Online advertising:
the impact of targeted
advertising on advertisers,
market access and
consumer choice
Policy Department for Economic, Scientific and Quality of Life Policies
Directorate-General for Internal Policies
Authors: Niklas FOURBERG, Serpil TAŞ, Lukas WIEWIORRA,
Ilsa GODLOVITCH, Alexandre DE STREEL, Hervé JACQUEMIN,
Jordan HILL, Madalina NUNU, Camille BOURGUIGON,
Florian JACQUES, Michèle LEDGER and Michael LOGNOUL
PE 662.913 - June 2021
EN
STUDY
Requested by the IMCO committee
Abstract
In this research paper, we provide a comprehensive overview of
online advertising markets and we analyse the challenges and
opportunities concerning digital advertising. We review the degree
to which existing and proposed legislation at EU level addresses the
identified problems, and identify potential solutions, with reference
to experience from EU Member States and third countries. We
conclude with a synthesis and specific policy recommendations,
drawing on stakeholder interviews.
This document was provided by the Policy Department for
Economic, Scientific and Quality of Life Policies at the request of the
committee on the Internal Market and Consumer Protection (IMCO).
Online advertising:
the impact of targeted
advertising on advertisers,
market access and
consumer choice
This document was requested by the European Parliament's committee on the Internal Market and
Consumer Protection.
AUTHORS
WIK-Consult: Niklas FOURBERG, Serpil TAŞ, Lukas WIEWIORRA and Ilsa GODLOVITCH
University of Namur: Alexandre DE STREEL, Hervé JACQUEMIN, Camille BOURGUIGON, Florian
JACQUES, Michèle LEDGER and Michael LOGNOUL
VVA: Jordan HILL and Madalina NUNU
ADMINISTRATOR RESPONSIBLE
Christina RATCLIFF
EDITORIAL ASSISTANT
Irene VERNACOTOLA
LINGUISTIC VERSIONS
Original: EN
ABOUT THE EDITOR
Policy departments provide in-house and external expertise to support European Parliament
committees and other parliamentary bodies in shaping legislation and exercising democratic scrutiny
over EU internal policies.
To contact the Policy Department or to subscribe for email alert updates, please write to:
Policy Department for Economic, Scientific and Quality of Life Policies
European Parliament
L-2929 - Luxembourg
Email: Poldep-Economy-[email protected]
Manuscript completed: May 2021
Date of publication: June 2021
© European Union, 2021
This document is available on the internet at:
http://www.europarl.europa.eu/supporting-analyses
DISCLAIMER AND COPYRIGHT
The opinions expressed in this document are the sole responsibility of the authors and do not
necessarily represent the official position of the European Parliament.
Reproduction and translation for non-commercial purposes are authorised, provided the source is
acknowledged and the European Parliament is given prior notice and sent a copy.
For citation purposes, the publication should be referenced as: Fourberg, N., Taş, S., Wiewiorra, L.,
Godlovitch, I., De Streel, A., Jacquemin, H., Hill, J., Nunu, M., Bourguigon, C., Jacques, F., Ledger, M., and
Lognoul, M., 2021, Online advertising: the impact of targeted advertising on advertisers, market access and
consumer choice, Publication for the committee on the Internal Market and Consumer Protection, Policy
Department for Economic, Scientific and Quality of Life Policies, European Parliament, Luxembourg.
© Cover image used under licence from Adobe Stock
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
3 PE 662.913
CONTENTS
LIST OF BOXES 7
LIST OF FIGURES 8
LIST OF ABBREVIATIONS 9
EXECUTIVE SUMMARY 12
HOW DO ONLINE ADVERTISING MARKETS WORK? 16
1.1. Types of digital advertising 16
1.2. The role of targeting in digital advertising 18
1.3. Main actors and business models 21
1.3.1. Ad-Networks 22
1.3.2. Ad-Exchanges and advertising auctions 24
1.3.3. The role of SMEs 26
1.3.4. Consumers 26
1.4. Evolution of the market and implications of COVID-19 27
CHALLENGES AND OPPORTUNITIES CONCERNING DIGITAL ADVERTISING 31
2.1. Practices and their impacts 31
2.1.1. Practices affecting consumers and SMEs as consumers/viewers of advertisement 31
2.1.2. Practices affecting purchasers of advertising including SMEs 38
2.1.3. Practices affecting suppliers of digital advertising services (publishers and smaller
SMEs functioning as Ad-Networks) 42
2.2. Steps taken by consumers and commercial providers to address perceived challenges
46
2.3. Analysis of costs and benefits of different practices 48
HOW FAR DOES EXISTING LEGISLATION GO IN ADDRESSING THE PROBLEM? 53
3.1. Legal framework 53
3.1.1. Overview of the legal framework 53
3.1.2. Soft Law instruments 55
3.1.3. Complexity of the legal framework 55
3.2. How far do existing substantial rules address the problems and what are the gaps? 57
3.2.1. Processing of personal data and privacy 57
3.2.2. Transparency obligations 61
3.2.3. Internal market 65
3.2.4. Prohibition of certain practices 68
3.2.5. Many players involved and key role played by the platforms 71
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 4
3.2.6. Summary 73
3.3. How far do existing enforcement measures address the problems and what are the
gaps? 73
3.3.1. Overview of the main enforcement measures 73
3.3.2. Main issues and gaps 74
3.3.3. Summary on enforcement 75
3.4. What are the new measures proposed by the Commission in DSA and DMA
proposals? 75
3.4.1. DSA Proposal 75
3.4.2. DMA Proposal 77
WHAT SOLUTIONS HAVE BEEN APPLIED? 79
4.1. Comparison of national context 79
4.2. Design and implementation of responses 81
4.3. Evaluation of EU Member State and third country experiences 85
CONCLUSIONS AND RECOMMENDATIONS 87
5.1. Recommendations to better protect consumers 87
5.1.1. Ensuring that data use and sharing in digital advertising conforms with privacy rules
87
5.1.2. Informing consumers about the fact that they are being targeted and improved
consent mechanisms 88
5.1.3. Addressing “dark patterns” through GDPR guidelines 89
5.1.4. Clarifying that digital targeted advertising must not breach rules on discrimination
and improving algorithmic transparency 89
5.1.5. Ensure that minors are not subject to harmful targeted advertising which exploits
their vulnerabilities 90
5.1.6. Dealing with multiple actors 91
5.1.7. Clarifying the redress mechanisms 91
5.1.8. Facilitating the functioning of the internal market 91
5.2. Recommendations to better protect SMEs as publishers and advertisers 92
5.2.1. Addressing exploitation by platforms which hold a dominant position in digital
advertising 92
5.2.2. Increase transparency concerning advertising auctions and the performance of
advertising 92
5.2.3. Tackling bundling and tying by gatekeeper intermediaries of premium advertising
space 93
5.2.4. Addressing asymmetric access to consumer data 94
5.2.5. Counteracting the monopolisation of the browser market 94
5.3. Summary of conclusions 95
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
5 PE 662.913
REFERENCES 101
ANNEX 1: LEGISLATIVE MEASURES AND PROPOSALS 119
Specific measures applicable to advertising in the digital environment 119
a) eCommerce Directive 119
b) P2B Regulation 120
c) DSA Proposal 122
d) DMA Proposal 124
e) ePrivacy (and Proposal) 125
General measures applicable to advertising 126
a) Unfair Commercial Practices Directive 126
b) Directive on misleading and comparative advertising 127
c) Consumer Rights Directive 128
d) GDPR 129
e) AVMS 130
Tables 132
ANNEX 2: CASE STUDIES 135
Australia 135
1. National context 135
2. Design and implementation of response 136
3. Analysis and Evaluation 137
Germany 138
1. National context 138
2. Design and implementation of response 138
3. Analysis and Evaluation 139
France 141
1. National context 141
2. Design and implementation of response 141
3. Analysis and Evaluation 142
Ireland 143
1. National context 143
2. Design and implementation of response 144
3. Analysis and Evaluation 145
United Kingdom 146
1. National context 146
2. Design and implementation of response 147
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 6
3. Analysis and Evaluation 147
United States 149
1. National context 149
2. Design and implementation of response 150
3. Analysis and Evaluation 151
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
7 PE 662.913
LIST OF BOXES
Box 1: Behavioural targeting and privacy concerns 33
Box 2: Obfuscation and location based targeting of advertisements reduce consumer choice 34
Box 3: Advertisement designs can exploit consumers’ behavioural biases and include dark
pattern elements 35
Box 4: Potential for discrimination and harmful targeting of vulnerable consumers through
Real-Time-Bidding (RTB) 36
Box 5: The prevalence gives rise to threats to cybersecurity in the form of malvertising and
spear-phishing 38
Box 6: Factors that constitute large advertising providers can both harm and benefit advertisers
40
Box 7: Lack of transparency may lead to increased costs and fraud 42
Box 8: Asymmetric access to information and data 44
Box 9: Conducts like exclusive clauses, self-preferencing, bundling and tied sales as well as
impediments to interoperability are associated with large players 45
Box 10: Widespread advertising facilitates the adoption of ad-blocking software (AAT) by users
which threatens the business models of publishers and advertisers 47
Box 11: Main issues and gaps: Processing of personal data and privacy 60
Box 12: Main issues and gaps: Transparency obligations 65
Box 13: Main issues and gaps: Internal market principles 68
Box 14: Main issues and gaps: Prohibition of some practices 71
Box 15: Main issues and gaps: Many players involved and key role played by the platforms 72
Box 16: Effect of COVID-19 on online advertising in France and Australia 81
Box 17: The path of online advertising regulation in Australia 83
Box 18: Ireland, an example of mixed regulatory systems 84
Box 19: The regulatory practice on online advertising in the United States 85
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 8
LIST OF FIGURES
Figure 1: Main players of the online advertising value chain 22
Figure 2: Ad delivery process: impression, click and conversion 24
Figure 3: Global ad spending forecast in million USD in 2020 28
Figure 4: Global CPM in USD – Metrics on Facebook & Instagram 29
LIST OF TABLES
Table 1: Elements of an ad brokerage network (Ad-network) 23
Table 2: Cost-benefit assessment 49
Table 3: Nature and scope of the legal measures 56
Table 4: Summary table on gaps 73
Table 5: Summary table on enforcement 75
Table 6: Main piece of legislation in case study countries 79
Table 7: Breakdown of online advertising spending in 2020 (USD) 80
Table 8: Effect of COVID-19 on online advertising in France and Australia 82
Table 9: Summary of conclusions 96
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
9 PE 662.913
LIST OF ABBREVIATIONS
Ad-Avoidance Technologies
Audiovisual Council
Australian Competition and Consumer Commission
Australian Consumer Law
Australian Communications and Media Authority
Alternative Dispute Resolution
Artificial Intelligence
Application Programming Interface
Autorité de régulation professionnelle de la publicité (French regulatory authority
for advertising)
Advertising Standards Authority
Advertising Standards Agency for Ireland
Audiovisual Media Services Directive
Business-to-Business
Business-to-Consumer
BBB National Programs
Committee of Advertising Practice
Competition and Consumer Protection Commission
Court of Justice of the European Union
Competition and Markets Authority
Commission Nationale de l'Informatique et des Libertés (French Data Protection
Authority)
Consumer Protection Act
Cost-Per-Action
Cost-Per-Click
Cost-Per-Impression
Cost-Per-Mille
Core Platform Services
Consumer Rights Directive
Conseil Supérieur de l’Audiovisuel (French Broadcasting Authority)
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 10
Department of Infrastructure, Transport, Regional Development and
Communications
Digital Markets Act
Data Protection Act
Data Protection Impact Assessment
Digital Services Act
Digital Service Coordinator
Demand Side Platform
European Advertising Standards Alliance
European Commission
European Data Protection Board
European Parliament
European Union
Frequently Asked Questions
Federated Learning of Cohorts
Federal Trade Commission
German Advertising Association
British pound sterling
General Data Protection Regulation
General Media Panel
Global Positioning System
Interactive Advertising Bureau
International Chamber of Commerce
International Consumer Protection Enforcement Network
Identifier
Internet Protocol
Incorporated Society of British Advertisers
Internet Service Provider
Member State
National Competition Authority
National Commission for Informatics and Freedoms
National Trading Standards
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
11 PE 662.913
Office of the Australian Information Commissioner
Online Behavioural Advertising
Online Dispute Resolution
Organisation for Economic Co-operation and Development
Platform to Business
Personally Identifiable Information
Quasi-Identifiers
Return on Investment
Real-Time Bidding
Small and medium-sized enterprises
Short Message Service
Syndicat des Régies Internet (Syndicate which aims to ensure the professionalisation
and development of digital advertising in France)
Supply Side Platform
Television
Unfair Commercial Practices Directive
United Kingdom
United States
United States Dollar
Gesetz gegen den unlauteren Wettbewerb (German Law against unfair competition)
Very Large Online Platform
Video-sharing Platform
World Federation of Advertisers
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 12
EXECUTIVE SUMMARY
Background
Online advertising is expanding at a significant rate
1
, and the sector is expected to take on increased
importance as progress towards digitisation accelerates in the wake of the COVID-19 pandemic. In this
context, consumers and SMEs as “end-users” of advertising are likely to become increasingly exposed
to online advertising which is targeted on the basis of their behavioural patterns, raising important
questions about privacy, and the potential for misleading or exploitative marketing or discrimination.
At the same time, concerns are emerging about the challenges faced by smaller companies which may
be seeking to provide or rely on these forms of advertising, but lack full information or the required
bargaining power to ensure their products are fairly represented.
Aim
The European Commission has recently released two legislative proposals, the Digital Services Act
2
and
the Digital Markets Act
3
, which include provisions which are relevant to online advertising. In addition,
there are other legislative measures targeted towards digital services
4
, as well as horizontal rules
5
regarding consumer protection, privacy and advertising which apply to the sector.
This study aims to inform the IMCO Committee about emerging challenges resulting from online
advertising practices, as well as identifying potential areas where legislative proposals could be
improved, or new initiatives taken.
How do online advertising markets work?
There are three main types of online advertising: search advertising, display advertising and classified
advertising. Online advertising has created a unique opportunity to tailor advertisements to reflect the
interests or needs of consumers. Targeting can be done based on the content of the visited website or
search query. However, targeting can also be based on information gathered about the consumer e.g.
via cookies or other tracking technologies. This type of targeted advertising - called behavioural
advertising - can involve extensive processing of consumers’ data. The provision of advertising can
involve a number of different players, including not only the advertiser and the consumer, but also the
“publisher” (the party which provides advertising space) and potentially an intermediary which acts to
provide a matching service between the advertiser and the publisher, often with the aid of data
analysis.
Challenges and opportunities concerning digital advertising
Online targeted advertising can generate benefits for both consumers and SMEs, specifically by
tailoring advertisements to match consumers’ interests and even by enabling the protection of certain
1
In 2019, Europe accounted for 19.9% of the global market for online advertising. Recent trends include a shift from desktop to mobile
content and associated advertising. Source: Statista, 2020, Digital Advertising Report.
2
European Commission, 2020, Proposal of the Commission of 15 December 2020 for a Regulation of the European Parliament and of the
Council on a Single Market For Digital Services (Digital Services Act) and amending Directive 2000/31/EC, COM(2020) 825 final, 2020/0361
(COD).
3
European Commission, 2020, Proposal of the Commission of 15 December 2020 for a Regulation of the European Parliament and of the
Council on contestable and fair markets in the digital sector (Digital Markets Act), COM(2020) 842 final, 2020/0374 (COD).
4
For example, the Platform to Business Regulation, and ePrivacy Directive and its proposed successor for an ePrivacy Regulation.
5
These include the UCPD, Directive on misleading and comparative advertising, CRD, GDPR and Better Enforcement Directive.
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
13 PE 662.913
consumers e.g. by screening out advertisements which may be unsuitable for minors. However,
targeted advertising can also raise a number of concerns. Issues impacting consumers include:
• A lack of understanding by consumers that their data are being used to target advertising;
• Targeting which exploits the vulnerability of certain groups of consumers;
• Consent forms and other design features which seek to steer consumers to make decisions
against their interests (dark patterns);
• Challenges to seek redress in cases where advertising or data collection / consent methods are
inappropriate due to the range of bodies involved; and
• The potential for the algorithms used to target advertising to result in intentional or
unintentional harmful discrimination.
Meanwhile the significant reach of the largest platforms, their access to extensive datasets (which
enable targeting) and participation at multiple levels of the advertising value chain (which can be
associated with bundling or self-preferencing practices) can create challenges for SMEs seeking to
advertise or develop a competing advertising platform.
How far does existing EU legislation go in addressing the problem?
Digital advertising is covered by a wide range of existing EU legislation including legislation specific to
digital platforms and services such as the eCommerce Directive, P2B Regulation, ePrivacy Directive and
Audiovisual Media Services Directive as well as horizontal measures such as the Unfair Commercial
Practices Directive, Directive on misleading and comparative advertising, Consumer Rights Directive
and General Data Protection Regulation. Moreover, competition law applies to all digital platforms.
However, the diversity of measures can make the legal framework complex to understand and apply.
Moreover, the legal framework is also fragmented across the EU27, especially since the older Directives
such as the eCommerce Directive are minimal harmonisation instruments, and some Member States
have introduced more far-reaching rules. Important gaps remain. For example, there are no obligations
to ensure that consumers are informed that they are being subject to targeted advertising, and to allow
them to change the parameters under which they are being targeted or to opt out. In addition, there
are no general rules which prevent minors from being subject to harmful targeted advertising.
Moreover, the complexity of the rules and diversity of players along the value chain risks creating a
vacuum in which the different players may not be aware of their responsibility to meet certain
obligations.
Some of these issues may be addressed within the proposed Digital Services Act (DSA) and Digital
Markets Act (DMA) as well as the proposed Artificial Intelligence Act (AIA). However, gaps remain,
notably regarding the treatment of “dark patterns”, algorithmic discrimination, how to ensure that the
relevant players take responsibility for their content and targeting methods, and how to enable
consumers to seek redress in an increasingly complex system. The DSA, the DMA and the AIA do include
provisions which aim to increase transparency towards consumers and purchasers of advertising.
However, these could be further strengthened and extended.
Conclusions and recommendations
Based on an analysis of the problems and their causes, as well as legislative gaps, we have identified
the following potential solutions that may be relevant to protect consumers and SMEs and facilitate
the development of the single market for advertising in the context of the DSA and DMA as well as
other legislative and soft law instruments.
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 14
a. Informing consumers about being targeted and improved consent mechanisms
The DSA could contain a requirement for meaningful transparency concerning the existence of
targeted advertising, alongside a requirement that information about targeted advertising should be
conveyed in a manner which is clear to consumers. If these measures are insufficient, an opt-in to
targeted advertising could be encouraged through the self and co-regulatory measures provided for
in Article 36 of the DSA.
b. Addressing “dark patterns” through guidelines
EDPB guidelines cover the issue of “dark patterns” to some extent. However, further action could be
taken such as defining design guidelines, among others, for cookie banners and consent forms and
providing a user-friendly tool enabling consumers to report websites that may not comply
6
.
c. Preventing discrimination and improving algorithmic transparency
The DSA could further contribute to ensuring that rules regarding discrimination are adhered to in the
context of digital advertising by enforcing greater meaningful transparency concerning the existence
of targeted advertising and the parameters used. In addition, regular vetting of systems and training
data by accredited researchers could offer new insights on how to mitigate systemic risks and reduce
information asymmetries.
d. Ensure that minors are not subject to targeted advertising which exploits their vulnerabilities
The AVMS Directive already includes obligations to protect minors from harmful content in
advertising
7
. However, the provisions apply only to video sharing platforms. The DSA could include a
similar provision to that in Article 28b, paragraph 3 AVMS, to clarify that minors (and potentially other
vulnerable customer groups) should be protected from harmful targeted advertising.
e. Ensuring responsibility for targeted advertising when multiple actors are involved
Actors which are responsible for targeted advertising solutions may not fall under the scope of the DSA,
which is limited to intermediation services. A possible solution may be to clarify that Article 5.3 of the
DSA proposal, which removes the liability exemption of the hosting platforms, could also apply to
platforms which may lead a reasonably well-informed consumer to believe that advertising is provided
by the online platform itself or by a recipient of the service who is acting under its authority or control.
This would ensure that the platform has an incentive to comply with all the transparency rules.
f. Improving consumers’ access to redress
The proposed DSA sets out a generic mechanism for users to flag illegal content and to seek redress.
However, this may not help consumers to identify how to make complaints and ensure that they reach
the right enforcement body. The Digital Service Co-ordinators could be tasked with providing
information to consumers on how to seek redress in relation to online advertising (among other areas).
Another option would be to adopt a sector-specific directive addressing all consumer protection issues
related to online advertising.
6
In the US, users can report websites to their respective Attorney General if they feel they are subjected to a dark pattern. Available at:
https://darkpatternstipline.org/report/.
7
Article 9(1) of AVMS Directive.
