking out the original item elsewhere or comparing prices for the more expensive alternative.

Default biases

Default biases may exacerbate issues related to online disclosures. For example, default settings may opt consumers in for additional services with associated fees or charges through the use of pre-checked boxes or negative option marketing (see below). The effectiveness of pre-checked boxes in influencing consumer behaviour can be demonstrated with an example from Goldstein et al. (2008):

“A large national railroad in Europe made a small change to its website so that seat reservations would be included automatically with ticket purchases (at an additional cost of one to two euros), unless the customer unchecked a box on the online booking form. Whereas 9% of tickets included reservations before the change, 47% did after, earning the railroad an additional [US dollars] $40 million annually.”

Similarly, default settings are relevant to negative option marketing,6 which has raised concerns for several consumer agencies across the OECD. In a Federal Trade Commission (FTC) workshop on the topic, panellists “discouraged the use of pre-checked boxes to obtain consumer consent because online research indicates consumers ignore them” (FTC, 2009). A representative from the National Advertising Division of the Council of Better Business Bureaus noted that pre-checked boxes signal that the information is routine or unimportant and hence, they are not an effective way of communicating with consumers (FTC, 2009)

To the extent that pre-checked boxes or other default settings (including negative option marketing) automatically sign consumers up for additional goods or services, financial commitments, disclosure of personal data or marketing material, it is likely in at least some circumstances that a significant proportion of consumers will fail to uncheck these options despite not actually wanting them or agreeing with them. This has great potential to result in consumer detriment. For example, consumers may be billed for goods or services they do not want, they may unwittingly share personal information or they may be hassled by unwanted marketing.

Overconfidence and myopia

Overconfidence and myopia may lead consumers to ignore certain types of information including warnings, disclaimers and T&Cs. These biases may also lead consumers to choose the wrong product or service if businesses take advantage of them, for example, by highlighting immediate benefits such as a “free” mobile phone but obfuscating the cost of this over the lifetime of the contract (Bar-Gill, 2012).

Social norms and other factors

Consumer behaviour in relation to disclosures is likely to be influenced by social and cultural norms. For example, if a consumer’s friends and family use an application, consumers may be less likely to read the T&Cs and check the privacy settings. Further, a consumer’s ability to comprehend online disclosures is likely to be influenced by their age, education and familiarity with the Internet, among other things. It is therefore important to test disclosures on the relevant population of interest.

Policy implications

In addition to identifying where behavioural biases can render online disclosures ineffective, or, in the absence of adequate disclosure, result in consumers being treated unfairly, the CCP also examined the policy implications. In particular, it assessed how BI can help policymakers determine when and how best to use online disclosures and how they can provide the basis for improved disclosure design. The resultant high-level recommendations are summarised below.

Information and pricing should be clear and accurate

Protections against false and misleading information remain important, especially given consumers’ susceptibility to behavioural biases. In particular, consumers need to be protected from misleading pricing practices such as drip pricing and bait pricing. Further, to the extent that reference prices are used, these should not be false or misleading.

Material information should not only be in the terms and conditions

As highlighted in Box ‎3.3, few consumers read online T&Cs in full. For this reason, businesses should not use T&Cs to communicate important information to consumers. Material information should be made clear and salient to consumers, potentially in multiple places on a firm’s website and at various points during the transaction.

Use of images, audio and video should be considered

Information should be made as clear as possible with alternatives to text considered, as appropriate. In some scenarios, images, audio and video can more effectively convey information to consumers than even the clearest and simplest text. As well as considering the use of these alternative media, businesses should consider the overall look of their website, including colour and visual layout, for example.

Timing of disclosures is important

Another key factor to consider is the timing of disclosures. Given the endowment effect, warnings or material information that is only provided towards the end of a purchase may have little impact, since consumers have already made the decision to purchase the good or service. In other scenarios, information that would be quite useful to consumers may be rendered useless if it is received at a time when consumers cannot react. The timing of disclosures is something that should be subject to consumer testing.

Consumer consent should be express

If businesses want consumers to confirm a transaction for goods or services, consumer consent should be expressly sought and obtained. Such consent should not rely on default settings, negative option marketing or pre‑checked boxes that consumers are unlikely to notice or change.

Information should be as simple as possible

Given the potential for information overload, online disclosures should be as simple as possible. Simplicity can be achieved by reducing the amount of text, using “signposts” to direct consumers to relevant information, and the use of “layering”, where different levels of information of increasing detail are presented to consumers as needed (for example, through active links).

Personalised disclosures require further consideration

Another way to improve consumer understanding of online disclosures is to better tailor the message to the individual concerned. In particular, if disclosure is only relevant to an identifiable target group, then it should ideally only be shown to that group. However, personalisation raises other potential consumer policy concerns. For this reason, it is an area that requires further research.

Technology-enabled information provision could facilitate comparison shopping and switching in complex markets

In complex markets, consumers might need additional information in order to compare offers in the market. In particular, consumers might need detailed information about their past or likely future usage. When such information is made available in a machine-readable form, it will be easier for consumers to make use of the services offered by public and private intermediaries.7

The importance of testing

Finally, the CCP’s online disclosure work also emphasises the importance of undertaking consumer testing – including through the use of behavioural experiments – whenever new disclosure requirements are being considered. The following section highlights how the CCP itself plans to use a behavioural experiment to test the impact of different disclosure requirements in relation to the issue of personalised pricing on line.

Methodology

Building on its work on online advertising and online information disclosures discussed in the preceding sections, the CCP plans to run its first behavioural experiment on personalised pricing online. Although there is little empirical evidence that personalised pricing (as opposed to dynamic pricing or personalised search rankings) is occurring in online markets, interest in this subject has grown. It has become clearer that online vendors are capable of tailoring prices to individuals based on granular personal data. The experiment will be an opportunity to test some of this chapter’s recommendations for enhancing the design of online disclosures as listed in the policy implications above. The personalised pricing experiment will also complement the CCP’s analysis of online advertising, as both personalised pricing and targeted online advertising are enabled by and are dependent on consumers’ personal information.

After providing a brief overview of personalised pricing, this section outlines the relevant consumer biases. It then presents the research orientations that will underpin the experiment, before concluding with initial thoughts around the experimental approach.

Once complete, it is anticipated that the report detailing the experiment’s findings will be published and made available at www.oecd.org/sti/consumer.

Personalised pricing: An overview

Definitions of online personalised pricing formulated in the literature typically comprise of three elements: i) that it is a sophisticated form of price discrimination; ii) that it hinges on online vendors’ access to and utilisation of consumer data to generate meaningful inferences of what an individual consumer, or group of consumers, is willing to pay for a given good or service; and iii) prices offered to the individual consumer are set on the basis of these insights.

As such, personalised pricing is distinct from (but risks being conflated with) other forms of dynamic pricing that are prevalent online. For services and products where prices are adjusted in response to availability and/or overall demand yield management strategies for selling products and services that are perishable, time-sensitive and/or scarce. This includes event tickets, flights and inter-city rail travel, surge pricing in app-based cab services, and hotel rooms. These strategies can result in different consumers seeing different prices (or the same consumer seeing different prices if they engage with the vendor at different points in time).

Although coverage of personalised pricing frequently highlights concerns and the potential for consumer detriment, the practice is not inherently problematic. The characteristics of a market (i.e. to what extent is it competitive?) and the motives of each vendor adopting the practice, will determine the extent to which it proves beneficial or harmful for consumers in each instance; and/or whether the harms caused to those made worse off by this form of price discrimination outweigh the gains experienced by its beneficiaries (Office of Fair Trading (UK), 2013).

There are, as yet, few irrefutable occurrences of online personalised pricing and no evidence to suggest it is widespread. This may in part reflect the challenges that consumers and those working in their interest face in detecting the practice; and/or vendor hesitancy in deploying a practice that could fuel a consumer backlash if implemented covertly and then exposed, or that consumers could resent even if deployed transparently.

Given the above, the CCP’s experiment will be developed on the basis of two untested assumptions: i) that personalised pricing happens at least occasionally and has the potential to become more frequent; and ii) that it is not normally detectable by consumers (although they may suspect it in some scenarios).8

The stated difficulty in detecting instances of personalised pricing will mean that – absent effective disclosure – consumers are unlikely to be able to: i) identify a personalised price when they encounter one; or ii) assess whether any personalised price they do encounter is higher or lower than the price they would otherwise have been offered.