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
15 PE 662.913
g. Facilitating the functioning of the internal market
The risk of legal uncertainty deriving from the potential application of national rules to advertising
services could be mitigated by clarifying the meaning of common terms and enforcing cooperation
mechanisms between Member States. This could be done, for instance, by amending the eCommerce
Directive (possibly through the DSA) by introducing precise deadlines and procedural conditions for
the implementation of derogations by Member States
8
. Adopting EU-wide Codes of Conduct which
could be ‘vetted’ by the European Commission to define which types of national restrictions would be
compatible with the internal market clause might also be helpful.
h. Addressing exploitation by platforms which hold a gatekeeper position in digital advertising
Smaller firms in particular may be reliant on large gatekeepers to reach consumers, potentially allowing
exploitation. The proposed DMA includes provisions which aim to shine a light on potential
exploitation by requiring information on advertising prices and performance to be shared with
advertisers. Alongside approving the DMA provisions, potential exploitation could be addressed by
taking advantage of this information to pursue case by case enforcement under competition law.
i. Increase transparency concerning advertising auctions and the performance of advertising
The distribution of ads via ad-auctions is marked by a lack of transparency towards both advertisers
and publishers. The current DMA proposal provides for some useful transparency provisions, but could
be extended to require transparency for the criteria used by the ad-tech platform services in the auction
process, including details of the price components as well as other factors which are taken into account
in the auction process and their weighting.
j. Tackling bundling and tying by gatekeeper intermediaries of premium advertising space
Ad-inventory of vertically integrated large intermediaries (e.g., Facebook, YouTube, Google) is
considered very valuable from an advertiser’s perspective but is often exclusively marketed via their
own Ad-Network or Ad-Exchange. This could raise entry barriers and impede competition in the
provision of advertising intermediary services. It may be appropriate to encourage the European
Commission to closely monitor competition across the online advertising value chain, and if necessary
consider separating intermediary services from the ad-inventory of their publisher’s sites.
k. Addressing asymmetric access to consumer data
Large providers of advertising services like Google and Facebook can access a huge amount of data
that other companies do not have access to. Large providers may directly prohibit or introduce
considerable obstacles to the use of the data via a competitor’s advertising services. The prohibition
on bundling and self-preferencing in the DMA proposal may address the issue to some degree, but
further analysis could be conducted to understand whether other measures may be necessary.
8
Such recommendations are inspired by the mechanisms laid down in Article 3 of the AVMS Directive (as amended as amended by
European Parliament and the Council of the European Union, 2018, Directive (EU) 2018/1808 of the European Parliament and of the
Council of 14 November 2018 in view of changing market realities, OJ L 303/69 of 28.11.2018
).
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 16
HOW DO ONLINE ADVERTISING MARKETS WORK?
Online advertising involves the interaction of a number of different players. In addition to the firms that
wish to advertise their products and services and the consumers and businesses at the receiving end
of the process, the market also involves a number of intermediaries, such as website operators
(publishers) and advertising networks.
In this chapter, we provide an overview of the different types of digital advertising, and the main
players in the market, as well as on the role played by “targeting” in digital advertising.
1.1. Types of digital advertising
There are three main types of digital advertising: search advertising, display advertising and
classified advertising
9
.
Advertising firms engaging in search advertising pay search engine operators to link their website (on
which they sell products or services) in a user’s search results. When a user enters a search query, the
specific advert (or sponsored link) is then shown together with the organic search results. The
sponsored link is typically presented in short form so as to resemble a search result, however they are
usually highlighted as advertisements by a short text identifier or additional colour highlights
10
. The
operator of the search engine usually determines which sponsored link is shown to an individual user,
and advertising firms compete to receive priority for display based on specific keywords and user
segments. Typically, with this advertising method, the dependence on user data is relatively low
because the selection of advertisements for presentation is mainly determined by the specific search
query. However, in some cases user data can be used to support the matching process
11
.
Search advertising is one of the most successful forms of online advertising in terms of conversion rate,
i.e. the percentage of visitors to a website that complete the desired goal of clicking on an ad (a
conversion) out of the total number of visitors who sees a specific ad. For example, the average
conversion rate of a sponsored search result via Google Ads across all industries is 4.40%, whereas it is
only 0.57% for display advertising
12
. The reason for this is that search advertising reaches consumers
while they are already willing to make an active purchasing decision, since they have already expressed
interest when searching for specific (product related) keywords
13
. Potentially as a result of its
effectiveness, search engine advertising is the second largest online advertising segment worldwide in
terms of revenue with a share of 43.3% in 2019. In the US and Europe, this segment is heavily dominated
by Google and its service Google Ads. The ongoing development of search algorithms that will
improve the targeting quality of search advertising as well as the emergence of more specialised
product related search engines will drive future growth of this market segment. A steady annual
growth of 6.7% in this segment is expected over the next years
14
.
9
See CMA, 2020, Online platforms and digital advertising, Market study final report.
10
Autorité de la Concurrence, 2010, No 10-A-29 Opinion, 28:‘display is reserved for branding objectives and search-based ads for
performance objectives.
11
See CMA, 2020, Online platforms and digital advertising, Market study final report.
12
See Wordstream, 2020, Conversion Rate Benchmarks: Find Out How Your Conversion Rate Compares.
13
See Geradin, D. & Katsifis, D., 2019, ‘An EU competition law analysis of online display advertising in the programmatic age’, European
Competition Journal, 15:1, 55-96.
14
See Statista, 2020, Digital Advertising Report 2020.
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
17 PE 662.913
Display advertising refers to advertisements that are displayed on websites or applications in various
forms. It includes social media advertising, video advertisements as well as banner-
advertisements. Display advertising is the largest segment in the online advertising market behind
search advertising and accounts for 51.8% of total online advertising revenue
15
.
Social media advertising accounts for the largest share of display advertising revenue. This segment
includes advertisements that are shown to users in the context of social networks (e.g., Facebook,
Instagram, TikTok, etc.) and business networks (e.g. LinkedIn, etc.), irrespective of whether the
advertisement is in the form of a text-based post, a picture or even a video. The advertised content is
embedded in an individual user’s newsfeed and the operator of the network itself mostly directly
markets the advertising space. Users’ personal profiles, interactions on the platform (e.g. likes, sharing
of content) and browsing behaviour within the social networks provide very rich consumer data to the
platform operator. This data can be used to deploy deeper targeting methods for the distribution of
advertisements (see Section 1.2). The leading company in this segment is Facebook with a market
share of between 75-80% in the US and Europe
16
. Furthermore, with the trend towards integrating
eCommerce and payment solutions into social networks, social network providers could benefit from
new monetisation opportunities in the future. Due to this development, consumer engagement and
conversion rates from advertisements on social networks are expected to increase significantly
17
.
Video advertising includes all advertisements provided within web- or app-based video-players. The
advertising content can be displayed in the form of a video (e.g. pre-roll advertisements), as well as
text- or image-based overlays. The increasing popularity of user-generated video content is a key driver
for this advertising type. In 2019, the video advertising segment accounted for 9.4% of the global online
advertising market
18
. The importance of this advertising type will further expand as developments in
mobile infrastructure and the widespread availability of 5G may provide a new boost for this kind of
advertising on mobile devices. While the field is dominated in the US and Europe by Google’s service
YouTube, other subscription-based video streaming services may adapt their current business model
and become publishers and offer video advertising space in the near future.
Banner-advertisements are typically placed on a publisher’s website or in an app alongside their
original content. The advertisement space is regularly marketed via a complex chain of intermediaries
(e.g., Ad-Networks, Ad-Exchanges) that determine the placement of a specific banner-ad based on a
real-time auction format (see Section 1.3.2)
19
. Banner-advertising accounted for 15% of global online
advertising revenues in 2019. An important innovation in the distribution of these ads was the
development and introduction of a system that allows for fully automated and individualised
purchases of advertising space, also known as programmatic advertising
20
. This process of
“programmatic advertising describes […] [a] method of buying, displaying, and optimizing of
advertising space driven by audience data in order to better target certain potential customers”
21
.
Nowadays, the majority of digital ad spending is processed via programmatic advertising
22
.
15
Ibid.
16
Ibid.
17
Ibid.
18
Ibid.
19
See CMA, 2020, Online platforms and digital advertising, Market study final report.
20
Bundeskartellamt, 2018, Online advertising. Series of papers on “Competition and Consumer Protection in the Digital Economy“.
21
Statista, 2020, Digital Advertising Report 2020, p. 25.
22
Statista, 2020, Digital Advertising Report 2020.
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 18
The third main category of online advertising is classified advertising. For this type of advertising,
advertisers directly purchase an advertising slot on a publisher’s website. This allows for the specific
targeting of the website’s user group and avoids losses that could arise from the extensive scattering
of advertising across different settings, which is typical for display advertising. Sectors in which
classifieds are frequently observed include recruitment, eCommerce, consumer finance, travel, real
estate and cars
23
. Pricing schemes for classified advertisements are relatively transparent and often
similar to offline advertisements, such as one-time payments which are independent of the
advertisement’s success. This segment is the smallest in online advertising and represents only 5.2% of
the global market revenue
24
. However, it is interesting to note that, the European market for classified
online advertising is the world's largest
25
, and thus this type of advertising plays a larger role in Europe
than it does elsewhere.
1.2. The role of targeting in digital advertising
Traditional advertising formats in TV, radio, print outlets or billboards have only limited targeting
potential and thus advertisers spend money on consumers who are not receptive to the advertised
content. In contrast, online advertising presents multiple opportunities for targeting. First, online
advertising can be linked to the activity that users are currently engaged in (e.g., searching, browsing,
watching videos, etc.) which allows for a segmentation of consumers
26
. Second, it is possible for
advertisers to measure the efficacy of their advertising expenditure by analysing data on the visitors
directed to their services by the ad, which enables more efficient spending of financial resources
27
.
Thus, targeting methods in online advertising should in principle allow more efficient allocation of
advertising resources and mitigate scattering losses from uninterested customers.
Exploiting this “value added” is the core business of many intermediary stakeholders, which analyse
available user data to offer sophisticated segmentation and targeting of consumer groups. In doing so,
they use a range of different targeting methods specific to the respective type of online advertising,
which can vary as regards their data requirements and potential efficacy
28
.
In the following sections, we elaborate on contextual targeting, behavioural targeting and
segmented targeting of online advertisements.
Contextual targeting methods are based on the content of a visited website or the specific search
query a user entered into a search engine. For example, a user browsing on a blog about ‘running’ will
likely see ads about ‘running shoes’ or ‘sports apparel’. Thus, it can be characterised as the online
equivalent of a tailored print-ad in a niche magazine. The placement of the specific advertisement or
sponsored link is therefore only determined by the content of the website and not by information
about the user itself. Hence, a record of specific user data or browsing history is not required. Elements
of contextual targeting are prevalent especially in classified advertisements and to a lesser degree in
search and various forms of display advertisement.
23
See CMA, 2020, Online platforms and digital advertising, Market study final report.
24
See Statista, 2020, Digital Advertising Report 2020.
25
Ibid.
26
See Goldfarb, A., 2014, ‘What is different about online advertising?’, Review of Industrial Organization, 44(2), 115-129.
27
Blake, T., Nosko, C. & Tadelis, S., 2015, ‘Consumer Heterogeneity and Paid Search Effectiveness: A Large‐Scale Field Experiment’,
Econometrica, 83: 155-174.
28
European Commission, 2018, Consumer market study on online market segmentation through personalised pricing/offers in the
European Union.
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
19 PE 662.913
In behavioural targeting, information that consumers have shared with the platform e.g. posts, likes,
written reviews, purchase history, search history, browsing history and technical information about the
individual user like e.g. device manufacturer, operating system, browser vendor, screen resolution and
IP-address are used to attribute a specific consumer to a consumer group and their interests and needs.
Hence, the data requirements to engage in behavioural targeting are relatively high compared to other
types of advertising. Data collection is carried out via cookies and other tracking technologies (e.g.
browser & canvas fingerprinting, unique device identifiers)
29
. Generally, the matching quality between
a consumer and an advertisement increases with the depth of data available and when a narrow
segmentation is possible. In practice this targeting method is regularly implemented in the so-called
‘re-targeting’ of consumers who reveal their interest in buying a specific product but leave a seller’s
website without a purchase. Over the course of a few days such a consumer then experiences an
increased frequency of display advertisement of the previously searched product he or she did not buy.
Similar methods are also used in search advertising.
In contrast to the previous two types of advertisements, which are based on data on user preferences
which has been gathered through the general use of the website or platform, advertisements delivered
by segmented targeting rely on information that the user has provided voluntarily. This is especially
relevant on websites which require a user to register a user profile and enter their name, age, gender
and other contact information
30
. Although this information allows for a perfect identification of the
user, it needs to be combined with other data to form a clearer picture of a user’s interests.
The targeting of online advertisements has become increasingly popular among marketers, because
they assume these ads are more effective and have a higher rate of return than standard non-targeted
advertisings. Behavioural advertising is a particularly fast-growing segment, with revenues estimated
to reach 21.4 billion EUR in Europe by 2020
31
. It is estimated that the click-through rate (the likelihood
that a user clicks on a displayed ad) for behavioural advertising is 5.3 times higher than for standard
advertising. For re-targeted consumers (for explanation see above) who have previously shown an
interest in a product, the click-through rate is estimated to be 10.8 times higher
32
. Nevertheless, it is
generally difficult to measure the causal effect of targeted advertisements since one cannot rule out
the possibility that a user would have bought a certain product even without seeing the advertisement.
However, Blake et al. (2015) have measured the causal efficacy in a controlled experimental setting for
search advertisements on eBay
33
. They find that there is a positive effect from search advertising, but
this is mainly present for consumers that are new to the website or use it infrequently. The more familiar
a user is with the search engine, the more she overlooks sponsored search results.
A study by Bleier & Eisenbeiss (2015) which investigates the efficacy of personalised advertisements in
banner advertising, observes a similar effect
34
. Advertisements in the field of retargeting that are
substantially personalised (heavily targeted to a specific user) are especially effective in promoting
purchases of consumers who are at an early information stage in their purchasing decision process.
However, the effectiveness of these personalised ads decreases significantly over time after the last
29
Briz, N., 2018, This is Your Digital Fingerprint, Mozilla Blog.
30
European Commission, 2018, Consumer market study on online market segmentation through personalised pricing/offers in the
European Union.
31
European Parliamentary Research Service, 2020, Digital Service Act: European Added Value Assessment.
32
Ibid.
33
Blake, T., Nosko, C. & Tadelis, S., 2015, ‘Consumer Heterogeneity and Paid Search Effectiveness: A Large‐Scale Field Experiment’,
Econometrica, 83: 155-174.
34
Bleier, A., & Eisenbeiss, M., 2015, ‘Personalized online advertising effectiveness: The interplay of what, when, and where’. Marketing
Science, 34(5), 669-688.
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 20
visit of an advertiser’s online store. Therefore, medium or less personalised (re-targeted) ads
outperform their heavily targeted counterparts in the long run. In addition, some studies even show
that very high degree of personalisation led to less engagement in the first place. For instance, Tucker
(2014)
35
and Aguirre et al. (2015)
36
each conducted field studies on Facebook to examine the
effectiveness of personalised ads. The authors considered different levels of personalisation. Although
both studies showed that personalised advertising generally has a greater effect on consumers and
click-through rates than non-personalised ads, this effect is less pronounced if the personalisation is
too precise. Thus, advertising messages that are not entirely generic, but address a user profile and
interest in a broader way, were most effective. On the other hand, the most personalised forms of
advertising were only effective in specific circumstances. Specifically, the studies found that consumers
respond more positively to high levels of personalisation if they have trust in the respective vendor, or
if the latter creates openness about how data is collected or/and provides consumers with greater
control over their own data
37
.
Usually, ad-networks combine behavioural advertising, which can be highly personalised, and
contextualised advertising with lower levels of personalisation in their targeting service. The
complementary nature of contextual and behavioural methods has also been recognised in academic
literature and it has been found that the combined application of both elements results in significantly
higher click-through rates
38
.
From an economic perspective, behavioural targeting affects publishers of advertising space and
advertising firms in two different ways. Chen and Stallaert (2014) characterise this as a competitive
effect and a propensity effect
39
. The competitive effect is beneficial for advertisers but is negative for
publishers. Behavioural targeting enables a more fine-grained segmentation of consumers with the
result that fewer advertising firms compete for the same target group. The reduction in fixed
advertising prices may be due to this. In contrast, the propensity effect implies that, due to behavioural
targeting, the likelihood of a user clicking on an ad is higher. This should result in a higher click-through
rate and, thus, more revenue from the success-based price components for publishers. Therefore, the
overall effect hinges crucially on the efficacy of behavioural targeting (and other targeting methods),
and on which of the two effects dominates.
35
Tucker, C. E., 2013, ‘Social Networks, Personalized Advertising, and Privacy Controls’, Journal of Marketing Research, 51 (5):546-562.
36
Aguirre, E., Mahr, D., Grewal, D., de Ruyter, K. & Wetzels, M., 2015, ‘Unraveling the Personalization Paradox: The Effect of Information
Collection and Trust-Building Strategies on Online Advertisement Effectiveness’, Journal of Retailing, 91 (1):34-49.
37
e.g. Chellappa, R. K., & Sin, R. G., 2005,’Personalization versus Privacy: An Empirical Examination of the Online Consumer’s Dilemma’.
Information Technology and Management, 6 (2):181-202; Tucker (2013); Aguirre et al (2015); Bleier, A., & Eisenbeiss, M., 2015, ‘The
Importance of Trust for Personalized Online Advertising’, Journal of Retailing, 91 (3):390-409.
38
Lu, X., Zhao, X., & Xue, L., 2016, ‘Is combining contextual and behavioral targeting strategies effective in online advertising?’, ACM
Transactions on Management Information Systems (TMIS), 7(1), 1-20.
39
Chen, J., & Stallaert, J., 2014, ‘An economic analysis of online advertising using behavioral targeting’, Mis Quarterly, 38(2), 429-A7.
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
21 PE 662.913
1.3. Main actors and business models
The online advertising environment involves different actors across the value chain. While there are
subtle differences within each role, the main roles can be identified as advertisers, publishers,
intermediators and consumers. This classification is based on Cai et al. (2020) and their assessment
on the distinction made by the Interactive Advertising Bureau (IAB)
40
. A conceptual description of the
online advertising ecosystem is provided in Figure 1. The characteristics of each of the players are
described below.
• Advertisers: Large businesses or SMEs which are interested in advertising their products or
services. They may have a preference for specific target groups, which could give them
guidance on where to place their ads. If the chosen advertising type allows for design elements
(text, images, etc.), these tend to be designed by the advertisers themselves. Advertisers pay
according to a sophisticated success-based measure which usually includes “cost-per-
impression” (CPI) and “cost-per-click” (CPC) elements.
• Publishers: Platform operators and owners of websites that offer space for advertisements to
be shown. Publishers’ dependence on advertising revenue varies significantly as there are
websites that are entirely financed by advertising, whereas others rely on advertising only as a
supplementary revenue stream.
• Intermediators: Companies which engage in the distribution and targeting of advertisements.
Market players that fall under this category can be further characterised as either Supply Side
Platforms (SSPs), Demand Side Platforms (DSPs) or Ad-Exchanges. While SSPs usually buy
advertising spaces on publishers’ websites and sell these on Ad-Exchanges (brokers between
both sides), DSPs buy these advertising inventories on Ad-Exchanges to fill them with matching
advertisements from interested advertisers they represent. However, there are also large
intermediaries who assume a dual role or combine each of the above mentioned three roles in
their services. In this case they are referred to as “Ad Brokerage Network” in Figure 1. These
intermediaries are also responsible for transferring payments from advertisers to publishers.
• Users: Consumers and business users are the recipients of advertisements. They are likely to
encounter a range of advertisements (search, display, etc.) during their regular browsing
experience. Which specific advertisement is shown to an individual user is determined by
sophisticated targeting methods.
The following figure summarises the main actors across the online advertising value chain. We
elaborate further below on the business models, with a focus on the intermediary actors who engage
in the delivery of the advertisements and discuss how they interact with the other stakeholders
involved. We also discuss the specific role played by SMEs and consumers at different points in the
value chain.
40
Cai, Y., Yee, G. O., Gu, Y. X., & Lung, C. H., 2020, ‘Threats to online advertising and countermeasures: A technical survey’, Digital Threats:
Research and Practice, 1, 2, Article 11.
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 22
Figure 1: Main players of the online advertising value chain
41
Source: Cai et al. (2020).
1.3.1. Ad-Networks
The business model of an intermediary Ad Brokerage Network is based around aggregating advertising
space from publishers and matching it against advertisers’ demand. It can offer any combination of
services that would relate to a Supply Side Platform (SSP), Demand Side Platform (DSP) or Ad-Exchange.
Table 1 provides definitions of these roles as described by the Interactive Advertising Bureau (IAB)
42
.
Although Ad-Networks in the narrow sense represent only one element in an Ad Brokerage Network’s
service umbrella, the term is often used as a synonym for the entire sector of intermediaries. For the
remainder of the study we refer to “Ad-Networks” in relation to all Ad Brokerage services and refer to
the elements of DSPs, SSPs and Ad Exchanges specifically where necessary.
Ad-Networks in their role as intermediaries interact with advertisers and publishers. Advertising firms
provide the network with the relevant campaign materials while the publishers offer advertising space
on their website to be filled by the network. When planning their advertising campaign, advertisers can
typically choose between different targeting options offered by the network. Targeting can for
example be based on geolocation, specific keywords which define the contextual environment, time
of day, browser type or operation system amongst other options. To identify these different consumer
segments, Ad-Networks rely heavily on the analysis of click-stream data from users (i.e. data on the
detailed interaction of users with elements of the web-service).
41
Cai, Y., Yee, G. O., Gu, Y. X., & Lung, C. H., 2020, ‘Threats to online advertising and countermeasures: A technical survey’, Digital Threats:
Research and Practice, 1, 2, Article 11.