Disclosure, therefore, offers a potential means of making the practice transparent to consumers – enabling them to determine whether a personalised price for a given good or service serves their best interests and, as a result, make better-informed purchasing decisions. On that basis, testing the impact of different forms of disclosure of personalised pricing on consumer behaviour will form the crux of the experiment.

Examples of behavioural biases that are potentially relevant to personalised pricing

Framing and loss aversion

Online vendors might exploit this bias by, for example, framing a personalised price offered as being superior to an inflated reference price and/or by creating the impression of scarcity, so that the consumer is compelled to buy through fear of losing out.

Fairness

In addition to considerations of whether a personalised price is fair, consumers might also deem the information asymmetry on which personalised pricing is predicated to be unfair and modify their behaviour accordingly. A recent behavioural experiment examining online targeted advertising and the influence of different approaches to (and degrees of) disclosure on consumers’ purchasing intentions, produced evidence showing that when consumers realise that their personal information is flowing in ways they dislike, purchase interest declines. It also found that when third-party sharing of consumer data had occurred in ways that consumers deemed unacceptable, concerns about privacy outweighed people’s appreciation for ad personalisation. Offering consumers the means to meaningfully control their privacy settings appeared to buffer any backlash to unacceptable data collection (John, Kim and Barasz, 2018). It would be interesting to see whether similar findings emerge in relation to personalised pricing.

Overconfidence

With regards to personalised pricing, it may be possible to observe whether limited forms of disclosure (e.g. the provider informs the consumer they are being offered a price that is unique to them, based on their purchase history for example) lead the consumer to assume the price they are offered is better than the average.

Research objectives and orientations

Ensuring consumers are able to make well-informed decisions is a key objective for effective consumer policy regimes (OECD, 2010). As noted above, by seeking to ensure that consumers have access to and can comprehend the information required to reach a well-informed decision, disclosure policies play a vital role in this respect. It is anticipated that the CCP’s experiment will seek to address two interrelated questions in relation to disclosure, namely:

• Which approaches to disclosure are most effective9 in enabling participants to: i) identify when a transaction is subject to personalised pricing; and ii) comprehend the implications of this practice for the transaction (e.g. how it affects the price they are asked to pay)? This could include testing variations in both the content and form of disclosures.

• To what extent does the disclosure of personalised pricing to consumers have a material impact on their decision-making?10

Proposed experimental approach

A laboratory experiment may be the most feasible option to test these initial questions because it will be in a controlled environment with minimal noise. The results of this laboratory experiment could inform a future field experiment or a natural experiment in the future.

The laboratory experiment will be built around high quality, realistic simulations of e‑commerce and m-commerce sites, potentially resembling those of popular online retailers. It is anticipated that smartphones will be used by at least a proportion of participants, given their growing role in online purchasing. In controlled conditions, participants would engage in a range of online purchasing tasks and be presented with various personalised pricing and disclosure scenarios that support the testing of the various research objectives outlined above.

Conclusion

Policy officials can leverage BI as a tool to analyse the behavioural biases that are relevant to consumer policy. The CCP has applied a behavioural approach to analysing three important areas in consumer policy: online advertising, online disclosures and personalised pricing. The CCP has found that online advertising can take advantage of consumer biases, which can mislead consumers to buy full cost products or disclose personal information. Further behavioural experiments are necessary to better understand the associated risks. In the case of online disclosure, the CPP has used a behavioural lens to assess how consumers react to online disclosures. In addition, they have provided practical recommendations for policymakers on how to apply BI to improve online disclosures. Finally, the CCP has applied this behaviourally-driven approach to inform a future experimental design to understand the most effective ways for the consumer to identify personalised pricing and its implications.    

References

Ahmetoglu, G. et al. (2010), Pricing Practices: Their Effects on Consumer Behaviour and Welfare, Prepared for the Office of Fair Trading, http://webarchive.nationalarchives.gov.uk/20140402172955/http://oft.gov.uk/shared_oft/business_leaflets/659703/Advertising-of-prices/Pricing-Practices.pdf (accessed on 02 August 2017).

Allcott, H. (2011), “Social norms and energy conservation”, Journal of Public Economics, Vol. 95, pp. 1082-1095, https://bit.ly/2AMm6je.

Bakos, Y., F. Marotta-Wurgler and D.R. Trossen (2014), “Does anyone read the fine print? Consumer attention to standard-form contracts”, The Journal of Legal Studies, Volume 43(1), pp. 1-35, http://dx.doi.org/10.1086/674424 (accessed 4 November 2018).

Bar-Gill, O. (2012), Seduction by Contract: Law, Economics, and Psychology in Consumer Markets, Oxford University Press.

Behavioural Insights Team (UK) (2014), EAST: Four Simple Ways to Apply Behavioural Insights, www.behaviouralinsights.co.uk/publications/east-four-simple-ways-to-apply-behavioural-insights/.

Benartzi, S. and J. Lehrer (2017), The Smarter Screen: Surprising Ways to Influence and Improve Online Behavior, Penguin.

Boerman, S., S. Kruikemeier and F. Zuiderveen Borgesius (2017), “Online behavioral advertising: A literature review and research agenda”, Journal of Advertising, Vol. 46/3, pp. 363-376, http://dx.doi.org/10.1080/00913367.2017.1339368.

Calo, R. (2014), “Digital market manipulation”, The George Washington Law Review, Vol. 82(4), pp. 995-1051, http://www.gwlr.org/wp-content/uploads/2014/10/Calo_82_41.pdf (accessed 4 November 2018).

Clay, R. (2002), “Advertising as science”, Monitor on Psychology, Vol. 33/9, p. 38, http://www.apa.org/monitor/oct02/advertising.aspx.

Ellison, G. and S. Fischer Ellison (2009), “Search, obfuscation, and price elasticities on the Internet”, Econometrica, Vol. 77/2, pp. 427-452, http://dx.doi.org/10.3982/ECTA5708 (accessed 30 December 2018)

Elshout, M. et al. (2016), European Commission (2016), Study on Consumers’ Attitudes Towards Terms and Conditions (T&Cs) Final Report, European Commission, Brussels, http://bit.ly/2FrybKl (accessed 4 November 2018).

European Commission (2018), Consumer Market Study on Online Market Segmentation Through Personalised Pricing/offers in the European Union, European Commission, https://ec.europa.eu/info/publications/consumer-market-study-online-market-segmentation-through-personalised-pricing-offers-european-union_en (accessed on 10 August 2018).

European Commission (2014), DG Justice Guidance Document (The Directive on Consumer Rights), European Commission, https://ec.europa.eu/info/sites/info/files/crd_guidance_en_0.pdf.

Federal Trade Commission (US) (2016), Putting Disclosures to the Test Workshop: Staff Summary, https://www.ftc.gov/system/files/documents/reports/putting-disclosures-test/disclosures-workshop-staff-summary-update.pdf (accessed on 02 August 2017).

Federal Trade Commission (US) (2015), Native Advertising: A Guide for Businesses, www.ftc.gov/tips-advice/business-center/guidance/native-advertising-guide-businesses (accessed 30 December 2018).

Federal Trade Commission (US) (2009), Negative Options - A Report by the Staff of the FTC’s Division of Enforcement, https://bit.ly/2CTWimI (accessed 4 November 2018).

Goldstein, D. et al. (2008), “Nudge your customers toward better choices”, Harvard Business Review, Vol. 12, https://hbr.org/2008/12/nudge-your-customers-toward-better-choices (accessed on 08 August 2017).

Greenleaf, E. et al. (2016), “The price does not include additional taxes, fees, and surcharges: A review of research on partitioned pricing”, Journal of Consumer Psychology, Vol. 24/1, pp. 105-124, https://bit.ly/2VkNkWo (accessed 30 December 2018).

Ham, C.-D. and M.R. Nelson (2016), “The role of persuasion knowledge, assessment of benefit and harm, and third-person perception in coping with online behavioral advertising”, Computers in Human Behavior, Volume 62, pp. 689-702, https://bit.ly/2Op8oq3 (accessed 4 November 2018).

Hernandez, M. et al. (2017), Applying Behavioral Insights to Improve Tax Collection: Experimental Evidence from Poland, World Bank, Washington, DC, https://bit.ly/2PaOVyJ (accessed 4 November 2018).