42
Ibid.
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
23 PE 662.913
Table 1: Elements of an ad brokerage network (Ad-network)
43
Player Definition
Agency
An organisation that, on behalf of its clients, plans marketing and advertising
campaigns, drafts and produces advertisements, places advertisements in the
media. Agencies often use third-party technology (ad servers) and may place
advertisements with publishers, ad networks, and other industry participants.
Demand Side
Platforms
(DSPs)
Organisations that provide centralised (aggregated) media buying from multiple
sources including Ad Exchanges, Ad networks, and Supply Side Platforms, often
leveraging real-time bidding capabilities of said sources.
Ad-Exchanges
Organisations that provide a sales channel to Publishers and Ad-Networks, as well
as aggregated inventory to Advertisers. They bring a technology platform that
facilitates automated auction-based pricing and buying in real-
time. Ad
Exchanges’ business models and practices may include features that are similar to
those offered by Ad-Networks.
Supply Side
Platforms
(SSPs)
Organisations that provide outsourced media selling and Ad-Network
management services for publishers. Also known as sell-side platforms. Their
business models and practices are similar to Ad Networks. SSPs are typically
differentiated from Ad-Networks in not providing services for Advertisers. DSPs
and Ad-Networks often buy from SSPs.
Ad-Networks
Organisations that provide an outsourced sales capability for publishers and a
means to aggregate inventory and audiences from numerous sources in a single
buying opportunity for media buyers. Ad Networks may provide specific
technologies to enhance value to both Publishers and Advertisers, including
unique targeting capabilities, creative generation, and optimisation. Ad Networks’
business models and practices may include features that are similar to those
offered by Ad Exchanges. Usually, Ad-Networks rely on Content Delivery Networks
to server advertising contents.
Source: Cai et al. (2020).
Traditional payment models in advertising have been dominated by one-time (or potentially recurring)
up-front payments for ad space, like the rental expenses of billboards or a fixed cost for a printed ad in
a newspaper. While this still holds true for the classified sector within online advertising, in search and
display advertising, however, performance-based measures such as cost-per-click (CPC), cost-per-mille
(CPM – thousand impressions) and cost-per-action (CPA) (the desired action of the advertiser which is
usually a purchase) are prevalent. These success-based measures serve to reduce the risk of advertisers
of wasting resources on ad spaces which receive limited views or achieve only poor conversion. The
delivery process for the example case of display advertising is displayed in Figure 2.
43
Ibid.
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 24
Figure 2: Ad delivery process: impression, click and conversion
44
Source: Dave et al. (2013).
To improve efficiency in the intermediation process (and to maximise the profits generated by
targeting) large Ad-Networks and Ad-Exchanges tend to employ an auction format to determine which
advert is shown to a given user. After the Ad-Network/Exchange receives the JavaScript request for the
display of an ad, this real-time auction process is triggered and returns the winning advertisement. The
successful impression of the winning ad is recorded and the relevant advertiser is billed. Advertisers
who are interested in targeting a specific consumer group can choose to do so by preselecting a
keyword, industry sector or other identifiable metric.
The largest players in the Ad-Networks space are Google, Facebook and Amazon, who not only act as
SSP, DSP (thereby aggregating advertising demand and supply) but also operate an integrated Ad-
Exchange. Thus, these firms operate well beyond their self-generated ad spaces and additionally act as
market makers, e.g. via Google Ads, Google AdSense, DoubleClick, the Facebook Audience Network or
Amazon Advertising. The vertical integration of these large players contrasts with smaller market
participants, which typically act only as DSP, SSP or provide Ad-Exchange services. We identify potential
problematic practices arising from vertical integration by major players in online advertising in Section
2.
1.3.2. Ad-Exchanges and advertising auctions
In contrast to Ad-Networks which aggregate the supply and demand side, Ad-Exchanges are brokerage
platforms on which all market players interact. Large publishers may post their available inventory
directly and link it to specific keywords to allow filtering and segmentation, while large advertisers
provide their marketing material.
44
Dave, V., Guha, S., & Y. Zhang, 2013, ‘Viceroi: Catching click-spam in search ad networks’, Proceedings of the ACM SIGSAC Conference on
Computer and Communications Security. ACM, 765–776.
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
25 PE 662.913
Other Ad-Networks (DSPs & SSPs) may also use Ad-Exchanges themselves to buy relevant advertising
space or offer advertising space from the publishers they represent.
The Ad-Exchange is central to the Real-Time Bidding (RTB) process and hosts the auction in which the
demand and supply of advertising space is brokered. Following a user’s visit to a website that includes
advertisements, the request is sent to the Exchange along with relevant data that further characterises
the user, his or her past browsing history, the context of the website, the device being used, and the
location of the request among many others. Based on these characteristics, DSPs who represent
advertisers that want to target the respective consumer group enter the RTB process and submit their
maximum bid. The winning DSP will be determined and fills the inventory space with the advertising
materials of their clients
45
.
The auction format of the RTB follows the design of a second-price auction in which the advertisers’
maximum bids compete against each other. Under a second-price auction, the winner submitting the
highest bid pays a price set by the second highest bid in the auction. These auction formats motivate
bidders to increase the level of their bid as, if they win, they pay a lower amount than their bid price.
Due to these dynamics, sellers or auction hosts may secure higher revenues under a second price
auction than under a first-price auction where the winner pays the amount that they proposed in their
own bid.
In these ad auctions, advertisers tend to face higher price competition for popular search keywords and
more general characteristics, which may increase incentives to bid for space on more niche keywords
or user characteristics to avoid competition from other advertisers. The buyers’ maximum bids are
usually measured in a CPM (cost-per-mille) price which is charged for each thousand impressions of
the advertisement which is equivalent to an auction win in an identified consumer segment. This
maximum bid is complemented by specific quality scores of the advertisement material provided. This
auction format, equivalent to a stock-market, is easily scalable and Ad-Networks seek to optimise their
RTB process such that the advert with the highest success rate (in terms of subsequent conversion) is
shown to consumers
46
.
A prime example of the auction process can be seen in Google’s so-called “Ad-Manager”
47
. This service
is a fully automated auction, also referred to as a “programmatic auction”, and receives bids from
advertisers based on their targeting settings and advertisers pay less or exactly their maximum bid
48
.
Thus from a pricing perspective, the auction format is traditionally that of a second-price auction
49
,
although the design has shifted more towards first-price elements recently
50
. The winning bidder,
however, is determined not only by the price, but also by other factors which are weighted in a way to
ensure the auction maximises the profits of the auction operator. According to Varian (2009) this is
achieved through ranking the ads “…by bid times expected click through rates, and those ads with the
highest expected revenue are shown in the most prominent positions”
51
. However, users’ current
propensity to click on an ad is not the only weighting factor that is applied to a submitted bid.
45
For more details on the technical implementation of the RTB process the interested reader is referred to Wang, J., Zhang, W., & Yuan, S.,
2016, Display advertising with real-time bidding (RTB) and behavioural targeting. ArxivPreprint Arxiv:1610.03013(2016).
46
EDRi, 2019, Real Time Bidding: The auction for your attention.
47
Google, 2021, Ad Manager.
48
Ibid.
49
Edelman, B., Ostrovsky, M., & Schwarz, M., 2007, ‘Internet advertising and the generalized second-price auction: Selling billions of dollars
worth of keywords’. American economic review, 97(1), 242-259.
50
Despotakis, S., Ravi, R., & Sayedi, A., 2019, First-price auctions in online display advertising. Available at SSRN 3485410.
51
Varian, H. R., 2009,’ Online ad auctions’, American Economic Review: Papers & Proceedings, 99(2), 430-34.
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 26
Measures that indicate an ad’s quality may also reflect future adaptions in users’ behaviour based on
their anticipated positive or negative experiences when clicking on an ad
52
. Minimum bids or
reservation prices can also be implemented in the auction process that are either specific to keywords
or individual advertisers to increase the realised revenue
53
. All these different elements contribute to a
lack of transparency within the automated ad distribution process and give rise to problems which may
disadvantage other market players (see Section 2).
Smaller or more unknown advertisers whose adverts are expected to generate only limited click
through potential may be at a disadvantage in this process
54
. Due to the design of the auction and
weighting of gross bids, those advertisers could be forced to submit higher maximum bids than they
would otherwise submit in order to compete for the best advertisement space. This two-class society
of advertisers is also reflected in ad-auctions operated by Google. Advertisers have the option to escape
competition in the open auction format and enter a so-called “Preferred Deal” or private auction with
a more restricted pool of competitors
55
.
1.3.3. The role of SMEs
SMEs can participate in all stages of the advertising value chain. However, they most often take on the
role of advertisers in their goal to win more customers. Online advertising benefits SMEs in particular
as a quick, flexible and relatively effective method to reach a large audience at low (up front) cost,
compared to traditional offline media advertising. SMEs can also become publishers themselves in that
they can generate and sell their advertising inventory. Both in the role of advertiser and publisher, SMEs
are directly affected by their dependence on intermediaries and the advertising targeting mechanisms
used.
The market for designated Ad-Networks and Ad-Exchanges is dominated by large, mostly US based,
firms and conglomerates. Nonetheless, some smaller EU based firms are also active in the broader
intermediation market. Prime examples include the British Ad-Network adnow
56
or the French Ad-
Network Criteo
57
.
Lastly, SMEs can also represent a crucial consumer group that other SMEs try to sell to in the context of
B2B transactions. As such SMEs can also be “consumers” of advertising.
Along the online advertising value chain, SMEs have also developed complementary consulting
services to support different stakeholders. Such business models can involve improving the design of
adverts, the strategic choice of keywords and advising publishers on how to effectively monetise their
advertisement inventory.
1.3.4. Consumers
Consumers are typically on the receiving end of the online advertising value chain. They encounter
advertisements in various forms when they use a search engine or visit a website. Depending on the
content and the context in which the ad is experienced, the user may view it as potentially valuable,
52
Ibid.
53
Ostrovsky, M., & Schwarz, M., 2011, ‘Reserve prices in internet advertising auctions: A field experiment’, Proceedings of the 12th ACM
conference on Electronic commerce (pp. 59-60).
54
Liu, D., Chen, J., & Whinston, A. B., 2010, ‘Ex ante information and the design of keyword auctions’, Information Systems Research, 21(1),
133-153.
55
Google, 2021, Ad Manager.
56
AdNow, 2021, Website.
57
Criteo, 2021, Website.
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
27 PE 662.913
providing useful information, which is economically characterised as informative advertising in the
sense of Grossman & Shapiro (1984)
58
. In this case, advertisements can help consumers to make
purchasing decisions, which better reflect their preferences and are therefore welfare enhancing
59
.
However, if adverts are perceived as rather intrusive instead, consumers experience nuisance costs that
limit their browsing experience
60
. This effect led to an increase in the use of ad avoidance technologies
in recent years.
In addition to their role as recipients of advertisements, consumers which participate in deploying user-
generated content, may also enter the advertising value chain themselves as publishers selling their
inventory from blogs or online tools. Unless they reach a very large audience group, consumers as
publishers tend individually to have very limited bargaining power in relation to other actors in the
value chain
1.4. Evolution of the market and implications of COVID-19
The first ever online advertisement was placed in 1994 by AT&T; it was a banner ad displayed on the
website of HotWired
61
. Since then, the relevance of online advertising has continued to grow, as has
the amount spent by companies on digital advertising. During the last decade, the sector has grown at
a compound annual growth rate of about 18%
62
. According to a study conducted in 2017, digital
advertising contributes EUR 526 billion to the EU economy every year and supports about 6 million
jobs
63
.
As for all branches of the economy, the outbreak of the pandemic has also left its mark on the online
advertising industry. Global digital ad spending in 2020 was lower than previously expected. Prior to
the outbreak of the pandemic, the global online advertising spend forecasted by Statista for 2020 was
in the range of USD 374 billion
64
, but it only amounted to about USD 356 billion
65
, a difference of
approximately 5%
66
.
58
Grossman, G. M., & Shapiro, C., 1984, ‘Informative advertising with differentiated products’, The Review of Economic Studies, 51(1), 63-81.
59
Meurer, M., & Stahl II, D. O., 1994,’ Informative advertising and product match’, International Journal of Industrial Organization, 12(1), 1-19.
60
Johnson, J. P., 2013,’ Targeted advertising and advertising avoidance’. The RAND Journal of Economics, 44(1), 128-144.
61
Kaye, B. K., & Medoff, N., 2001, Just a Click Away: Advertising on the Internet. Massachusetts: Allyn and Bacon as cited in Evans, D. S., 2009,
‘The Online Advertising Industry: Economics, Evolution, and Privacy’, Journal of Economic Perspectives 23:3, 37–60. Bundeskartellamt, 2018,
Online advertising. Series of papers on “Competition and Consumer Protection in the Digital Economy”.
62
Own calculation based on data provided by Zenith, 2020, Internet advertising spending worldwide from 2007 to 2022, by format, Statista.
63
IAB Europe, 2017, Research report: The economic contribution of online advertising in Europe.
64
This corresponds roughly to EUR 295 billion [exchange rate 31.12.2020].
65
This corresponds roughly to EUR 292 billion [exchange rate 31.12.2020].
66
Statista, 2020, Digital Advertising Report 2020.
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 28
Figure 3: Global ad spending forecast in million USD in 2020
Source: Statista, 2020, Digital Advertising Report 2020.
This underspend may be due to the fact that companies have faced uncertainties in their operating
businesses and, therefore, scaled back on ad spending. Especially in the first month of the pandemic,
both publishers and intermediaries experienced a decline in the value of their inventory and service
67
.
The pandemic had particularly negative consequences for news publishers, who regularly reported on
the pandemic. Since advertisers generally avoid displaying their ad alongside information about
natural disasters or tragedies, many advertisers refrain from placing ads on news publishers’ websites
68
.
However, the pandemic had a far more dramatic effect on traditional advertising segments than on the
digital advertising industry in 2020
69
. According to an article posted by the World Economic Forum
(2020), “out-of-home and cinema advertising shrank almost instantly”, as did spending on TV and print
advertising. On the other hand, spending on online advertising still grew in 2020, albeit less than
expected and lower than in previous years. Compared to 2019, global online advertising spending
increased by around 7% and is expected to rise again by 12% in 2021, while in Europe online ad
spending increased by about 4% in 2020 compared to the previous year. In 2021, expenditure is
expected to increase again by 13%
70
. This development is also reflected in the rising costs per 1,000
impressions. Figure 4 shows the development of the global CPM in USD for ads on Facebook and
Instagram. The dip in prices from February to April of 2020 is clearly visible. With many pulling out of
the advertising business, the availability of programmatic inventory skyrocketed during this time.
Those companies that were in a position to keep investing in advertising faced less competition in the
auctions, which resulted in greater efficiency and a decrease in CPMs
71
. Nevertheless, Figure 4 also
illustrates that prices rose again from around April onwards, and by the end of 2020 had already
surpassed the price level recorded in the previous year.
67
Widawska, I., 2021, One Year With COVID-19: Programmatic’s Performance Prior To And During The Course Of The Pandemic.
68
IAB Europe, 2020, IAB Europe Backs ‘Don’t Block Quality European Journalism’.
69
Statista, 2020, Digital Advertising Report 2020.
70
Statista, 2021, Digital Markets, Digitale Werbung, Weltweit; Statista, 2021, Digital Markets, Digitale Werbung, Europa.
71
Comstock, M., 2020, COVID-19 and Programmatic Advertising, PMG.
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
29 PE 662.913
Figure 4: Global CPM in USD – Metrics on Facebook & Instagram
Source: GuptaMedia, 2021, Understanding the effects of COVID-19 on the Facebook Ads Marketplace.
Although many companies worked and are still working with tighter budgets, completely pulling away
from advertising would do more harm than good, as it could lead to a significant drop in total brand
communication awareness and, thus, even fewer sales
72
. Therefore, the goal of advertisers is to improve
the allocation of available resources, leading them to adapt to changing consumer behaviour as a result
of the COVID-19 crisis. The pandemic has forced citizens to spend most of their time at home
73
.
Consequently, digital consumption has increased significantly. According to an article written by
Fernandez et al. (2020), digital adoption in Europe jumped from 81% to 94%
74
. Social media and
messaging services saw the most growth. ECommerce has also gained in importance
75
.
These
developments are reflected in the current as well as the future marketing strategies of various
companies which prioritise digital advertising. In a survey of marketing executives from various
industries worldwide by Criteo (2021), a significant share of marketers confirmed that they plan to
spend the same or even more in digital advertising in 2021, despite the impact of the pandemic
76
.
More
than 40% of the respondents indicated that their budgets for social media advertising, website and
content marketing as well as advertising on retail websites and apps is likely to increase in 2021. Video
ads will also account for an increasing share of marketing budgets. However, marketers also
emphasised that they face some challenges related to digital advertising campaigns. For instance,
marketers complain that “campaigns do not always target the right people”, “campaigns are too
dependent on Facebook / Google / Amazon”, and “it is difficult to measure the Return on Investment
(ROI) of campaigns”. There are also many marketers who plan to diversify their budgets and cover
multiple channels
77
. While larger companies and brands have been taking a multi-channel approach
for years, smaller enterprises rely more on or tend to use channels that reach a large audience and,
72
Roberts, K., & Southgate, D., 2020, Impact Of COVID-19 On Consumer Behaviour, Attitudes and Expectations, Kantar, iabEurope.
73
Ford, G. S., 2020, ‘COVID-19 and Broadband Speeds: A Multi-Country Analysis’, Phoenix Centre Policy Bulletin 49.
74
Fernandez, S., Jenkins, P., & Vieira, B., 2020, Europe’s digital migration during COVID-19: Getting past the broad trends and averages.
75
Roberts, K., & Southgate, D., 2020, Impact Of COVID-19 On Consumer Behaviour, Attitudes and Expectations, Kantar, iabEurope.
76
Most insights presented in this paragraph are based on Criteo, 2021, State of Digital Advertising 2021: i budget e le priorità del marketing in
un mondo nuovo.
77
Criteo, 2021, State of Digital Advertising 2021: i budget e le priorità del marketing in un mondo nuovo.
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 30
thereby, enable engagement on a large scale
78
. These companies may become more dependent than
ever on large players like Facebook, Google, and Amazon in order to reach their audience.
Publishers are also looking for new ways to improve and adapt their business models. A study found
that around 71% of publishers are seeking to diversify their revenue in the future
79
. As a result, they will
become less reliant on advertising revenue. Brazzoni (2020) features three potential strategies for
publishers. One approach could be to switch to a subscription model, as this offers an additional source
of revenue as well as extensive audience data. Publishers could also consider investing in eCommerce.
This could include selling products and services online through, for instance, newsletters and ads.
Virtual events could also prove successful
80
.
78
CMA, 2020, Online platforms and digital advertising: Market study final report; Li, C. & Hall, S., 2020, This is how COVID-19 is affecting the
advertising industry, World Economic Forum.
79
Brazzoni, A., 2020, 71% of Publishers Seeking to Diversify Revenue After COVID: Here Are the Models They're Considering. lineup.
80
Ibid.
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
31 PE 662.913
CHALLENGES AND OPPORTUNITIES CONCERNING DIGITAL
ADVERTISING
2.1. Practices and their impacts
In this chapter, we discuss how different stakeholders including consumers, SMEs and suppliers of
advertising services are affected by trends and practices in the online advertising sector. In this context,
we identify, with reference to relevant literature and recent competition and consumer protection
cases, current issues and practices that are prevalent in the online advertising landscape, distinguishing
between practices affecting consumers, purchasers and suppliers of online advertising.
2.1.1. Practices affecting consumers and SMEs as consumers/viewers of advertisement
Online Behavioural Advertising enables the delivery of messages tailored to the interests of individual
consumers. As such, it typically provides more useful and relevant information, and prevents
consumers from being presented with random ads
81
. This is considered to be one of the main benefits
of behavioural advertising for consumers
82
. However, the application of behavioural advertisement
typically requires the extensive use of user data, which raises significant privacy concerns. Research
shows that consumers’ awareness and understanding of how personal behaviour is tracked online is
limited. A study for the European Commission which examined consumers’ awareness and perception
of online personalisation practices showed that around two-third of surveyed participants reported
that they have some understanding of how personalisation practices work. However, the results
gathered by a complementary behavioural experiment showed that the proportion of respondents
who correctly identified targeted adverts or personalised rankings was considerably lower
83
. Another
study by Dehling et al. (2019) used semi-structured interviews to examine consumer awareness and
perceptions of online behavioural advertising
84
. They document large differences in knowledge about
online advertising among consumers. Whereas some interviewees called online behavioural
advertising a magic trick of the Internet, others were able to identify an existing connection between
different websites or to differentiate between different targeting techniques. While there may be some
awareness of personalisation practices, consumers are largely unaware of the volume and granularity
of the data that is being collected and used to deliver personalised advertisements. In this regard,
Which? (2019) found that consumers in general perceive data collection as a series of single bounded
transactions, where individual pieces of data are “given” to an organisation in exchange for a specific
service
85
. Furthermore, new techniques and technologies are constantly being applied to track
behaviour that ordinary consumers are not aware of, making tracking more difficult to block
86
. As such,
Kingaby (2020) notes that consumers have little agency within the current online ecosystem and have
limited ways to stop or control data exploitation
87
. When consumers are asked directly how they assess
81
Ham, C., 2017, ‘Exploring how consumers cope with online behavioral advertising’, International Journal of Advertising, 36 (4):632-658.
82
Strycharz, J.; van Noort, G.; Smit, E. & Helberger, N., 2019, Consumer View on Personalized Advertising: Overview of Self-Reported Benefits and
Concerns, In: Advances in Advertising Research X: Multiple Touchpoints in Brand Communication, edited by E. Bigne & S. Rosengren, 53-
66. Wiesbaden: Springer Fachmedien Wiesbaden.
83
Ipsos, London Economics & Deloitte, 2018, Consumer market study on online market segmentation through personalised pricing/offers in the
European Union – Request for Specific Services 2016 85 02 for the implementation of Framework Contract EAHC/2013/CP/04. Study prepared
for the European Commission.
84
Dehling, T., Zhang, Y. & Sunyaev, A., 2019, ‘Consumer Perceptions of Online Behavioral Advertising’, Proceedings of the 21st IEEE Conference
on Business Informatics.
85
Which?, 2019, Control, Alt or Delete? The future of consumer data.
86
Ham, C., 2017, ‘Exploring how consumers cope with online behavioral advertising’, International Journal of Advertising, 36 (4):632-658.
87
Kingaby, H., 2020, AI and Advertising: A Consumer Perspective.
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 32
behavioural advertising, they recognise the benefits of more relevant ads, but the concerns about
privacy and data protection seem to outweigh these advantages. In interviews, terms like manipulative,
creepy, intrusive and privacy-invasive are often provided as answers in relation to such marketing
techniques
88
. Some even wish not to be targeted at all
89
. However, a recent study by IAB Europe (2021)
revealed that most consumers, faced with a choice between a version of the Internet without targeted
advertising but with a financial fee for most content, and today's Internet with mostly free content and
targeted advertising, would choose the latter
90
.
Protecting user’s privacy often implies restricting the potential for behavioural targeting methods.
Examples include technical solutions such as those employed by alternative web-browsers based on
the Tor routing network
91
or disabling the use of cookies
92
. These possibilities indeed prevent the
tracking of users. However, they also prevent any (potentially desired) personalisation based on a
derived user profile
93
. Thus, current research has focused on investigating options for a middle ground
which offers privacy enhancement compared to the status-quo while also allowing for some kind of
imperfect user profiling on the other. One potential solution is so-called private ad ecosystems. These
methods employ locally derived user profiles in that they download a wide range of advertisements
from an ad-network and store them on an individual user’s device
94
. These pre-loaded advertisements
are not based on any available user data but rely on random samples drawn from the available
advertisements distributed by an ad-network. If a user visits a website that includes advertisements,
the targeting method then chooses the most suitable candidate among the pre-loaded ads. In this case,
the final display of advertising is only based on the locally derived user profile (and pre-loaded ads) on
the end-user device and does not reflect all the available data or the entire user profile. However, an
advantage of conducting targeting entirely on the user’s device is that it does not require personal data
to leave the closed system.
Another privacy enhancing remedy that limits behavioural targeting to some degree but does not
prevent the formation of a user profile in general, is the anonymisation of quasi-identifying variables
(QIDs). These data variables include information on a user’s gender, zip-code or IP-address which can
be later used to uniquely identify individuals. If these variables are not gathered at all or are at least
aggregated to a higher hierarchical level (e.g., GPS location into a postal code), behavioural targeting
based on interests or shopping patterns will remain feasible while an individual user’s privacy is
simultaneously preserved
95
.
88
Moore, R., Moore, M., Shanahan, K, Horky, A. & B. Mack, 2015, ‘Creepy Marketing: Three Dimensions of Perceived Excessive Online Privacy
Violation’, Marketing Management Journal, 25:42-53.; Strycharz, J., van Noort, G., Smit, E. & Helberger, N., 2019, Consumer View on
Personalized Advertising: Overview of Self-Reported Benefits and Concerns, In: Advances in Advertising Research X: Multiple Touchpoints in
Brand Communication, edited by E. Bigne & S. Rosengren, 53-66. Wiesbaden: Springer Fachmedien Wiesbaden.
89
Turow, J., King, J., Hoofnagle, C., Bleakley, A., & M. Hennessy, 2009, ‘Americans Reject Tailored Advertising and Three Activities That Enable
It’, SSRN Electronic Journal; Global Witness, 2021, Do people really want personalised ads online?.
90
IAB Europe, 2021, What would an Internet without targeted ads look like?
91
Dingledine, R., Mathewson, N., & Syverson, P., 2004, ‘Tor: The Second-Generation Onion Router’, Proceedings of the13th USENIX Security
Symposium.
92
Aggarwal, G., Bursztein, E., Jackson, C., & Boneh, D., 2010, ‘An Analysis of Private Browsing Modes in Modern Browsers’, Proceedings of 19th
USENIX Security Symposium (p. 79-93).
93
Ullah, I., Boreli, R., & Kanhere, S. S., 2020, Privacy in targeted advertising: A survey, arXiv preprint arXiv:2009.06861.
94
Toubiana, V., Narayanan, A., Boneh, D., Nissenbaum, H., & Barocas, S., 2010, ‘Adnostic: Privacy preserving targeted advertising’, Proceedings
Network and Distributed System Symposium; Guha, S., Cheng, B., & Francis, P, 2011, ‘Privad: Practical privacy in online advertising’,
Proceedings of the 8th USENIX Conference on Networked Systems Design and Implementation (pp. 169-182).
95
Ullah, I., Boreli, R., & Kanhere, S. S., 2020, Privacy in targeted advertising: A survey, arXiv preprint arXiv:2009.06861.
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
33 PE 662.913
Box 1: Behavioural targeting and privacy concerns
• Behavioural advertising provides more useful and relevant information for consumers and more
locally targeted information, thereby improving user experience.
• The higher relevance of behaviourally targeted ads is realised by gathering and analysing
personal user data. This data is regularly shared across various players, often without the users’
knowledge resulting in significant privacy concerns.
• The extent to which visible ads are delivered based on behavioural targeting methods is widely
underestimated and cannot be easily assessed by consumers.
• Technical remedies that protect privacy while providing sufficient potential for behavioural
targeting methods are available (e.g., private ad ecosystems, anonymisation of QIDs)
Perceived benefits of more relevant ads due to behavioural advertising:
Perceived costs due to consumers’ privacy infringement:
Source: Authors’ own elaboration.
In addition to not understanding how and who collects data about them, consumers may also not
recognise when they are being targeted by advertising. For instance, the use of paid personalised
rankings could be viewed as a form of nudging to encourage consumers to view and purchase certain
offers, which can be especially misleading when platforms present themselves as providing a “neutral”
comparison service. Such practices subtly divert consumer attention, thereby negatively impacting
consumer choice.
Another case where invisible targeting may limit choice relates to location-based targeting. Location-
based targeting can help companies with a more limited geographic footprint to advertise more
efficiently and effectively. However, location-based advertising also enables companies to restrict their
advertising of services or products, discounts and promotions so that they only reach a certain group
of consumers, while others would not even be aware of these offers. Depending on the extent of the
application of geo-tracking, it could be a barrier to the EU Single Market. Recent examples from a
closely related context are the geo-blocking efforts of some providers. They avoid actively advertising
certain products or services outside a predefined geographical area. Duch-Brown and Martens (2016)
examined the welfare effects of removing geo-blocking restrictions on cross-border eCommerce based
on a dataset of consumer electronics across several European countries for the period 2012-2015
96
.
Reviewing different scenarios, the authors conclude that lifting geo-blocking restrictions would be
positive and would benefit both consumers and producers. Another more recent study by Procee et al.
(2020) assessed the impact of the extension of the scope of the Geo-blocking Regulation with regard
to audio-visual and non-audio-visual services, which are currently excluded
97
. The authors considered
different industries, and found, at least in some cases, that the removal of geo-blocking could have
positive welfare effects.
96
Duch-Brown, N. & B. Martens, 2016, The Economic Impact of Removing Geo-blocking Restrictions in the EU Digital Single Market. Institute for
Prospective Technological Studies Digital Economy Working Paper 2016/02. JRC Technical Reports.
97
Procee, R., Arnold, R., Marcus, J. S., Fina, D., Lennartz, J., Kazakova, S. & E. Lykogianni, 2020, Study on the impacts of the extension of the scope
of the Geoblocking Regulation to audiovisual and non-audiovisual services giving access to copyright protected content – Final Report. Study
for the European Commission.
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 34
Furthermore, there has been a lively debate on whether IP addresses and geolocation information
should be categorised as personal data
98
. According to Article 4 of the GDPR (Regulation (EU) 2016/679)
personal data “means any information relating to an identified or identifiable natural person […]; an
identifiable natural person is one who can be identified, directly or indirectly, in particular by reference
to an identifier such as a name, an identification number, location data, an online identifier […]”. Recital
30 of the GDPR even explicitly mentions IP addresses as personal data in certain cases. These cases
occur when the IP address in combination with other information can serve to identify a person
99
.
Box 2: Obfuscation and location based targeting of advertisements reduce consumer choice
• Advertisements are regularly concealed by presenting them as seemingly neutral product
review product comparison sites. These sites effectively narrow consumers’ set of options.
• Location based targeting enables advertisers to display more relevant products or services to
consumers in their proximity. However, consumers may be excluded from offers which are
made available in other jurisdictions (a form of geo-targeting).
Perceived benefits of more relevant ads due to location targeting:
Perceived costs due to reduced consumer choice due to obfuscation and location targeting:
Source: Authors’ own elaboration.
In addition to concerns around privacy and restrictions on choice, online advertising can also raise
concerns around the exploitation of consumers’ behavioural biases. By targeting behavioural biases,
consumers can be subconsciously influenced in the decision-making process. Major biases that impact
consumers’ decisions in an online context are: inclination to stick with the default settings (status-quo
bias), tendency to focus on short-term implications of decisions by disregarding the long-term
implications (present bias or myopia). There are numerous design elements available that can serve to
nudge consumers towards behaviours that advantage advertisers, ad intermediaries or other data
collection organisations rather themselves
100
. These design practices are also called dark
patterns
101
.The objective of dark patterns is to confuse users, make it difficult to express actual
preferences or manipulate users into taking certain actions that are not always in their best interest
102
.
Different from traditional marketing efforts, which are designed to alter consumer preferences, dark
patterns attempt to manipulate consumers into doing something that is inconsistent with their true
preferences
103
. The use of dark patterns has been documented by academic research in the fields of
social networks, especially in the form of log-ins and consents to data sharing between services
104
.
98
Clifford, D., 2014, ‘EU Data Protection Law and Targeted Advertising: Consent and the Cookie Monster - Tracking the crumbs of online
user behaviour’, JIPITEC 194, para 1.
99
GDPR.EU, 2021, The EU’s GDPR only applies to personal data, which is any piece of information that relates to an identifiable person. It’s
crucial for any business with EU consumers to understand this concept for GDPR compliance; Munz, M. and Hickman, T., 2016, Court
confirms that IP addresses are personal data in some cases. White & Case Technology Newsflash.
100
CMA, 2020, Online platforms and digital advertising: Market study final report.
101
See further Brignull, H., 2021, What are dark patterns.
102
Luguri, J. and Strahilevitz, L. J., 2019, Shining a Light on Dark Patterns, University of Chicago, Public Law Working Paper. No. 719.
103
Ibid.
104
Bösch, Ch., Erb, B., Kargl, F., Kopp, H. & S. Pfattheicher, 2016, ‘Tales from the Dark Side: Privacy Dark Strategies and Privacy Dark Patterns’,
Proceedings on Privacy Enhancing Technologies. 2016 (4): 237–254.
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
35 PE 662.913
In 2018, the Norwegian Consumer Council (Forbrukerrådet) conducted a study on the prevalence of
dark pattern techniques in services of Facebook, Google and Windows 10
105
. They found that privacy
intrusive default options increase the effort for consumers to opt-out of the gathering of their user data
(e.g., cookie consent screens). Additionally, they found that providers of online advertising presented
consent to targeted advertising as exclusively beneficial while also suggesting that consumers would
experience a loss of functionality if they decline to give consent. In addition, dark patterns, if
recognised, generated backlash and negative feelings among users/consumers
106
. They are also
particularly problematic for minors, who use the same interfaces as adults but do not necessarily have
the same cognitive abilities to recognise persuasive design techniques.
Box 3: Advertisement designs can exploit consumers’ behavioural biases and include dark pattern
elements
•
Design features within advertisements may aim to exploit consumers’ behavioural biases.
• The design elements of interfaces or targeting practices can “nudge” consumers towards certain
choices which are contrary to their best interests and preferences. These so-called “dark
patterns” are especially prominent in consent forms concerning permissions relating to the
provision of personal data.
Perceived benefits of advertising design/dark patterns that exploit behavioural biases:
None
Perceived costs of advertising designs/dark patterns that exploit behavioural biases:
Source: Authors’ own elaboration.
The allocation of advertising space is dependent on the design of the Real-Time-Bidding (RTB)
processes employed by large intermediary ad-networks. These procedures take large amounts of
consumer data into account to identify and segment the consumer base into very precise groups,
which results in ads being shown to some consumers and not to others. Thus, RTB designs could result
in unintentional, yet structural discrimination. Privacy International
argues that RTB enables companies
to reach out to vulnerable consumers in ways that are difficult to audit
107
. When combined with
techniques to exploit behavioural biases, this could contribute to exacerbating bad habits or emotional
weaknesses. For instance, based on past behaviour someone might be shown fast-food ads which may
be counterproductive when trying to overcome a poor diet. Adults with earlier gambling problems
may be shown gambling ads based on their past online behaviour, making it more difficult to overcome
addiction
108
.
Moreover, data that predicts when consumers are in particular emotional states is already
in use and can be used to target consumers when they are particularly vulnerable or otherwise
receptive to certain messages. Research has also shown that social media can influence the mood of its
users. In a controversial study, Facebook tweaked the newsfeed of hundreds of thousands of users,
skewing the feed to more positive or negative content in an attempt to manipulate their mood. It was
found that this technique does have an influence on users
109
.
105
Forbrukerrådet, 2018, Deceived by Design - How tech companies use dark patterns to discourage us from exercising our rights to privacy.
106
Luguri, J. and Strahilevitz, L. J., 2019, Shining a Light on Dark Patterns, University of Chicago, Public Law Working Paper. No. 719, 2019.
107
Privacy International, 2019, Why am I really seeing that ad? The answer might be Real Time Bidding (RTB).
108
Kingaby, H., 2020, AI and Advertising: A Consumer Perspective.
109
Sullivan, G., 2014, Cornell ethics board did not pre-approve Facebook mood manipulation study, Washington Post.
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 36
However, the very same data-driven targeting methods can also be put to good use as vulnerable
groups can be identified and avoided by specific advertised content. In an interview conducted for this
study it was highlighted that these methods are already used to protect consumers, e.g., minors are
reliably identified and as a result see less advertisements for alcoholic beverages
110
. Hence, while the
amount of data that feeds into behavioural targeting methods bears risks for vulnerable consumers, it
can be also used to protect them.
Moreover, due to the large amounts of personal data that is transferred between different actors, there
is little control about how the data is shared or whether it is used against consumers in contexts other
than advertising. In this vein, EDRi argues that RTB poses challenges to the rights recognised in the
GDPR
111
. Similarly, the UK’s Information Commissioner’s Office published a report highlighting the risks
with RTB including profiling, combining and matching data from multiple sources, tracking of
behaviour and geolocation as well as invisible processing
112
. This is one of the most pertinent
challenges as regards online advertising.
Box 4: Potential for discrimination and harmful targeting of vulnerable consumers through Real-Time-
Bidding (RTB)
•
The data-driven ad distribution through RTB processes could result in structural discrimination
or harmful targeting of vulnerable consumers.
• Past browsing behaviour that indicates e.g. bad eating habits or a potential addiction to
drinking or gambling, could induce increased exposure to advertisement in these fields. This
could pose a serious risk to consumers’ wellbeing.
• The data-driven identification of consumer groups can, however, also be beneficial for
vulnerable consumers since the advertisements can be specifically filtered for them (e.g., no
advertisements for alcoholic beverages to minors).
Perceived benefits of more relevant and secure ads for vulnerable consumers:
Perceived costs of discrimination and harmful advertising for vulnerable consumers:
Source: Authors’ own elaboration.
Alongside the potential for nudging and malicious targeting, concerns may also arise around
‘malvertising’ which refers to online ads that spread malware, or illegitimately collect personal data
about the consumers’ activities online. If a consumer clicks on an advertisement that has been
distributed by a hacker or other cybercriminal, the user is usually transferred to a landing page which
automatically downloads a malicious executable or binary file
113
. The increasingly automated
distribution of ads plays into hackers’ hands since they only have to submit their malicious ads to a
regular ad-network to spread it to an entire ecosystem
114
.
110
WFA, 2020, Alcohol marketers seek to prevent minors from seeing alcohol ads.
111
EDRi, 2019, Real Time Bidding: The auction for your attention.
112
ICO, Our work on adtech.
113
Cai, Y., Yee, G. O., Gu, Y. X., & Lung, C. H., 2020, ‘Threats to online advertising and countermeasures: A technical survey’, Digital Threats:
Research and Practice, 1, 2, Article 11.
114
Zarras, A., Kapravelos, A., Stringhini, G., Holz, T., Kruegel, C., & Vigna, G., 2014, ‘The dark alleys of madison avenue: Understanding malicious
advertisements’, Proceedings of the 2014 Conference on Internet Measurement Conference (pp. 373-380).
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
37 PE 662.913
Recent malware that was widely spread through malvertising are for instance AdGholas
115
and
Ramnit
116
which infested thousands of computers of consumers and SMEs. Although ad-network and
ad-exchanges have an incentive to tackle their security issues and have done so already in the past,
attackers still have many options to bypass them
117
. Hence, end users such as consumers and SMEs
alike must keep their system’s anti-virus protection up to date to reduce their risk of exploitation.
Similar issues persist with email advertising, which is still a common marketing channel
118
. Phishing
and spam emails continue to present a challenge as roughly 50% of worldwide email traffic in Q2 2020
was spam
119
. The prevalence of spam promotes the spreading of malicious software that harms
consumers and ransomware that is targeted at SMEs. With both legitimate advertising and spam
becoming more personalised and targeted, some users struggle to correctly identify spam and
fraudulent messages
120
. Spear-phishing is a form of spam that targets carefully selected user groups in
order to gain access to important information. For this purpose, email messages are usually more
personalised
121
. Bullee et al. (2017) conducted an experiment among several hundred employees. The
authors sent out two types of phishing emails – general phishing emails and personalised spear-
phishing emails – and requested the provision of personally identifiable information (PII). While 19% of
the employees provided personal information after receiving a general phishing email, 29 % shared
information in the spear phishing condition. One can assume that less technologically savvy users, like
elderly and people with language difficulties might be more vulnerable to such practices. Research also
indicates that internet users with lower educational attainment are less likely to be able to spot more
sophisticated practices than their well-educated counterparts.
More generally, online practices that involve misleading or harmful content can be considered
problematic from the perspective that they contribute to undermining consumer trust in digital
markets. Trust is a necessary precondition for well-functioning markets. However, as harmful practices
continue to evolve, they may work to impede some of the growth potential of the digital economy. In
this vein, tacking unfair and problematic online practices has become a focal point in the new EU
Consumer Agenda
122
.
115
Constantin, L., 2016, The AdGholas malvertising campaign infected thousands of computers per day.
116
Segura, J, 2017, Canada and the U.K. hit by Ramnit Trojan in new malvertising campaign.
117
Cai, Y., Yee, G. O., Gu, Y. X., & Lung, C. H., 2020, ‘Threats to online advertising and countermeasures: A technical survey’, Digital Threats:
Research and Practice, 1, 2, Article 11.
118
Statista, 2020, Digital Advertising Report 2020, p. 29; Econsultancy, & Upland Adestra, 2019, 2019 Email Marketing Industry Census , p. 23.
119
Kaspersky, 2020, Spam and Phishing in Q2 2020.
120
For example, online and email spam attempts try to appear credible through stating location, IP or browser information.
121
Truyens, M., Van Eecke, P., 2009, EU study on the legal analysis of a single market for the information society, European Commission.
122
European Commission, 2020, Communication from the Commission to the European Parliament and the Council, New Consumer Agenda
Strengthening consumer resilience for sustainable recovery.
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 38
Box 5: The prevalence gives rise to threats to cybersecurity in the form of malvertising and spear-
phishing
• Widespread online advertising and the increased level of personalisation of email messages
makes it difficult for users to detect malvertising and phishing or spam emails.
• The automated distribution of ads through ad-networks/ad-exchanges enables malware to be
widely distributed.
• Less technologically savvy people are expected to have their personal or their employers’
systems compromised. This can cause significant damage to consumers, companies and reduce
trust in legitimate marketing services.
Perceived benefits of prevalence advertisements and automated widespread distribution of ads:
Perceived costs due to subtle distribution of malvertising and resulting damages to consumers and
companies:
Source: Authors’ own elaboration.