Hyrax Films (2013), Terms & Conditions May Apply (Film), Directed by C. Hoback, http://tacma.net/ (accessed 4 November 2018).

Ipsos (2017), Ipsos Global Trends: Section 8 - Online Behaviour, https://bit.ly/2HJJnUK (accessed 4 November 2018).

Ipsos (2014), Global Trends 2014 - Navigating the New, http://bit.ly/2FEMmiA (accessed 4 November 2018).

John, L., T. Kim and K. Barasz (2018), Ads That Don’t Overstep, pp. 62-69, https://hbr.org/2018/01/ads-that-dont-overstep.

Johnson, E. and D. Goldstein (2004), “Defaults and donation decisions”, Transplantation, Vol. 78, pp. 1713-1716, http://dx.doi.org/10.1097/01.TP.0000149788.10382.B2.

Kahneman, D., J.L. Knetsch and R.H. Thaler (1991), “Anomalies: The endowment effect, loss aversion, and status quo bias”, The Journal of Economic Perspectives, Vol. 5(1), pp. 193-206.

Kolodko, J., D. Read and U. Taj (2016), Using Behavioural Insights to Reduce Littering in the UK, Clean Up Britain, Warwick Business School, http://www.nudgeathon.com/wp-content/uploads/2016/01/CLUB-REPORT.pdf (accessed on 25 August 2017).

Letang, V. and L. Stillman (2016), Global Advertising Forecast Winter Update, Magna.

Marotta-Wurgler, F. (2012), “Does contract disclosure matter?”, Journal of Institutional and Theoretical Economics, Vol. 168(1), https://bit.ly/2p0ZVyw (accessed 4 November 2018).

McAuley, I. (2013), “Behavioural economics and public policy: Some insights”, International Journal of Behavioural Accounting and Finance, Vol. 4(1), pp. 18-31.

OECD (2019), “Online advertising: Trends, benefits and risks for consumers”, OECD Digital Economy Papers, No. 272, OECD Publishing, Paris, https://doi.org/10.1787/1f42c85d-en.

OECD (2018), “Improving online disclosures with behavioural insights”, OECD Digital Economy Papers, No. 269, OECD Publishing, Paris, https://doi.org/10.1787/39026ff4-en (accessed on 10 August 2018).

OECD (2017a), Use of Behavioural Insights in Consumer Policy, OECD, Paris http://bit.ly/2Ic01fJ.

OECD (2017b), Trust in Peer Platform Markets - Consumer Survey Findings, OECD Publishing, Paris, https://doi.org/10.1787/1a893b58-en (accessed 4 November 2018).

OECD (2016), OECD Recommendation of the Council on Consumer Protection in E-Commerce, OECD Publishing, Paris, http://dx.doi.org/10.1787/9789264255258-en.

OECD (2010), Consumer Policy Toolkit, OECD Publishing, Paris, https://doi.org/10.1787/9789264079663-en.

OECD (2007), Roundtable on Economics for Consumer Policy – Summary Report, OECD, Paris, http://www.oecd.org/sti/consumer/39015963.pdf (accessed 4 November 2018).

OECD (2006), “The roundtable on demand-side economics for consumer policy”, OECD Digital Economy Papers, No. 116, OECD Publishing, Paris, https://bit.ly/2Q7U0UK.

OECD (forthcoming), Online Advertising: Trends, Benefits and Risks for Consumers, OECD Publishing, Paris.

Office of Fair Trading (UK) (2013), Personalised Pricing - Increasing Transparency to Improve Trust, OFT, https://bit.ly/2ydgzmo.

Office of Fair Trading (UK) (2012), Drip Pricing: UK Experience, https://bit.ly/2R1xNMs (accessed 30 December 2018).

Office of Fair Trading (UK) (2010), Advertising of Prices, OFT, http://webarchive.nationalarchives.gov.uk/20140402173016/http://oft.gov.uk/shared_oft/market-studies/AoP/OFT1291.pdf (accessed on 02 August 2017).

Office of Fair Trading (UK) (2007), Internet Shopping: An OFT Market Study, OFT, http://webarchive.nationalarchives.gov.uk/20140402163042/http://oft.gov.uk/OFTwork/markets-work/internet (accessed on 01 August 2017).