2.1.2. Practices affecting purchasers of advertising including SMEs
Behavioural advertising in general offers new opportunities for advertisers, especially for the
implementation of marketing campaigns. According to the Australian Competition and Consumer
Commission (ACCC) (2019) “digital platforms have provided a new advertising avenue for small to
medium sized businesses that may not have been able to afford advertising on the high-reach of
traditional newspapers or commercial television and radio networks. For some small to medium
businesses, online advertising has become a significant part of their business models, and many have
become successful through an online only focused strategy, building a brand and community of
customers entirely through social media”
123
. An advantage for SMEs is that online advertising is usually
cheaper than placing traditional/offline ads. According to Top Draw Inc. (2020), “[t]he average cost to
reach 1,000 people with online advertising [r]anges from $3-$10, while the average cost to reach 1,000
people with traditional/offline advertising is $22 and up”
124
.
However, online advertising markets are increasingly concentrated amongst a few key players. Google
and Facebook dominate the two most commercially important markets, namely the search and the
display advertising market, and generate the greatest shares of revenue in the entire advertising
sector
125
.
123
ACCC, 2019, Digital Platforms Inquiry: Final Report.
124
Top Draw Inc., 2020, Online Advertising Costs in 2021.
125
Datanyze, 2020, Advertising Networks; Statista, 2020, Digital Advertising report 2020; Autorité de la concurrence, 2018, Sector-specific
investigation into online advertising.
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
39 PE 662.913
The competitive advantage of these companies stems primarily from the following factors: reach
126
,
large scale of advertising inventories
127
and vertical integration
128
. Each of these aspects can be positive
for advertisers in their own right, as they can enable both strong targeting and the ability to reach a
large number of people. In addition, integration can lead to efficiency gains. However, the quasi-
duopoly situation raises serious competition concerns, which have been addressed by a number of
competition authorities in recent reports
129
.
CMA's market study (2020) on digital advertising devoted some sections to Google’s and Facebook’s
abilities and incentives to exploit market power. Google and Facebook are, in theory, able to
manipulate market outcomes such as the price for advertising, even within an auction framework, by
determining the quantity and presentation of ads, by controlling the trade-off between price and
quality of adverts or certain auction parameters. All this can affect advertisers, for example, in the form
of higher costs or lower returns on investment. CMA (2020) noted that several advertisers reported
changes in Google’s policies on ad load and the presentation of search advertising, which “had the
effect of increasing the propensity for users to click on ads rather than organic links”
130
. As a result,
businesses that rely primarily on general search may have to bear higher costs for advertising, for
example by paying (more) for ads, to maintain the same amount of traffic. Changes in ad-load and
presentation of ads are also observable for Facebook
131
. Such practices can also harm consumers by
directly affecting the user experience of the platforms. Consumers can receive a poor quality of service,
seeing too much advertising or having to disclose too much personal data. In addition, consumer harm
can arise indirectly. If businesses have to bear higher costs, these costs will presumably be passed on
to consumers, who would have to pay higher prices for goods and services
132
.
126
Both Google and Facebook operate services that are popular among a vast majority of Internet users and have high audience numbers.
This enables these companies to collect constantly updated, very large and diversified volumes of data. Large data sets combined with
great data mining capabilities result in Facebook and Google having great targeting abilities. (Autorité de la concurrence, 2018, Sector-
specific investigation into online advertising.)
127
Both Google and Facebook companies are able to sell ad space of their own services but also advertising inventories of many third-party
websites. (Autorité de la concurrence, 2018, Sector-specific investigation into online advertising.)
128
Facebook and Google are vertically integrated along the advertising value chain. (Autorité de la concurrence, 2018, Sector-specific
investigation into online advertising.)
129
Autorité de la concurrence, 2018, Sector-specific investigation into online advertising; Bundeskartellamt, 2018, Online advertising. Series of
papers on “Competition and Consumer Protection in the Digital Economy”; ACCC, 2019, Digital Platforms Inquiry: Final Report; CMA, 2020,
Online platforms and digital advertising: Market study final report.
130
CMA, 2020, Online platforms and digital advertising: Market study final report, p. 237.
131
CMA, 2020, Online platforms and digital advertising: Market study final report.
132
Ibid.
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 40
Box 6: Factors that constitute large advertising providers can both harm and benefit advertisers
• Major providers like Facebook and Google benefit from great reach, large scale of advertising
inventory and from being vertically integrated. These aspects provide an opportunity for
advertisers to advertise more efficiently and effectively and reach a large audience. Small
business that operate on a limited budget and might not be able to diversify their advertising
effort across different advertising service providers may particularly benefit from the service
provided by large players.
• The very same factors may enable those players to manipulate market outcomes, thereby
harming advertisers, publishers, and consumers.
Perceived benefits of greater targeting power and reach as well as efficiency gains:
Perceived costs due to exploitation of market power:
Source: Authors’ own elaboration.
Furthermore, the design of advertising auctions itself as well as the entire structure of the advertising
value chain has the potential to disadvantage advertisers and publishers. Two of the main problems
are lack of transparency and hidden fees. Both ACCC (2019) and CMA (2020) refer explicitly to these
issues
133
. In digital behavioral advertising, advertising slots are usually allocated within milliseconds
with the help of sophisticated auction systems and algorithms. The functioning of the algorithms in
particular remains poorly understood by outsiders. As a result, most advertisers and publishers simply
accept the results of the auctions. As noted by the ACCC (2019) “it is often difficult for advertisers and
websites to understand how ad tech platforms determine what advertising inventory to buy or sell,
how winners in auctions are determined, or how advertising bid requests are passed along the ad tech
supply chain”
134
. In some processes, the success of an auction is also determined by various scores that
might be beyond the strategic control of the individual market participant. Success in an auction
depends on the automatically calculated quality of the ad, quality of the website and the demand for
specific keywords or the value of specific target groups. The method of calculating these scores is often
a “black-box” for bidders. Although less transparency means that market participants cannot outplay
the algorithm, this also makes it very difficult for them to observe whether they are making the right
decisions and investments or being exploited
135
. The value chain itself is also a factor in the lack of
transparency. We have already described the multitude of players operating along the value chain and
their functions in value creation in Chapter 1. Advertisers and publishers are often unable to monitor
who is involved and what the intermediaries are doing. This also applies to the so-called “ad tech” fees,
which are fees charged by various intermediaries along the advertising value chain
136
. Market
participants are kept in the dark when it comes to the calculation of these fees. The lack of transparency
leads to businesses making uniformed decisions or being lured into certain actions that can be
133
ACCC, 2019, Digital Platforms Inquiry: Final Report; CMA, 2020, Online platforms and digital advertising: Market study final report.
134
ACCC, 2019, Digital Platforms Inquiry: Final Report, p. 151.
135
CMA, 2020, Online platforms and digital advertising: Market study final report.
136
Geradin, D. and D. Katsifis, 2019, ‘An EU competition law analysis of online display advertising in the programmatic age’, European
Competition Journal, 15 (1):55-96.
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
41 PE 662.913
harmful
137
. Morton & Dinielli (2020) argue further that additional costs which advertisers might face
could result in companies investing less in brands and products.
After all, if companies expect a lower return on a particular innovation or investment because reaching
consumers is more expensive, they are less likely to invest and innovate in the first place
138
.
These problems are highlighted in a paper published by the World Federation of Advertisers (WFA) in
April
139
. According to a survey, 65% of global advertisers stated that they had “difficulties accessing
data related to the way actions are carried out and prices determined on large online advertising
platforms”
140
. Such information is rarely shared. According to WFA, advertisers also have difficulties
accessing ad performance data. Access to raw data on impression, viewability, reach, conversion,
engagement is rare to non-existent. The possibility of having data validated by independent auditors
is also usually restricted.
Less transparency can also be a loophole for fraud. Fraud in digital advertising or ad fraud refers to any
practice that involves manipulating views, clicks, pages and other metrics for the purpose of deception.
These conducts aim to increase the costs for companies buying advertising space without them getting
any real value in return
141
.
This direct harm to advertising companies is difficult to quantify but some
estimate the annual losses/damage from ad fraud to be in the billions
142
. Ad fraud is a major potential
threat to the whole advertising ecosystem
143
. SMEs, often on a tight marketing budget, may be hit
particularly hard by these practices. Aside from direct monetary self-interests, ad fraud has also been
reported to be used in an anti-competitive way
144
. Fake clicks on competitors’ ads not only waste their
overall marketing budget, but can also impact their daily ad auction budgets, thereby capping the
competitor’s potential market reach and brand awareness. Thus, small businesses and large companies
alike are struggling with ad fraud and some high-profile examples like Uber and P&G have drastically
cut their online advertising budgets in response
145
.
137
ACCC, 2019, Digital Platforms Inquiry: Final Report.
138
Morton , F. M. S., & Dinielli, D. C., 2020, Roadmap for Digital Advertising Monopolization Case Against Google. Omidyar Network.
139
WFA, 2021, WFA global position on advertiser access to data in the digital advertising market.
140
WFA, 2021, WFA global position on advertiser access to data in the digital advertising market, p 5.
141
Zhu, X., Tao, H., Wu, Z. Cao, J., Kalish, K. & J. Kayne, 2017, Ad Fraud Taxonomy and Prevention Mechanisms, in: Fraud Prevention in Online
Digital Advertising, SpringerBriefs in Computer Science 19-23.
142
Juniper Research, 2017, Ad Fraud to Cost Advertisers $19 billion in 2018, Representing 9% of Total Digital Advertising Spend; WFA, 2016, WFA
issues first advice for combatting ad fraud.
143
Cf. Haider et al., 2018, ‘An ensemble learning based approach for impression fraud detection in mobile advertising’, Journal of Network
and Computer Applications, 112, 126-141.
144
Cf. Gordon et al., 2020, ‘Inefficiencies in Digital Advertising Markets’, Journal of Marketing, 85(1), 7-25.
145
Fou, A., 2020, Out Of The Darkness (Of Ad Fraud) And Into The Light. Forbes.
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 42
Box 7: Lack of transparency may lead to increased costs and fraud
• Online advertising auctions are highly opaque. Although less transparency means that market
participants cannot outplay the algorithm, this also makes it very difficult for them to observe
whether they are being exploited.
• Lack of transparency can lead to exploitation of market participants and fraud, raising costs for
purchasers of advertising
Perceived benefits of lack of transparency:
None
Perceived costs due to lack of transparency, potential exploitation and fraud :
Source: Authors’ own elaboration.
2.1.3. Practices affecting suppliers of digital advertising services (publishers and smaller
SMEs functioning as Ad-Networks)
Although a large number of companies are active in the online advertising industry, Facebook and
Google account for most of the market share. These two companies have a significant market share in
both the display and search advertising markets. For instance, Google’s share in video- and search
advertising market is close to 80%, respectively
146
. The company has a particular advantage in these
two segments due to their vertical integration; they operate their own consumer-centric platforms,
which allow them to provide their own inventory and collect rich data on their users, as well as selling
advertising services. Google operates as a DSP and SSP, as well as being engaged in ad networks/ad
exchanges
147
. Although Google’s but also Facebook’s market position have remained stable over the
past few years, Amazon in particular seems to be gaining some momentum recently, mainly by
providing advertising on their own website, thereby benefiting from vertical integration as well
148
.
Small suppliers of digital advertising services which are not vertically integrated may be at a
disadvantage compared with larger players due to the current market conditions and the high barriers
for entry and expansion. The dependence on behavioural data is one of the main barriers for smaller
competing advertising networks. Larger platforms benefit from extensive volumes of data, which they
can gather in the context of their proprietary user-facing services and from third parties using their
advertising and data analytics services and tools
149
. This superior access to a combination of first- and
third-party data attracts advertisers. Although large advertisers tend to use more than one supplier of
digital advertising services to reach consumers, smaller advertisers are more likely to use just one.
146
Sunderland,J., Herrera, F., Esteves, S., Godlovitch, I., Wiewiorra, L., Taş, S., Kroon, P., Stronzik, M., Baischew, D., Nett, L., Tenbrock, S., Strube
Martins, S., de Streel, A., Kalliala, J., Huerta Bravo, J., Maxwell, W., & Renda, A., 2021, Digital Markets Act: Impact assessment support study:
annexes.
147
Statista, 2020, Digital Advertising Report 2020. Some sources covering the market for adexchanges/adnetworks indicate Google as the
clear market leader. According to these sources, Google has the largest market share, followed by other companies such as AppNexus,
OpenX and Amazon, which have less than 10% market share. European companies operating in these markets are: Axonix, Netric, and
PubNativ. W3Tech, 2021, Market share yearly trends for advertising networks; Datanyze, 2021, Market Share Category – Ad Exchanges Market
Share.
148
Sunderland,J., Herrera, F., Esteves, S., Godlovitch, I., Wiewiorra, L., Taş, S., Kroon, P., Stronzik, M., Baischew, D., Nett, L., Tenbrock, S., Strube
Martins, S., de Streel, A., Kalliala, J., Huerta Bravo, J., Maxwell, W., & Renda, A., 2021, Digital Markets Act: Impact assessment support study:
annexes.
149
Autorité de la concurrence, 2018, Sector-specific investigation into online advertising.
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
43 PE 662.913
As Google and Facebook have the ability to reach a wide audience, smaller advertisers usually opt for
these providers
150
.
In this context, bundling or tied sales can have negative consequences for small suppliers of digital
advertising services
151
. Concerns in this regard have been raised with the competition authority in
France. Stakeholders noted that some companies only provide access to targeting data if it is
purchased together with their advertising solutions and services. In addition, suppliers may prohibit
purchasers of the data from using it on competitors' advertising services
152
. Even if the option were to
exist theoretically, it would likely be very cumbersome to realise. Srinivasan (2020) attributes this to the
fact that buying tools employ different user IDs
153
. Therefore, “[a]n advertiser that uses Google’s
DoubleClick ad server […] has a harder time using a non-Google buying tool because the two tools
operate on different user IDs”
154
. According to Srinivasan (2020), even Google itself uses different IDs
for the same user when sharing data with various companies using different IDs, which makes it difficult
or impossible for companies to compile records. This practice provides Google with an informational
and thus a competitive advantage
155
.
The information disparity between Google and its rivals may widen in the future. In 2020, Google
announced that it would withdraw from allowing third-party cookies altogether by 2022. Other
browsers like Mozilla's Firefox and Apple's Safari browsers already block third-party cookies by
default
156
. These are also likely to be blocked on Google’s Chrome, one of the most used browser
worldwide, in the future as well
157
. As far as users' privacy is concerned, this development may certainly
be welcomed. However it should be noted that, Google recently acknowledged that the current
implementation of their Privacy Sandbox technology FLoC “might not be compatible with European
privacy law”
158
. Moreover, some experts believe that such a step is likely to harm advertisers, competing
exchanges and buying tools more than it will Google
159
. According to Morton & Dinielli (2020) the
former will not be able to bid efficiently or deliver payments for effective ads without knowing
consumers and how they behave
160
. Publishers’ might also lose a substantial share of ad revenue
without third-party cookies
161
.
That said, it has become apparent that publishers are taking a proactive approach to the problem and
preparing for a future without third-party cookies. A key strategy involves strengthening and
improving the collection of first-party data. One popular tool used by individual publishers is the
150
CMA, 2020, Online platforms and digital advertising: Market study final report.
151
ACCC (2019) identified four different types of bundling and sales practices prevailed in the online advertising market, namely bundling
of advertising services, or ad tech services, with a) access to website inventory, b)
with access to advertiser demand, c) with data or general
bundling of multiple services. Some of these practices are discussed in subsequent paragraphs of this subsection. (ACCC, 2019, Digital
Platforms Inquiry: Final Report.)
152
Autorité de la concurrence, 2018, Sector-specific investigation into online advertising.
153
Srinivasan, D., 2020, ‘Why Google Dominates Advertising Markets – Competition Policy Should Lean on the Principles of Financial Market
Regulation’, Standford Technology Law Review 24 (1): 55-175.
154
Ibid.
155
Ibid.
156
DEUS Marketing GmbH, 2020, Welche Auswirkungen hat die Abschaffung von Third Party Cookies auf die Werbebranche?,
OnlineMarketing.de
157
Chrome holds a market share of about 65% on the global browser market. Statcounter.com, 2021, Browser Market Share Worldwide –
Statistic.
158
Schiff, A., 2021, Google Will Not Run FLoC Origin Tests In Europe Due To GDPR Concerns (At Least For Now). Adexchanger.
159
Srinivasan, D., 2020, ‘Why Google Dominates Advertising Markets – Competition Policy Should Lean on the Principles of Financial Market
Regulation’. Standford Technology Law Review 24 (1): 55-175; Morton, F. M. S., & Dinielli D.C., 2020, Roadmap for Digital Avertising
Monopolization Case Against Google. Omidyar Network.
160
F. M. S., & Dinielli D.C., 2020, Roadmap for Digital Avertising Monopolization Case Against Google. Omidyar Network.
161
Rybnik, R., 2020, How can publishers prepare for the end of the third-party cookies?; Sluis, S., 2020, Google Chrome Will Drop Third-Party
Cookies In 2 Years, Adexchanger; Thorpe, E. K., 2020, RIP third-party cookies: What Google’s ‘pivot to privacy’ means for publishers.
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 44
implementation of registration walls. Publishers benefit from such methods both through a more
direct relationship with their audience and the ability to process valuable data themselves
162
. In
addition, publishers have been forming new alliances around first-party data and collaborating with
each other
163
.
Box 8: Asymmetric access to information and data
•
Players like Facebook and Google have a data and information advantage over other players
in the market, which naturally makes them more attractive to advertisers. Therefore, certain
practices and behaviours, such as tying the sale of data and advertising services or
prohibiting/preventing buyers from using data for competitors' advertising services, will
weaken the market position of other players (SSPs, DSPs, and ad networks/ad exchanges).
• Prohibiting third-party cookies in Chrome will, at least in the short term, do more harm to
other companies in the value chain than to Google and widen the information gap. However,
there is potential for the market to develop without relying on third-party cookies.
Perceived benefits of banning third-party cookies :
Perceived costs due to asymmetric access to information and data :
Source: Authors’ own elaboration.
Another kind of bundling and tied sales occurs in relation to Google's approach to YouTube’s ad space.
According to Morton & Dinielli (2020), YouTube inventory is only available to advertisers via Google’s
own demand side service. Considering the reach and impact of YouTube ads, advertisers might
naturally opt to use Google’s services even when they would otherwise choose a competing demand
side service. This practice increases entry barriers for new demand side services, as services that are not
able to place ads on YouTube would be less useful for advertisers in general. Morton & Dinielli (2020)
also provided a real life example of Google’s conduct to foreclose rival services from inventory.
AppNexus was an expanding service provider signing a major deal with the largest advertising agency
worldwide, when Google suddenly cut off AppNexus’s ability to place ads on YouTube and other
supply that was available through real-time bidding. AppNexus’s ad buyers had to set up a contract
directly with Google before they were allowed to access Google’s inventory again
164
.
162
Thorpe, E. K., 2020, RIP third-party cookies: What Google’s ‘pivot to privacy’ means for publishers.
163
Thorpe, E. K., 2020, First-party data has empowered publishers to experiment with personalisation and better advertising.
164
Kafka, P., 2010, Google Cuts Off AppNexus, and the Ad Tech World Shudders. AllThingsD.com; Kincaid, J., 2010, Google (Temporarily) Blocks
AppNexus From Its Ad Exchange.
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
45 PE 662.913
Large players, which are mostly vertically integrated, often face accusations of impeding
interoperability and self-preferencing. One concern expressed by some stakeholders is that Google
could potentially be favouring its own SSP when its DSPs decides where to submit its bids
165
. Another
concern relates to SA360
166
. Some stakeholders claimed that Google benefits from greater
interoperability between SA360 and Google Search compared to other search engines. According to
CMA (2020), Microsoft claimed, among others, that new features from Google Ads are adopted on
SA360 much faster than the innovations developed by Bing
167
.
Exclusivity clauses, like agreements requiring third-party websites to source their ads largely or even
exclusively from the respective contracting party, can also restrict competition. The inclusion of such
clauses in contracts with third party websites was subject to an antitrust proceeding by the European
Commission against Google, which was closed in 2019. The European Commission considered
Google’s practice to be harmful to competition
168
. Another example that shows that Google takes
action that strengthens its competitive position is “Project Bernanke”. According to an article written
by Lyons (2021) the company used its ability “to access historical data about bids made through Google
Ads, to change bids by its clients and boost the clients’ chances of winning auctions for ad impressions,
putting rival ad tools at a disadvantage”
169
.
Box 9: Conducts like exclusive clauses, self-preferencing, bundling and tied sales as well as
impediments to interoperability are associated with large players
• Google has been accused of exploiting their market power through various types of practices.
• Several Authorities have taken a closer look at the individual conducts, while some are
confirmed, for others it can only be assumed that the companies in question may have
incentives to engage in such activities. It is clear, however, that if exercised those conducts
would harm competitors and other companies in the value chain.
Perceived benefits of such practices:
None
Perceived costs due to these practices :
Source: Authors’ own elaboration.
165
CMA, 2020, Online platforms and digital advertising: Market study final report. Appendix M: intermediation in open display advertising, p. M108.
166
SA360 is a Google-owned search management platform used by agencies and marketers to manage and optimise campaign across
multiple search networks and media channels (see CMA, 2020, Online platforms and digital advertising: Market study final report; Google,
2021, Search Ads 360 overview - What’s Search Ads 360).
167
CMA, 2020, Online platforms and digital advertising: Market study final report.
168
European Commission, 2019, Antitrust: Commission fines Google €1.49 billion for abusive practices in online advertising.
169
Lyons, K., 2021, Google reportedly ran secret ‘Project Bernanke’ that boosted its own ad-buying system over competitors. The information was
revealed in unredacted court documents. The Verge.