Oxera (2013), Behavioural Economics and its Impact on Competition Policy: A Practical Assessment with Illustrative Examples from Financial Services, www.oxera.com/Oxera/media/Oxera/downloads/reports/Behavioural-economics-and-its-impact-on-competition-policy.pdf?ext=.pdf.

Packard, V. (1957), The Hidden Persuaders, D. McKay Co.

Sauro, J. (2011), Do Users Read License Agreements?, https://measuringu.com/eula/ (accessed 4 November 2018).

Schuuring, M. et al. (2017), The Digital Marketing Revolution Has Only Just Begun, BCG, https://www.bcg.com/publications/2017/sales-consumer-insights-digital-marketing-revolution-has-only-just-begun.aspx (accessed on 10 January 2018).

Shafir, E. (2008), “A behavioural perspective on consumer protection”, Competition and Consumer Law Journal, pp. 302-317.

Shahnazari, S. et al. (2016), “Promoting healthy eating habits”, MaRS Solutions Lab, Canada, https://www.marsdd.com/wp-content/uploads/2016/03/MSL_ABI-Working-Paper.pdf (accessed on 25 August 2017).

Sloane, G. (2017), Snapchat Gives Advertisers New Tracking Tool to Target Ads, AdAge, http://adage.com/article/digital/snapchat-advertisers-pixel-tracking-tool-target-ads/311129/ (accessed 10 January 2018).

Stark, D.P. and J.M. Choplin (2009), “A license to deceive: Enforcing contractual myths despite consumer psychological realities”, NYU Journal of Law & Business, Vol. Spring, https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1340166 (accessed 4 November 2018).

Statista (2018), Google: Ad Revenue 2001-2016, https://www.statista.com/statistics/266249/advertising-revenue-of-google/ (accessed on 23 January 2018).

Thaler, R. and C. Sunstein (2008), Nudge: Improving Decisions about Health, Wealth and Happiness, Yale University Press, New Haven.

Xia, L. and K. Monroe (2004), “Price partitioning on the Internet”, Journal of Interactive Marketing, Vol. 18/4, http://dx.doi.org/10.1002/dir.20017 (accessed 30 December 2018).

Notes

← 1.  For example, the Midata initiative in the United Kingdom. See: https://bit.y/2hkNYBN.

← 2.  This approach is consistent with Step 1 of the 6 steps presented in the Consumer Policy Toolkit (OECD, 2010).

← 3.  The retargeting of online advertising occurs when a consumer starts seeing advertising for a product or service they have been researching appear on numerous sites across the web. Consumers can feel like the product or service is “following them” around the Internet. Retargeting can take various forms and can be based on different information, such as search activities, responses to online advertisements, responses to email advertisements and “clicks”. For examples, see Sloane (2017).

← 4.  The Federal Trade Commission has defined native advertising as: “content that bears a similarity to the news, feature articles, product reviews, entertainment and other material that surrounds it online” (FTC, 2015).

← 5.  This approach is in line with Step 6 of the Consumer Policy Toolkit (OECD, 2010), which sets out the end for a policy review process to evaluate the effectiveness of a policy.

← 6.  Negative option marketing refers to a category of commercial transactions in which sellers interpret a customer’s failure to take an affirmative action – either to reject an offer or cancel an agreement – as assent to be charged for goods or services. Negative option marketing can pose serious financial risks to consumers if appropriate disclosures are not made and consumers are billed for goods or services without their consent. See: https://bit.ly/2CTWimI.

← 7.  See Note 1, above.

← 8.  The recent behavioural experiment commissioned by the European Commission (2018) found that less than 20% of participants correctly identified price personalisation when they encountered it.

← 9.  In the context of this experiment, disclosure would be deemed effective if it succeeds in enabling participants to identify instances of personalised pricing and comprehend its implications for the transaction at hand.

← 10.  While it is anticipated that the experiment will observe and capture the impact of disclosure (and various forms of disclosure) on participants’ subsequent decision-making behaviours in the simulated transactions, it will not seek to define optimal outcomes in terms of decision-making and measure whether disclosure “nudges” participants towards these.