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 46
2.2. Steps taken by consumers and commercial providers to address
perceived challenges
Consumers who are weary of online advertising and perceive the adverts they see as rather intrusive
have the option to improve their browsing experience through the use of ad-avoidance technologies
(AAT). Ad-blockers mitigate the nuisance costs from advertisements and promise advertisement-free
visits of websites. Approximately 25% of all internet users in the US used such software in 2019
170
.
Malloy et al. (2016)
171
analysed a large scale data set consisting of over 2 million users in the US and
other key countries that records over a trillion website interactions. They find that ad-block usage
amounts to approximately 18% in the US and up to 49% in Germany. They estimate that the
effectiveness of ad-blocking software is limited to the extent that they only block roughly half the ads
seen by users without ad-blocking software. The consequential losses in monthly ad revenue for the
largest 10 scraped publisher websites are estimated to be approximately USD 120,000 to USD 3.9
million
172
(given an assumed average CPM of USD 1)
173
.
However, ad-blocking users might be short-sighted with respect to their own interests. While their
behaviour may indeed improve their online browsing experience in the short-run, it might decrease
the availability of free, advertising financed content in the long-run. Cutting off the advertising revenue
stream constitutes a severe threat to the business models of numerous content providers (publishers)
on the Internet. Publishers which employ technologies to exclude ad-blocking users or ask for the AAT
to be disabled on their site face commercial challenges as well: 61% of consumers would not return to
such a site again and only 38% of polled online users in the US would turn off their ad-blocker if asked
to by a website
174
. But, AATs also offer an edge for publisher. Since it is possible to detect whether a
visitor uses AAT, it can be interpreted as a signal. Those who use AAT are relatively more sensitive to
adverts; while those who do not are not. Aseri et al. (2020)
175
and Despotakis et al. (2020)
176
show that
this identification enables publishers to differentiate the advertisement intensity between these
different user groups on their site in the form of ad-light website designs and raise aggregate
advertising revenues accordingly. Consequently, this also leads to increased ad intensity on websites
and online services for non-ad blocking users
177
.
At the same time, developers of ad-blocking tools and privacy enhancing technologies may be doing
so to benefit from ad avoidance activities by consumers. So-called “whitelisting” business models
rely on passing through some advertisements in return for a revenue share of the regained ad revenues
of large publishers. These AATs could therefore constitute a new form of gatekeeping role in the open
Internet. The recent dispute between Apple and Facebook about new privacy enhancing technologies
that restrict ad tracking in Apple’s upcoming operating system update (iOS 14.5) underpins this
conclusion
178
.
170
eMarketer.com, 2020, Ad Blocking Is Slowing Down, but Not Going Away.
171
Malloy, M., McNamara, M., Cahn, A., & Barford, P., 2016, ‘Ad blockers: Global prevalence and impact’, Proceedings of the 2016 Internet
Measurement Conference (pp. 119-125).
172
This corresponds roughly to EUR 113.841 to EUR 3,7 million [exchange rate 31.12.2016].
173
This corresponds roughly to EUR 0.95 [exchange rate 31.12.2016].
174
eyeo.com, 2020, Ad-blocking usage report 2019.
175
Aseri, M., Dawande, M., Janakiraman, G., & Mookerjee, V. S., 2020, ’Ad-Blockers: A Blessing or a Curse?’, Information Systems Research, 31
(2) 627-646.
176
Despotakis, S., Ravi, R., & Srinivasan, K., 2020, ’The beneficial effects of ad blockers’, Management Science.
177
Anderson, S. P., & Gans, J. S., 2011,’ Platform siphoning: Ad-avoidance and media content’, American Economic Journal:
Microeconomics, 3(4), 1-34.
178
Warren, T., 2020, Facebook hits back at Apple with second critical newspaper ad. The Verge.
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
47 PE 662.913
The conflicting interests on advertising companies and publishers on one hand and developers of AATs
on the other has led to a cycle of technological developments to circumvent the technologies of the
respective opposing party. This is often referred to as the ad-blocking “arms race” in scientific
research
179
. Traditional ad-blocking browser extensions maintain so-called filter lists which are run
through with every website visit. If a visited website contains advertisements, they are filtered out if
their respective script is detected through an entry in the filter list. Hence, ad-blocking filter lists are per
definition a lagging tool in that they contain only known ad scripts. Ad-blocking detection, however,
evolved quickly over the recent years and works in two ways. Either the detection tool checks whether
ads are not downloaded upon a user’s visit or a potential ad blocking extension is detected because it
greatly reduces the loading time of the visited website
180
181
. While the academic consensus is that this
arms race between ad-block developers and consumers on the one side and advertisers and publishers
on the other side will be won finally by the latter party, recent research by Storey et al. (2017) proposes
new technological approaches to ad-blocking that employ elements of rootkits to prevent detection
by anti-ad blocking scripts and could potentially challenge this view
182
.
Box 10: Widespread advertising facilitates the adoption of ad-blocking software (AAT) by users which
threatens the business models of publishers and advertisers
• Ad-blocking usage is increasing worldwide as a means to limit ad exposure while browsing.
• Ad-blocking creates opportunities for publishers to strategically differentiate between ad
blocking and non-ad-blocking users and offer different website experiences (e.g. ad-light
versions).
• Ad blocking signifies a desire by consumers for greater privacy and can motivate the
development of new business models; however
• The short-term relief from intrusive advertisements could undermine advertising based
business models and the availability of “free” content in the long-term.
• Developers of AAT could act as a new bottleneck in reaching consumers and capture a fraction
of publishers’ advertising revenue.
Perceived benefits of ATT and the reduced ad exposure and increased privacy for consumers:
Perceived costs due to increased adoption levels of ATT for non-users and advertisers/publishers:
Source: Authors’ own elaboration.
179
Storey, G., Reisman, D., Mayer, J., & Narayanan, A., 2017, The future of ad blocking: An analytical framework and new techniques, arXiv
preprint arXiv:1705.08568.
180
Mughees, M. H., & Qian, Z., 2017, ‘Detecting anti ad-blockers in the wild’, Proceedings on Privacy Enhancing Technologies, 2017 (3):127–
142.
181
Nithyanand, R., Khattak, S., Javed, M., Vallina-Rodriguez, N., Falahrastegar, M., Powles, J. E., De Cristofaro, E., Haddadi, H., Murdoch, S.J.,
2016, ‘Adblocking and counter blocking: A slice of the arms race’, Proceedings of the 6th USENIX Workshop on Free and Open
Communications on the Internet (FOCI '16).
182
Storey, G., Reisman, D., Mayer, J., & Narayanan, A., 2017, The future of ad blocking: An analytical framework and new techniques. arXiv
preprint arXiv:1705.08568.
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 48
2.3. Analysis of costs and benefits of different practices
In the above Sections 2.1 and 2.2 we identified practices and issues that are either inherent to the online
advertising value chain or stem from countermeasures that have already been undertaken as a
response. However, the potential harm that might result for one or multiple involved parties does not
paint the complete picture. Practices or issues that arose over the last decade have done so because
they offer also a range of benefits and advantages in the form of higher efficacy, a reduction in costs or
create new revenue streams that enable the provision of content and innovative services.
For this reason, a holistic approach is needed when evaluating specific practices and considering
whether a remedial intervention is required. The evaluation of benefits and harms cannot be based on
quantitative metrics (i.e. in term of monetary costs or gains), since the plethora of issues is not readily
quantifiable without access to proprietary data. However, the concluding boxes in the previous two
sections seek to provide a qualitative assessment of the potential harms and benefits associated with
each identified issue. The assessment is based on extensive research of the academic and non-
academic literature and is also informed by insights from interviews conducted with different
stakeholders for this study. This assessment could provide a first indication of the severity of one
practice in relation to the other practices identified and, thus, offer guidance as to which should receive
priority when considering regulatory intervention.
In Table 2, we summarise the benefits and potential harms associated with each practice and problem.
The order in which the practices and issues are listed reflects the sequence in Sections 2.1 and 2.2. Thus,
the order of practices shown in the table should not be interpreted as a ranking of potentially harmful
practices.
We visualise the severity of associated costs or the potential for harm that originates from an issue in
red from moderate (1), somewhat harmful (2) to very harmful (3). Similarly, the associated benefits are
visualised in green and classified from moderate (1), somewhat beneficial (2) to very beneficial (3).
Please note that these ratings should only be interpreted in an ordinal manner and not as a cardinal
scale. Hence, a practice that is classified as being harmful should be more problematic compared to
another one whose harms are considered only moderate but it is not twice as harmful as the latter.
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
49 PE 662.913
Table 2: Cost-benefit assessment
Potentially
harmful practice
Benefits
Assessment
(
moderate (1) to
very beneficial (3))
Harm
Assessment
(m
oderate (1) to
very harmful (3))
Practices affecting consumers (and SMEs as viewers of advertising)
Behavioural
targeting gives rise
to privacy concerns
Behavioural targeting relies on the analysis of
profound user data. It offers more relevant
advertisement that aligns more with an individual’s
interests.
In the targeting process personal user data (e.g., IP-
address, browsing history, age, gender, etc.) is gathered
and analysed. This data is potentially also shared across
various market players without the consumers’
knowledge or consent. Behavioural targeting methods,
thus, raise severe privacy concerns.
Obfuscation and
location based
targeting reduces
consumer choice
Product reviews on comparison sites are often
sponsored content to cover the most popular
products. Further, advertisements are often also
targeted based on a consumers’ location or are not
purchasable in specific areas/countries (geo-
blocking). This provides the consumer with
potentially more relevant offers in their proximity.
The obfuscation of advertisement as neutral product
reviews narrows the consumers’ focus on the covered
products. Location based targeting of advertisement
may implicitly foreclose specific products to some
consumer segments. Both practices imply a reduced
consumer choice.
Advertising design
exploits
behavioural biases
and includes dark-
patterns
Behavioural biases of consumers are used to increase
the likelihood that an advertisement is clicked on.
Dark pattern elements are employed to nudge
consumers to give their consent to share their
personal data.
None
Design elements of advertising (dark patterns) that aim
at exploiting consumers’ behavioural biases lead to
consumers behaving in a way that is contrary to their
own preferences (e.g., consent to data sharing, clicking
on non-avoidable ads).
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 50
Potentially
harmful practice
Benefits
Assessment
(
moderate (1) to
very beneficial (3))
Harm
Assessment
(m
oderate (1) to
very harmful (3))
Discrimination and
harmful targeting
of vulnerable
consumers through
data-driven
targeting with RTB
The data-driven targeting implemented in the RTB
processes provides more relevant advertisements to
users. The effectiveness of behavioural targeting
relies on personal data such as the analysis of the
browsing history. Based on this, vulnerable groups
can also be protected (in not seeing certain ads, e.g.,
minors and alcoholic beverages).
The data-driven targeting may target vulnerable
consumers in a harmful way. Due to their past browsing
behaviour, consumers with, e.g., a bad eating habit or
gambling problem may be especially exposed to
advertisements in these fields. This poses an increased risk
to the wellbeing of such consumers.
Malvertising &
Spear-Phishing
Automated distribution of advertising through Ad-
Networks and RTB reduces costs and increases the
reach of advertising campaigns.
The automated distribution of advertisements makes it
easy for cybercriminals to spread malware in the form of
malvertising or spear-phishing. This can cause significant
damages to affected consumers and SMEs.
Practices affecting advertisers including SMEs
Market
concentration of
intermediaries
Large advertising intermediaries like Facebook and
Google offer a great variety of ad inventory and thus a
wide reach to advertisers. Due to their extensively
gathered data, they can provide profound targeting of
ads to increase the ROI for the advertisers. Especially
smaller advertisers benefit from the large scale of these
intermediaries since they regularly do not have the
funds to distribute their marketing efforts across many
intermediaries.
The large concentration implies only a marginal
competitive pressure among these two large
intermediaries. This bears the potential for higher prices
towards advertisers and buyer power towards publishers
which would in the long run lead to market inefficiencies.
Non-transparent price setting in RTB, as well as
performance and conversion data of distributed ads is
difficult to challenge by rivals.
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
51 PE 662.913
Potentially
harmful practice
Benefits
Assessment
(
moderate (1) to
very beneficial (3))
Harm
Assessment
(m
oderate (1) to
very harmful (3))
Lack of
transparency
towards advertisers
& Ad-fraud.
Details of the inner workings of the ad auctions (RTB)
are not disclosed in order to prevent other market
participants to game the algorithm.
None
The outcomes of the ad-auctions are non-transparent.
Advertisers receive no data on how their bid and ad
performed during the auction. Access to performance
data of the distributed ads is also not provided to
advertisers making it impossible to audit their ad
spending and calculate reliable ROI estimates. This
foreclosure of data also opens the door for ad-fraud as
fraudulent and artificial conversion cannot be detected
easily.
Practices affecting suppliers of advertising space (publishers) and SMEs functioning as Ad-networks
Asymmetric access
to information and
data
Large intermediaries such as Facebook and Google
have an information advantage over competing ad-
networks and publishers. Trends that facilitate this
information gap are the phasing out of third-party
cookies by Google. This creates new potential for the
market to develop without relying on cookies and
also improves consumers’ privacy.
The information advantage of these large intermediaries
makes their services more attractive to advertisers. Their
data sets may act as an entry barrier for smaller ad-
networks or increase costs for other publishers. The
phasing out of third-party cookies by Google may weaken
competitor’s positions in the long run as the knowledge
gap and the dependence on Google’s analytics services
increases.
Exclusivity clauses,
self-preferencing
and bundling
None
Large intermediaries such as Google and Facebook are
known to strengthen their market positions through
practices of bundling and self-preferencing. The vertical
integration of these players leads to bundling practices.
Exemplarily, advertising space on YouTube (Google),
Facebook (Facebook) are only accessible for advertisers
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 52
Potentially
harmful practice
Benefits
Assessment
(
moderate (1) to
very beneficial (3))
Harm
Assessment
(m
oderate (1) to
very harmful (3))
through the respective intermediaries at highly non-
transparent costs. This puts other publishers at an
disadvantage as they do not have access to this
exclusively marketed ad-inventory. Competitive concerns
usually associated with these conducts should also apply
here.
Effect of already undertaken countermeasures
Adoption of Ad-
blocking
technologies by
consumers
Ad-blocking software reduces the exposure to
advertisements while browsing. Regularly, these tools
also the tracking of users and hence increase privacy.
With the adoption consumers can express their
demand for less intrusive ads.
The adoption of ad-blockers may threaten advertising
based business models of publishers. The provision of
content ‘for free’ in the internet could be in jeopardy.
Adopters of ad-blocking exert a negative externality on
non-adopters since they are less sensitive to ads and
their exposure may increase. Whitelisting practices of ad-
block developers could constitute a new gatekeeping
role in the internet.
Source: Authors’ own elaboration.
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
53 PE 662.913
HOW FAR DOES EXISTING LEGISLATION GO IN ADDRESSING
THE PROBLEM?
3.1. Legal framework
3.1.1. Overview of the legal framework
Although there is no overarching legislative framework covering online targeted advertising at the EU
level, different legislative measures apply to advertising (including targeted advertising) and provide
certain rights to recipients. The current legislative framework is complex and only addresses certain
aspects of online targeted advertising and sometimes in an indirect manner.
These instruments do not include a definition of “targeted advertising” or “behavioural advertising”
but usually refer to the more general concepts of “commercial communication”, “commercial practice”
or “advertising”, which should be interpreted widely according to the Court of Justice of the European
Union (CJEU)
183
. It is therefore commonly understood that targeted advertising falls under the scope of
these legal instruments.
The main relevant instruments are listed below. Some are specific and their scope is limited to the
digital environment, although they may not cover all types of platforms:
• eCommerce Directive
184
(the Services Directive
185
and the Proportionality Test Directive for
regulated professions)
186
;
• P2B Regulation
187
;
• ePrivacy Directive
188
(incl. a new proposal of ePrivacy Regulation)
189
; and
• Audiovisual Media Services Directive (AVMS), as amended in 2018
190
.
183
CJEU, 11 July 2013, C-657/11, BEST, EU:C:2013:516, pt. 35.
184
European Parliament and the Council of the European Union, 2000, Directive 2000/31/EC of the European Parliament and of the Council
of 8 June 2000 on certain legal aspects of information society services, in particular electronic commerce, in the Internal Market, OJ L 178
of 17.07.2000 (hereafter the “eCommerce Directive”).
185
European Parliament and the Council of the European Union, 2006, Directive 2006/123/EC of the European Parliament and of the Council
of 12 December 2006 on services in the internal market, OJ L 376 of 27.12.2006.
186
European Parliament and the Council of the European Union, 2018, Directive (EU) 2018/958 of the European Parliament and of the Council
of 28 June 2018 on a proportionality test before adoption of new regulation of professions, OJ L 173 of 9.7.2018.
187
European Parliament and the Council of the European Union, 2019, Regulation (EU) 2019/1150 of the European Parliament and of the
Council of 20 June 2019 on promoting fairness and transparency for business users of online intermediation services, OJ L 186 of
11.7.2019.
188
European Parliament and the Council of the European Union, 2002, Directive 2002/58/EC of the European Parliament and of the Council
of 12 July 2002 concerning the processing of personal data and the protection of privacy in the electronic communications sector, OJ L
201 of 31.7.2002 (Directive on privacy and electronic communications).
189
European Commission, 2017, Proposal for a Regulation of the European Parliament and of the Council concerning the respect for private
life and the protection of personal data in electronic communications and repealing Directive 2002/58/EC, COM(2017) 10 final -
2017/0003 (COD).
190
European Parliament and the Council of the European Union, 2010, Directive 2010/13/EU of the European Parliament and of the Council
of 10 March 2010 on the coordination of certain provisions laid down by law, regulation or administrative action in Member States
concerning the provision of audiovisual media services (Audiovisual Media Services Directive) (codified version), OJ L 095 of 15.4.2010.
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 54
Others are general and apply to all kinds of advertising, irrespective of the means used to display the
message:
• Unfair Commercial Practices Directive (UCPD)
191
;
• Directive on misleading and comparative advertising
192
;
• Consumer Rights Directive (CRD)
193
; and
• General Data Protection Regulation
194
(GDPR).
EU legislation also bans the advertising of certain types of products such as tobacco products, which
are not covered in this study
195
.
Moreover, competition law applies to all digital platforms and may have far reaching implications for
the provision of online advertising. Several national competition authorities including authorities in
France, Germany and the Netherlands have concluded or launched sector enquiries or general
investigations into online advertising, while authorities in France
196
, Italy
197
, and the UK
198
have opened
or decided individual cases. The European Commission has also opened a case against Google
concerning online advertising
199
. Potential competition concerns in the online advertising market
include: tying, bundling, exclusivity, leveraging, discrimination, interoperability issues and access to
data.
This legal framework is set to be complemented shortly, by the Digital Services Act (DSA)
200
and Digital
Markets Act (DMA)
201
, which were proposed by the European Commission on 15 December 2020. These
proposals are currently being adopted under the co-decision procedure and as such may be amended.
Annex 1 of the report contains an overview of these legislative measures (including the DSA and DMA
proposals), especially with regard to their (i) nature and scope, (ii) substantial measures (relevant for
targeted advertising), and (iii) enforcement.
191
European Parliament and the Council of the European Union, 2005, Directive 2005/29 of the European Parliament and of the Council of
11 May 2005 concerning unfair business-to-consumer commercial practices in the internal market, OJ L 149 of 11.06.2005.
192
European Parliament and the Council of the European Union, 2006, Directive 2006/114/EC of the European Parliament and of the Council
of 12 December 2006 concerning misleading and comparative advertising (codified version), OJ L 376 of 27.12.2006.
193
European Parliament and the Council of the European Union, 2011, Directive 2011/83 of the European Parliament and of the Council of
25 October 2011 on consumer rights, OJ L 304 of 22.11.2011.
194
European Parliament and the Council of the European Union, 2016, Regulation 2016/679 of the European Parliament and of the Council
of 27 April 2016 on the protection of natural persons with regard to the processing of personal data and on the free movement of such
data, OJ L 119 of 04.05.2016.
195
European Parliament and the Council of the European Union, 2003, Directive 2003/33/EC of the European Parliament and of the Council
of 26 May 2003 on the approximation of the laws, regulations and administrative provisions of the Member States relating to the
advertising and sponsorship of tobacco products, OJ L 152, 20.6.2003, p. 16-19.
196
The NCA fined Google €150m for having abused its dominance in search advertising and reiterated Google’s obligation to “define the
operating Rules of its advertising platform in an objective, transparent and non-discriminatory manner”. Autorité de la concurrence, 2019,
Decision 19-D-26 of 20 December 2019 regarding practices employed in the online search advertising sector.
197
The NCA investigates whether Google is engaged in an exclusionary abuse of dominance in the online display advertising market, where
its ad intermediation services rely on “non-replicable” user data that Google collects within its ecosystem.
198
The NCA investigates Google's proposals to disable third-party cookies in its Chrome browser and replace them with new tools for
targeted advertising (see CMA, 2021, Investigation into Google’s ‘Privacy Sandbox’ browser changes.
199
Cases AT.40660 and AT.40670.
200
European Commission, 2020, Proposal of the Commission of 15 December 2020 for a Regulation of the European Parliament and of the
Council on a Single Market For Digital Services (Digital Services Act) and amending Directive 2000/31/EC, COM(2020) 825 final, 2020/0361
(COD).
201
European Commission, 2020, Proposal of the Commission of 15 December 2020 for a Regulation of the European Parliament and of the
Council on contestable and fair markets in the digital sector (Digital Markets Act), COM(2020) 842 final, 2020/0374 (COD).
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
55 PE 662.913
3.1.2. Soft Law instruments
Soft law instruments have an important influence in this sector. Obviously, the binding force of these
instruments is less than legal provisions
202
but they have the advantage of being more flexible while
also reflecting the latest market developments.
In particular, the ICC Advertising and Marketing Communications Code (2018) adopted by the
International Chamber of Commerce (ICC) contains rules which are particularly well adapted to digital
advertising such as rules on the identification, transparency and identity of the marketer, as well as a
prohibition on harassment, and the need to respect the consumer’s choice not to receive direct
marketing communication, including giving the consumer a right of withdrawal
203
. All parties involved
should also be bound by the principle of responsibility (not only the marketer, but also in particular
tech companies, platforms, market influencers or the persons preparing the algorithms used in the
marketing campaigns)
204
. The ICC Code also contains principles applicable to so-called Interest-
Based Advertising (IBA), or online behavioural advertising. According to these principles, data
collection and use should be transparent (with clear and visible notice), and users should be able
to exercise their choice in this respect (user control). Companies should also refrain from
creating IBA based on sensitive data except with prior consent and in accordance with the
applicable legal framework.
3.1.3. Complexity of the legal framework
The legal framework applicable to advertising in general, and to digital targeted advertising in
particular, is very complex. Depending on the parties involved (consumers, traders, minors, etc.), the
content of the advertising or the means used to display the message, some legislative measures will be
applicable, while other measures will not.
The legal framework is also fragmented within the 27 Member States, since the older directives –
eCommerce Directive, for instance – are minimal harmonisation instruments (which means that some
Member States have introduced more far-reaching rules, thereby threatening the achievement of the
Digital Single Market).
202
The violation, by a trader, of the commitments set forth by a Code of conduct could however be considered as a misleading commercial
practice, under Article 6(2)(b) of the UCPD (subject to the requirements prescribed by this provision).
203
This Code is available online at: ICC, 2018, ICC Advertising and Marketing Communications Code.
204
Article C21 of the ICC Code.
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 56
Table 3 below highlights for each of the legal instruments their nature and (personal and material)
scope of application.
Table 3: Nature and scope of the legal measures
Nature Personal Scope Material Scope
Specific
eCommerce
DIR
Directive (min
harmonisation)
B2B and B2C
(Information society service
provided to the recipient of a
service)
Commercial communications which are
part of, or constitute, an information
society service
P2B
Regulation
Regulation
Platform to Business (only when
business users target consumers)
Practices such as ranking, differentiated
treatment and access to data
DSA Proposal
Regulation
Technical intermediaries (ISPs)
Hosting service providers
Online platforms
Very large online platforms
(information to be made publicly
available)
(in their relationship with users)
Information duties
DMA Proposal
Regulation
Gatekeepers
8 types of digital services, including
online advertising when provided in
combination with intermediation
services, search engines, social
networks, video-sharing,
communication services, cloud
computing or operating systems
AVMS
DIR
Directive (min
harmonisation)
Video sharing platforms – end
users
Advertising (content rules)
Information duties
Processing of personal data of minors
General
UCPD
Directive (full
harmonisation)
B2C
Commercial practices (incl. marketing
and advertising) - broad
Misleading
and
Comparative
Advertising
DIR
Directive
B2C and B2B (depending on
provisions)
Misleading and Comparative
Advertising
CRD
Directive (full
harmonisation)
B2C
Conclusion of B2C contracts (incl.
obligations of information before
conclusion of B2C contracts) - broad
GDPR
Regulation
Relationships between data
controllers/processor and data
subjects
Processing of personal data (both
broad concepts)
Source: Authors’ own elaboration.
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
57 PE 662.913
3.2. How far do existing substantial rules address the problems and
what are the gaps?
3.2.1. Processing of personal data and privacy
Targeted advertising is only possible when the online activities of users
205
can be traced and when their
personal data can be processed, for instance to build profiles relating to the targeted audience
206
. The
ePrivacy Directive and the GDPR are the two main instruments that frame how this can take place.
The ePrivacy Directive mainly applies to the processing of personal data in the electronic
communication sector but the directive also contains rules on “cookies” which have a broader scope
since they protect the terminal equipment of end users. Accordingly, storing information and gaining
access to information already stored in terminal equipment of a subscriber/user (e.g. a phone,
computer, connected vehicle or smart speaker) requires the prior informed consent of subscribers
irrespective of any qualification of personal data of these information
207
. Prior consent of the
subscriber/user is not required when gaining access is necessary (i) for the purpose of carrying
transmission of a communication or (ii) to provide an information society service requested by the
user/subscriber. With regard to this second exception, the EDPB has recently stated that profiling for
the purpose of advertising will not be considered as a service expressly requested by the
user/subscriber
208
. Hence, this provision creates de facto an opt-in mechanism for the use of
tracking cookies
209
or tracking pixels
210
for the purpose of targeted advertising.
A new ePrivacy Regulation
211
is currently being negotiated between the European Parliament and the
Council. The Commission adopted its proposal in 2017, the European Parliament adopted the report of
the Committee on Civil Liberties, Justice and Home Affairs in October 2017 and the EU Council adopted
its general approach on 10 February 2021. In these three versions of the text, prior consent of the user
is required to (i) collect of information on the terminal equipment and (ii) use processing and storage
capabilities of terminal equipment
212
. This wording has the effect of ensuring that all tracking
mechanisms (such as device fingerprinting) requires the users’ prior consent
213
. Contrary to the
Commission’s proposal and the European Parliament’s position of 2017, the Council’s general
approach allows further processing for compatible purposes, provided that, inter alia the information
is not used to determine user characteristics or to build profiles
214
.
205
Clifford, D., 2014, ‘EU Data Protection Law and Targeted Advertising: Consent and the Cookie Monster - Tracking the crumbs of online
user behaviour’, JIPITEC, pp. 194 – 195.
206
Article 29 Data Protection Working Party, 2010, Opinion 2/2010 on online behavioural advertising, 22 June 2010, WP171, p. 4; Wachter,
S., 2020, ‘Affinity Profiling and Discrimination by Association in Online Behavioural Advertising’, Berkeley Technology Law Journal, Vol. 35,
No. 2, 2020.
207
CJEU, judgement Planet49 GmbH, 1st October 2019, C-673/17, EU:C:2019:801, points 70 and 71.
208
European Data Protection Board, 2021, Guidelines 02/2021 on Virtual Voices Assistants, 9 March 2021, p. 23.
209
Article 29 Data Protection Working Party, 2010, Opinion 2/2010 on online behavioural advertising, 22 June 2010, WP171, p. 25.
210
European Data Protection Board, 2020, Guidelines 8/2020 on the targeting of social media users, 2 September 2020, pp. 19-21.
211
European Commission, 2017, roposal for a Regulation of the European Parliament and of the Council concerning the respect for private
life and the protection of personal data in electronic communications and repealing Directive 2002/58/EC, COM/2017/010 final -
2017/0003 (COD).
212
Commission ePrivacy Regulation proposal, Article 8(1)(b); General approach of the Council, Article 8(1)(b); European Parliament report of
the Committee on Civil Liberties, Justice and Home Affairs, Article 8(1)(b).
213
See BEUC, 2017, Proposal for a regulation on Privacy and electronic communication (ePrivacy) – BEUC position paper, BEUC-X-2017-059,
09/06/2017, p. 9; Commission ePrivacy Regulation proposals, Recital 20.
214
General approach of the Council, Article 8(1)(h)(iii).
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 58
The proposed ePrivacy Regulation also creates the possibility for users to exert their opt-in through
software technical settings
215
.
In addition to the ePrivacy Directive, which relates to placing and accessing to cookies on terminal
equipment, the subsequent processing operation on personal data will trigger application of the GDPR
rules.
The GDPR requires that data processing operations rely on one of the limitative legal grounds provided
in Article 6. With regard to targeted advertising, interpretative guidelines adopted at the EU level
tend to emphasise the need to obtain the consent of the data subject.
The Article 29 Working Party
216
had already stated that the legitimate interest is not a suitable legal
ground for advertising purposes involving tracking data subjects across websites/devices/services,
involving data brokers and implying intrusive profiling
217
. The European Data Protection Board (EDPB)
has also excluded performance of a contract as a legal ground for behavioural advertising
218
. In its
recent guidelines on targeting of social media users, the EDPB tends to consider legitimate interest as
a valid legal ground only in relation to targeting on the basis of data directly provided by the data
subjects. For targeting based on inferred and observed data, consent is generally considered to be
more suitable.
Finally, when part of the processing requires consent under the ePrivacy Directive (e.g. accessing
cookies stored in the end users’ terminal equipment), the EDPB considers that consent under the GDPR
is an adequate legal ground for the subsequent processing operations. This aims to avoid use of the
legitimate interest ground as a way to lower the level of protection granted by the consent requirement
under the ePrivacy Directive
219
.
a. Opt-in or opt-out?
When data processing is based on consent, which is typically the case for targeted advertising as seen
above, this in practice means that consumers need to opt-in
220
to targeted advertising (i.e. as consent
implies a real choice from the data subject)
221
. In addition, as part of a valid consent, the data subject
must be able to withdraw consent and in which case, the data controller must stop the processing
222
.
This creates the possibility to opt-out for the data subject to processing operations linked to targeted
advertising.
215
Commission ePrivacy Regulation proposal, Article 9 and Article 10(2); European Parliament report of the Committee on Civil Liberties,
Justice and Home Affairs, Article 9 and Article 10; General approach of the Council, Article 4a(2).
216
The Article 29 Working Party was established under Article 29 of the Data Protection Directive (95/46/EC) as an independent EU advisory
body on data protection. The GDPR replaced the Working Party by the European Data Protection Board (EDPB).
217
Article 29 Data Protection Working Party, 2018, Guidelines on Automated individual decision-making and Profiling for the purposes of
Regulation 2016/679, 3 October 2018, WP251rev.01, p. 15.
218
European Data Protection Board, 2019, Guidelines 2/2019 on the processing of personal data under Article 6(1)(b) GDPR in the context of
the provision of online services to data subjects, 8 October 2019, p. 14.
219
See European Data protection Board, 2019, Opinion 5/2019 on the interplay between the ePrivacy Directive and the GDPR, in particular
regarding the competence, tasks and powers of data protection authorities, 12 March 2019, p. 14, and especially footnote 29; European
Data Protection Board, 2020, Guidelines 8/2020 on the targeting of social media users, 2 September 2020, p. 22; European Data Protection
Board, 2021, Guidelines 02/2021 on Virtual Voices Assistants, 9 march 2021, p. 11; European Data protection Board, 2021, Guidelines
01/2020 on processing personal data in the context of connected vehicles and mobility related applications, adopted on 9 March 2021,
pp. 7-8.
220
GDPR, Article 4(11).
221
European Data Protection Board, 2020, Guidelines 05/2020 on consent under Regulation 2016/679, 4 May 2020, p. 7.
222
European Data Protection Board, 2020, Guidelines 05/2020 on consent under Regulation 2016/679, 4 May 2020, p. 24.
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
59 PE 662.913
Even where processing is based on the “legitimate interests of the controller”, the data subject always
enjoys a right to object to processing of its personal data for direct marketing purposes (including any
related profiling). The right to object may therefore also be used as an opt-out mechanism by data
subjects facing targeted advertising.
In any event, when a data subject (i) withdraws consent for data processing and no other legal ground
can be invoked by the data controller or (ii) objects to the processing for direct marketing purposes,
this person can also request erasure of personal data
223
. In such cases, the data controller must, in
principle, communicate to each recipient of the personal data the erasure request of the data subject
224
.
Lastly, under the GDPR, data processing for the purpose of displaying targeted advertising will always
necessitate prior opt-in in three specific cases:
• First, when an information society service is offered directly to children between the age of
13 and 16 (depending on what age the Member States decide children can validly give their
consent), consent of the holder of the parental responsibility is required in order to process
their data
225
. Recital 38 highlights the need for specific protection when children data are used
for marketing purposes.
• Second, because data subjects have the right not to be subject to a decision based solely on
automated processing, including profiling, which may significantly affect them, prior explicit
consent of the data subject may also be required under Article 22 GDPR
226
. In that case
targeted advertising would be subject to an opt-in of the targeted person.
• Third, the same requirement applies to processing of special categories of data such as
health related data, which require prior explicit consent of the data subject unless the data
have been manifestly made public by the data subject
227
.
b. Access to data under the P2B Regulation
As recognised by the P2B Regulation, the data (whether personal or not) provided or generated
through “online intermediation services” can be of crucial importance. For instance, the data
concerned could be ratings or reviews accumulated by business users on the intermediation
services
228
. The ability to access these data or to use them can lead to important value creation, not
only from a wider economic perspective, but also for the intermediation services and their business
users
229
. Such data might be used, directly or indirectly, for targeted advertising purposes in particular
to provide data on the effect of the digital advertising campaign. The P2B Regulation does not
constrain the use of the personal data or other data that can be gathered through the online
intermediation services (there is no specific prohibition for instance). However, it requires the online
intermediation service provider to be transparent towards its business users regarding the use of
personal data or other data by the intermediation service. The terms and conditions of the provider
must contain a description of technical and contractual access, if any, to personal data and/or other
223
GDPR, Article 17(1)(b) and (c). Also see Politou, E., Alepis, E., & Patsakis,C., 2018, Forgetting personal data and revoking consent under the
GDPR: Challenges and proposed solutions, Journal of cybersecurity, vol.4, p. 12.
224
GDPR, Article 19. This requirement suffers exception if it proves to be impossible or involves disproportionate efforts.
225
GDPR, Article 8.
226
GDPR, Article 22(2).
227
GDPR, Articles 9(2)(a) and 9(2)(e).
228
P2B Regulation, Recital 33.
229
P2B Regulation, Recital 33.
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 60
data provided by users when using the service or generated through the use of the services
230
. That
description shall adequately inform the business user on the following:
(i) whether the provider has access to the data provided by the business user or the consumers
through their use of the intermediation service or generated through the provision of the
intermediation service; regarding such data, the provider must also indicate whether they are
provided to third parties, and if such a provision is not necessary for the functioning of the
intermediation service, the purpose of such data sharing and the possibility for the business
user to opt-out from such a data sharing;
(ii) whether the business user has access to those data; and
(iii) whether the business user has access to data, including in aggregated form, provided by or
generated through the provision of the online intermediation services to all of the business
users and consumers.
The terms and conditions must indicate, when applicable, the categories of the data concerned by such
accesses and the conditions under which such accesses are proposed.
Box 11: Main issues and gaps: Processing of personal data and privacy
•
Although placing tracking devices on terminal equipment and accessing the information
contained on it requires consent under ePrivacy Directive, data controllers remain free to
choose the legal ground under Article 6 GDPR for further processing operations.
• Some tracking mechanisms such as device fingerprinting which may not require storing or
accessing information in the end user equipment are not clearly covered by the rules of user
consent under the ePrivacy Directive. This gap may be closed by the adoption of the new
ePrivacy Regulation.
• Depending on the legal basis used by the data controller, data subjects will usually enjoy an
opt-in-based consent or (less frequently) as a minimum, an opt-out-based right to object when
facing targeted advertising. However, this opt-out right to object is not defined in the legislation
and this right may be difficult to exercise in practice, especially for targeted advertising.
• The GDPR recognises children as vulnerable data subjects and contains rules on child consent
with regard to information society services offered directly towards them
231
. Conversely, neither
the ePrivacy Directive nor the proposed ePrivacy Regulation contain specific rules on the
tracking children’s devices
232
. However, the revised AVMS Directive does protect children from
having their data collected and processed for targeted advertising purposes (see below).
• The P2B Regulation enshrines transparency requirements in the terms and conditions of the
online intermediation services on the sharing of personal and non-personal data to business
users but this requirement does not extend to end users.
Source: Authors’ own elaboration.
230
P2B Regulation, Article 9.
231
GDPR, Article 8.
232
Verdoodt, V. & Feci, N., 2017, Children and targeted advertising in the proposed ePrivacy (Regulation) colouring book: Time to use the
whole box of crayons’. BEUC, 2017, Proposal for a regulation on Privacy and electronic communication (ePrivacy) – BEUC position paper,
BEUC-X-2017-059, 09/06/2017, p. 15.
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
61 PE 662.913
3.2.2. Transparency obligations
c. Specific transparency obligations on ranking
Information duties related to the ranking of products are now contained in the UCPD, the CRD
(through amendments introduced by the Directive on Better enforcement and modernisation of Union
consumer protection rules) and in the P2B Regulation. These rules are relevant since targeted
advertising can in certain settings be seen as a form of ranking.
Ranking is defined in the UCPD as “the relative prominence given to products, as presented, organised
or communicated by the trader, irrespective of the technological means used for such presentation,
organisation or communication”
233
. Through ranking parameters, some elements (for instance
personal data, purchasing history, web surfing patterns, etc.) could be used to propose a personalised
ranking.
Under new Article 7(4a) of the UCPD, the main parameters used for ranking a product and their
importance (compared to other parameters) are considered as material information if the consumers
have the possibility to search products offered by different traders or consumers on the basis of a query
(with a keyword or a phrase for instance). It means that, subject to other requirements of Article 7 of
the UCPD, the violation of this information duty may be regarded as a misleading omission (and an
unfair, and prohibited, commercial practice) under the Directive.
Furthermore, the new Article 6a(1)(a) of the CRD states that online marketplaces are obliged to provide
consumers with additional information, such as the main parameters determining ranking of offers
presented to the consumer as a result of search queries. Where (part of) the offers presented to
consumers after search queries are based on targeted advertising, consumers must be given
information on the criteria used for ranking, and hence on the criteria for targeting the advertisements
themselves.
These online marketplaces are not expected to disclose the detailed functioning of the ranking
mechanism but they should provide a general description of the main parameters they use
234
.
Information needs to be made available in a specific section of the online interface that is directly and easily
accessible from the page where the query results or the offers are presented.
The P2B Regulation also lays down some requirements of transparency regarding the presentation
and ranking of the goods and services offered on a platform or by the search results appearing on a
search engine. Article 2(8) of the P2B Regulation defines ‘ranking’ as the “relative prominence given to
the goods or services offered through online intermediation services, or the relevance given to search
results by online search engines, as presented, organised or communicated by the providers of online
intermediation services or by providers of online search engines, respectively, irrespective of the
technological means used for such presentation, organisation or communication”. Pursuant to that
definition, ranking “can essentially be thought of as a form of data-driven, algorithmic decision-
making”
235
and, as result, could lead to targeted advertising.
233
UCPD, Article 2(m) (as introduced by Directive on Better enforcement and modernisation of Union consumer protection rules).
234
Directive on Better enforcement and modernisation of Union consumer protection rules, Recital 23.
235
European Commission, 2020, Guidelines of 7 December 2020 on ranking transparency pursuant to Regulation (EU) 2019/1150 of the
European Parliament and of the Council, OJ C 424/1 of 8.12.2020, paragraph 11.
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 62
With regard to ranking mechanisms, the P2B Regulation requires providers of online intermediation
services and providers of online search engines to be transparent regarding the ranking of the goods
or services offered on the intermediation platform, or the relevance given to search results on search
engines.
Providers of intermediation services must present, in their terms and conditions, the main parameters
determining ranking and the reasons for the relative importance of those main parameters as opposed
to other parameters
236
.
Through a publicly available description, search engine providers, must indicate the main parameters,
which individually or collectively, are the most significant in determining ranking and the relative
importance of those main parameters
237
. These descriptions must enable professional users of the
intermediation services or of the search engines to adequately understand whether the characteristics
of the goods and services offered through those services or search engines, and their relevance for the
consumers, are taken into account in the ranking mechanism, how and to what extent.
If the provider includes, in its ranking parameters, possibilities to influence ranking against any direct
or indirect remuneration paid by the professional users, the description mentioned above must
integrate those possibilities and the effects of such remuneration on ranking
238
.
Whilst the P2B Regulation does not require the disclosure of the detailed functioning of ranking
mechanisms, including algorithms
239
, it states that “providers should give meaningful explanations of
their ranking mechanisms and, in particular, the main parameters used”
240
. However, providers must
not provide excess information (or lengthy or complicated descriptions) that would render the given
explanation not meaningful in the end
241
. In addition, according to the European Commission, “not
providing excessive details should also help avoid the risk of enabling the deception of consumers or
consumer harm”
242
. As a way to provide such explanations to the professional users of their platforms,
providers should identify the main parameters determining ranking proposed on their platforms. To
that end, providers should accomplish first an assessment of all relevant features of ranking taken into
account for their specific service
243
. Such an assessment must be specific to the service concerned and
the ranking presented on it; for instance, some platforms may use a large number of ranking
parameters and others very few
244
.
The European Commission has listed examples of types of ranking parameters the providers may
use
245
, including the personalisation of the ranking, the consumer search behaviour and intent and the
default settings, sorting and filtering mechanisms chosen by the consumer. The European Commission
considers that the requirement to describe the main parameters used in personalised ranking “could
236
P2B Regulation, Article 5, paragraph 3.
237
P2B Regulation, Article 5, paragraph 2.
238
P2B Regulation, Article 5, paragraph 3.
239
P2B Regulation, Article 5, paragraph 6.
240
European Commission, 2020, Guidelines of 7 December 2020 on ranking transparency pursuant to Regulation (EU) 2019/1150 of the
European Parliament and of the Council, OJ C 424/1 of 8.12.2020, paragraph 17.
241
European Commission, 2020, Guidelines of 7 December 2020 on ranking transparency pursuant to Regulation (EU) 2019/1150 of the
European Parliament and of the Council, OJ C 424/1 of 8.12.2020, paragraph 25.
242
Ibid.
243
European Commission, 2020, Guidelines of 7 December 2020 on ranking transparency pursuant to Regulation (EU) 2019/1150 of the
European Parliament and of the Council, OJ C 424/1 of 8.12.2020, paragraph 38.
244
European Commission, 2020, Guidelines of 7 December 2020 on ranking transparency pursuant to Regulation (EU) 2019/1150 of the
European Parliament and of the Council, OJ C 424/1 of 8.12.2020, paragraph 38 to 40.
245
See European Commission, 2020, Guidelines of 7 December 2020 on ranking transparency pursuant to Regulation (EU) 2019/1150 of the
European Parliament and of the Council, OJ C 424/1 of 8.12.2020, Annex A.
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
63 PE 662.913
require that providers analyse the potentially very long list of factors that are used for this
personalisation, such as consumers’ personal profiles, interests, search behaviour, their actual
geographic location, the time of day the search takes place, their use of cookie blockers or other
technical tools and more generally, the wealth of data held on the specific consumer as well as their
use of default settings”
246
. Consumer search behaviour can also affect the ranking
247
. Finally, the
Commission adds that personalised ranking can also be impacted by “default settings which can be
rearranged, undone, or ‘overridden’ by consumers using sorting or filtering tools”
248
.
The P2B Regulation requires that the provider describes the main parameters that determine
ranking on its platform to the direct benefit of the professional users. However, it does not
require the provider (or the professional user) to also provide such information directly to the
consumer. Therefore, it is not clear how, to what extent and by which player, the consumer should be
informed of such ranking parameters (except for search engines that are required to give a public
access to the ranking information).
Nevertheless, the P2B Regulation must be read together with the new Articles 7(4a) of the UCPD and
6a(1)(a), of the CRD. Those directives render compulsory the provision of the information regarding
ranking parameters directly to the consumer by the online marketplaces. Such an obligation applies
also to online intermediation services referred to in the P2B Regulation, since such services include the
notion of online marketplaces referred to in UCPD and CRD
249
.
With regard to personalised pricing (that may be incorporated in targeted advertising), reference
must be made to the new Article 6(1)(ea) of the CRD. This provision states that, where applicable,
traders must provide consumers with information on the fact that prices are personalised based on
automated decision-making, before consumers are bound by distance contracts. Where targeted
advertising includes price personalisation (e.g. an advertising offering a discount to a consumer, based
on his previous buying history), this article provides consumers with transparency on (part of) the
content of the advertising.
d. General rules
Article 6 of the eCommerce Directive creates various identification and transparency requirements for
commercial communications which are part of, or constitute, an information society service. The
commercial communication must be clearly identifiable as such
250
. The person on whose behalf the
commercial communication is made must be clearly identifiable
251
. Furthermore, the eCommerce
Directive requires promotional offers, such as rebates, and promotional competitions or games, to be
clearly identifiable and the associated conditions easily accessible and presented clearly and
246
European Commission, 2020, Guidelines of 7 December 2020 on ranking transparency pursuant to Regulation (EU) 2019/1150 of the
European Parliament and of the Council, OJ C 424/1 of 8.12.2020, paragraph 51.
247
European Commission, 2020, Guidelines of 7 December 2020 on ranking transparency pursuant to Regulation (EU) 2019/1150 of the
European Parliament and of the Council, OJ C 424/1 of 8.12.2020, paragraph 54.
248
European Commission, 2020, Guidelines of 7 December 2020 on ranking transparency pursuant to Regulation (EU) 2019/1150 of the
European Parliament and of the Council, OJ C 424/1 of 8.12.2020, paragraph 56.
249
European Parliament and the Council of the European Union, 2019, Directive 2019/2161 of the European Parliament and of the Council
of 27 November 2019 amending Council Directive 93/13/EEC and Directives 98/6/EC, 2005/29/EC and 2011/83/EU of the European
Parliament and of the Council as regards the better enforcement and modernisation of Union consumer protection rules, OJ L 328 of
18.12.2019, Recital 21.
250
eCommerce Directive, Article 6(a).
251
eCommerce Directive, Article 6(b).
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 64
unambiguously
252 253
. This Article 6 therefore also applies to targeted advertising but does not
explicitly oblige the platform to inform users that they are receiving targeted advertising.
Several provisions of the CRD can also be relied on to increase consumers’ awareness of (targeted)
advertising. For instance, Article 6 requires traders to provide consumers, before they are bound by any
distance contract, with information on the identity of the traders, the main characteristics of the goods
or services, their total prices, etc. Where these obligations apply, this provision indirectly provides
consumers with transparency on (part of) the content of the advertising.
Furthermore, under the UCPD, any lack of transparency with regard to targeted advertising may also
be seen as a misleading action or omission under the semi-general rules of Articles 6 and 7 or under
the general rule of Article 5(2). These provisions are open norms and the claimant needs to
demonstrate that the requirements are fulfilled in the particular case at hand (which could be difficult,
especially with regard to the relationship between the information and the transactional decision –
potentially – taken by the average consumer). Some differences could also be observed in the case law
of the Member States.
In addition to the general transparency requirement (see above), the GDPR is interpreted as containing
more specific information requirements. First, when the data controller uses profiling techniques, the
Article 29 Working Party considers that the data subject must be informed about this profiling. When
the data subject exercises the right of access, the data subject must be able to obtain details concerning
segments the data subject has been placed into
254
. This transparency requirement may help data
subjects to understand why they are the recipient of some advertising. Second, Article 22 of the GDPR
relates to decision making processes based solely on automated processing. Under GDPR, the sole
decision to display a targeted advertising based on profiling is not, in principle, considered as a fully
automated decision with far reaching effect unless it takes advantage of data subject vulnerabilities for
instance
255
. In such cases, the data controller must provide meaningful information about the logic
involved by the processing
256
. According to the Article 29 Working Party this requirement means, among
others, providing information on criteria used during the decision process and the rationale behind the
processing
257
. In addition, the right of access
258
of the data subject enable him to obtain information on
the parameters used and on their weight on an aggregate level
259
. Hence, when the decision to display
advertising falls under the scope of Article 22, the data subject may obtain information on parameters
used to determine the content of an advertising.
252
eCommerce Directive, Article 6(c) and (d).
253
Article 7 contains additional requirements regarding unsolicited commercial communications sent by electronic mails. The commercial
communication shall be clearly and unambiguously identifiable as such as soon as it is received by the recipient. Nevertheless, the
requirements stated in Article 7 of eCommerce Directive are no longer relevant in most cases since ePrivacy Directive now lays down
stricter requirements as regards unsolicited commercial communications (the only hypothesis where Article 7 would still be relevant is
“the (limited) cases where natural persons would not be protected by Directive 2002/58/EC (e.g. natural persons who are not
subscribers)”; see Communication from the Commission to the European Parliament, the Council, the European Economic and Social
Committee and the Committee of the Regions on unsolicited commercial communications or 'spam', COM/2004/0028 final, paragraph22.
254
Article 29 Data Protection Working Party, 2018, Guidelines on Automated individual decision-making and Profiling for the purposes of
Regulation 2016/679, 3 October 2018, WP251rev.01, pp. 16-17.
255
Article 29 Data Protection Working Party, 2018, Guidelines on Automated individual decision-making and Profiling for the purposes of
Regulation 2016/679, 3 October 2018, WP251rev.01, p. 22.
256
GDPR, Article 13 (2)(f) and Article 14(2)(g).
257
Article 29 Data Protection Working Party, 2018, Guidelines on Automated individual decision-making and Profiling for the purposes of
Regulation 2016/679, 3 October 2018, WP251rev.01, p. 25.
258
GDPR, Article 15.
259
Article 29 Data Protection Working Party, 2018, Guidelines on Automated individual decision-making and Profiling for the purposes of
Regulation 2016/679, 3 October 2018, WP251rev.01, p. 27.
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
65 PE 662.913
Box 12: Main issues and gaps: Transparency obligations
•
Under consumer protection rules, there is no specific obligation to inform consumers in a clear
and comprehensible manner that they are facing targeted advertising, or regarding the
parameters used to determine the recipient or the content of the advertising (rules are currently
limited to ranking and personalised pricing).
• Under consumer protection rules, regarding the information on ranking, discussions may be
expected with regard to the interpretation of “main parameters”. The Guidelines issued by the
European Commission on ranking transparency pursuant the P2B Regulation
260
can serve as a
source of interpretation. However, the use of these guidelines for interpreting the consumer
rules are limited since (i) in terms of scope, the guidelines expressly concern the P2B Regulation
and do not cover the consumer rules; (ii) some differences of interpretation are possible since
the recipients of the information regarding ranking parameters are not the same in the P2B
regulation and in the consumer protection rules (the professional user for the first and the
consumer for the second) and they do not have the same levels of knowledge, which may result
in different levels of information to be provided.
• Under the GDPR, there is no obligation to provide information on the parameters used to display
advertising unless Article 22 applies. Consequently, this is likely to apply exceptionally.
Additionally, this requirement is not mentioned in the GDPR but rather in Article 29 Working
Party guidelines.
• Under the GDPR, obtaining information on the main criteria used for displaying advertising
and/or on details of which segment the data subject is placed when using profiling requires a
specific action of the data subject (i.e. exercising its right of access).
Source: Authors’ own elaboration.
3.2.3. Internal market
To ensure free movement of services and to avoid barriers to trade, the European legislator enacted
several legal instruments where, in line with consistent case-law, restrictions to advertising could be
acceptable only if they are necessary and proportionate.
In the eCommerce Directive, there is a very specific mechanism called the internal market clause
(Article 3). With that clause, instead of trying to harmonise all the areas concerned linked to online
activities
261
, the European legislator opted for a mechanism based on the principles of the country of
origin and mutual trust between Member States.
Pursuant to the clause, information society service providers are subject to the law of the Member State
in which they are established and they are supervised by the competent authority of that Member
State
262
.
260
See European Commission, 2020, Guidelines of 7 December 2020 on ranking transparency pursuant to Regulation (EU) 2019/1150 of the
European Parliament and of the Council, OJ C 424/1 of 8.12.2020.
261
It must be noted also that the Directive (EU) 2015/1535 (European Parliament and the Council of the European Union, 2015, Directive (EU)
2015/1535 of the European Parliament and of the Council of 9 September 2015 laying down a procedure for the provision of information
in the field of technical regulations and of rules on Information Society services, OJ L 241, 17.9.2015), p. 1–1 sets up a procedure whereby
Member States must notify to the European Commission all draft technical regulations on products and information society services
before they are adopted in national law. Since these measures can lead to trade barriers, Member States must ensure that they act in a
transparent manner, by giving the Commission and the other Member States the possibility to examine the drafts and to comment on
these drafts or to issue detailed opinions. The notification triggers a three month standstill period to allow the Member States and/or the
Commission to react
262
eCommerce Directive, Article 3, paragraph 1.
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 66
On the other hand, the other Member States cannot take measures to prevent such service providers
from offering their services on their territory
263
. The clause only applies to the national requirements
that are in the “coordinated field’. The “coordinated field” covers requirements with which the service
provider must comply in respect of “the pursuit of the activity of an information society service, such
as requirements […] regarding the quality or content of the service including those applicable to
advertising […]”
264
.
Consequently, the national requirements regulating advertising are covered by the coordinated field
and are within the scope of the internal market clause.
The CJEU confirmed this analysis in a recent decision of 1
st
October 2020 (C-649/18) regarding national
rules restricting the possibility for online pharmacists to conduct advertising including online
advertising
265
.
By way of derogation, however, Member States may impose national measures on providers offering
services in their jurisdiction provided those measures are: (i) necessary to the protection of public
interest objectives (such as the protection of public order or public health), (ii) taken against an
information society service, which prejudices such objectives or presents a serious and grave risk of
prejudice to those objectives and (iii) proportionate to that objective
266
. In addition, before being
authorised to impose such measures on the service provider, the Member State must ask the Member
State of establishment to comply with its national measures and must notify the European Commission
of its intention to enforce the so-called measure
267
.
If that second condition is not fulfilled (i.e. the measure restricting the freedom to provide an
information society service has not been notified to the Member State of establishment and to the
European Commission), the measure is considered unenforceable against the provider
268
.To analyse if
the first condition of Article 3, paragraph 4(a) is fulfilled as regards necessity and proportionality, the
Member State must take into account “the case-law relating to Articles 49 and 56 TFEU […], in so far as
those conditions largely overlap with the requirements that must be fulfilled by any obstacle to the
fundamental freedoms guaranteed in those articles of the TFEU”
269
. Although the internal market
clause is a powerful mechanism to ensure the free circulation of information society services within the
European Union, the possibility given to Member States to derogate still allows for barriers, under the
condition that they are justified, non-discriminatory and proportionate. It also creates legal uncertainty
for the service providers (and even more for SMEs) since it is complicated for providers which are active
in several national markets, to be able to anticipate if they will be subject to the national measures in
each Members States where they provide their service, or not.
The C-649/18 case provides an interesting illustration of the risk that the derogation mechanism entails.
In that case, four French measures were cited, before the French Courts, to restrict an online pharmacy
established in the Netherlands from targeting among others French consumers and patients
270
.
263
eCommerce Directive, Article 3, paragraph 2.
264
eCommerce Directive, Article 2(h).
265
CJEU, case A contre Daniel B e.a., 1 October 2020, C-649/18, EU:C:2020:764.
266
eCommerce Directive, Article 3, paragraph 4(a).
267
eCommerce Directive, Article 3, paragraph 4(b).
268
CJEU (GC), case Airbnb, 19 December 2019, C-390/18, EU:C:2019:1112, point 96; Ledger, M. 2020, CJEU ruling on legal qualification of
Airbnb services, R.D.T.I., 2020/3, p. 123, paragraph15.
269
CJEU, case A contre Daniel B e.a., 1
October 2020, C-649/18, point 64.
270
CJEU case A contre Daniel B e.a., 1 October 2020, C-649/18.
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
67 PE 662.913
Among those measures, three aimed to restrict the possibility for pharmacists to promote their online
services
271
. The CJEU had to analyse, for each measure, whether it could fulfil the condition laid down
in Article 3, paragraph 4a, or not. For each measure, the CJEU carried out a case-by-case analysis.
This clearly deprives market players of legal certainty
272
. The eCommerce Directive also aims “to
remove the barriers to the development of cross-border services […] which members of the regulated
professions might offer on the Internet”
273
. In order to reach that objective, Article 8 creates a positive
obligation for the Member States to authorise regulated professions to use commercial
communications which are part of, or constitute, an information society service. However, commercial
communications need to comply with the professional rules regarding, in particular the independence,
dignity and honour of the profession, professional secrecy and fairness towards clients and other
members of the profession
274
. On its decision Vanderborght
275
, the CJEU stated that a prohibition of
advertising for a certain activity is liable to restrict the possibility, for the persons carrying on an activity,
of making themselves known to their potential clientele and of promoting the services which they
propose to offer to that clientele. Such a prohibition therefore constitutes a restriction on the freedom
to provide services. The CJEU concluded that the professional rules can never result in a total prohibition
of the use of online advertising by the member of a regulated profession. Therefore, the Member States
must ensure that no total prohibition of the use of online advertising by the member of a regulated
profession remains or is adopted.
Under Article 24, paragraph 1, of the Services Directive, the same principle applies regardless of
whether the commercial communication is made online or not. It provides that Member States must
remove total prohibitions on commercial communications by the regulated professions. However, the
second paragraph of Article 24 goes on to recall that such commercial communications must
nevertheless comply with the professional rules that apply to each regulated profession which relate,
in particular, to the independence, dignity and integrity of the profession, as well as to professional
secrecy. Article 24, paragraph 2 explains how to achieve the balance between the right to make
commercial communications and the need to comply with these professional rules. It specifies that
professional rules that would restrict the use of commercial communications by the regulated
professions, are admitted provided that they are non-discriminatory, justified by an overriding reason
relating to the public interest and proportionate.
In addition, the Proportionality Test Directive was adopted not to remove the existing restrictive
professional rules, but to prevent new ones from being adopted
276
.
271
The first forbidded the use of paid referencing services, the other restricted the possibility of proposing price reductions and the last
prohibited the pharmacist from sending advertising through paper flyers.
272
On that point, in the field of the online sale of medicinal products, see Bourguignon, C., 2020, La vente en ligne transfrontalière de
médicaments au sein de l’Union européenne, D.C.C.R., 2020/2, n° 127, p. 127.
273
eCommerce Directive, Recital 32.
274
eCommerce Directive, Article 8(1).
275
CJEU, case Procédure pénale contre Luc Vanderborght, 4 May 2017, C-339/15, EU:C:2017:335, point 44; Regarding that decision: Ghislain, S.,
2018, ‘Publicité et professions libérales : questions actuelles au regard de l’arrêt Vanderborght du 4 mai 2017 de la Cour de Justice’,
D.C.C.R., 2018/3-4, nos 120-121, p. 19-43; Bourguignon, C., 2019, ‘Les réglementations des professionnels de la santé revues à l’aune des
libertés de circulation économiques : l’exemple de la publicité à l’ère du numérique’, R.D.T.I., 2019/1, n° 74, p. 25-54.
276
It completes the requirement imposed on Member States in Article 59, paragraph 3, of the Directive 2005/36/CE (European Parliament
and the Council of the European Union, 2005, Directive 2005/36/CE of the European Parliament and of the Council of 7 September 2005
on the recognition of professional qualifications, OJ L 255 of 30.9.2005). Under that Article, the Member States shall examine “whether
requirements under their legal system restricting the access to a profession or its pursuit to the holders of a specific professional
qualification, including the use of professional titles and the professional activities allowed under such title, referred to in this Article as
‘requirements’ are compatible with the following principles: (a) requirements must be neither directly nor indirectly discriminatory on
the basis of nationality or residence; (b) requirements must be justified by overriding reasons of general interest; (c) requirements must
be suitable for securing the attainment of the objective pursued and must not go beyond what is necessary to attain that objective”. The
Members States had to communicate to the European Commission the results of such an examination by 18 January 2016.
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 68
Pursuant to that Directive, before introducing new (or amending existing) legislative, regulatory or
administrative provisions restricting access to, or the pursuit of, regulated professions, the Member
States are required to undertake an assessment of proportionality in accordance with the rules laid
down in the Directive. Those rules are the following: the new provision must be neither directly nor
indirectly discriminatory on the basis of nationality or residence
277
; it must be justified on grounds of
public interest objectives
278
; it must be proportionate, i.e. “suitable for securing the attainment of the
objective pursued and not go beyond what is necessary to attain that objective”
279
. These assessments
must be notified by the Member State to the European Commission which makes it publicly
available
280
. Under Article 7 of that Directive, the compliance of national restrictions to advertising with
the principle of proportionality should be examined by the Member State before being adopted and
such an assessment shall be notified to the European Commission
281
. The Directive had to be
transposed into national law by 30 July 2020.
Box 13: Main issues and gaps: Internal market principles
•
The internal market clause of the eCommerce Directive clearly applies to online advertising. The
host Member State could apply its national restrictions only if they are non-discriminatory,
justified and proportionate. This may lead to legal uncertainty for market players especially for
SMEs.
Source: Authors’ own elaboration.
3.2.4. Prohibition of certain practices
The applicable legal framework may prohibit some practices currently carried out in the context of
targeted advertising. Such prohibitions must therefore be taken into account while assessing the
compliance of any advertising with the legal framework.
The revised AVMS Directive introduces qualitative rules for advertising that need to be complied with
by video sharing platforms for the commercial communications that are marketed, sold or arranged by
them. These rules
282
mirror exactly those that need to be complied with by audiovisual media service
providers (linear and non-linear) and will have an impact on users in particular because platforms will
need to make sure that their advertising is recognisable as such and does not discriminate on the basis
of sex, racial or ethnic origin, nationality, religion or belief, disability, age or sexual orientation. Also,
their advertising cannot cause any detriment to minors. Targeted advertising campaigns will therefore
need to take these factors into consideration and users will benefit from these protections. The video
sharing platforms in scope will need to take “appropriate measures” to make sure that advertising not
arranged by them also complies with these rules.
Such requirements were inserted in 2013 with the Directive 2013/55/EU (European Parliament and the Council of the European Union,
2013, Directive 2013/55/EU of the European Parliament and of the Council of 20 November 2013 amending Directive 2005/36/EC on the
recognition of professional qualifications and Regulation (EU) No 1024/2012 on administrative cooperation through the Internal Market
Information System (‘the IMI Regulation’), OJ L 354 of 28.12.2013).
277
Proportionality Test Directive, Article 5.
278
Proportionality Test Directive, Article 6.
279
Proportionality Test Directive, Article 7.
280
Proportionality Test Directive, Article 11.
281
On that point: Bourguignon, C., op. cit., p. 50, paragraph 37.
282
Article 9(1) of AVMS Directive.
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
69 PE 662.913
Also of interest, the revised AVMS Directive specifies that that data collected by video sharing platforms
to protect children may not be processed for commercial purposes, such as for direct marketing,
profiling and behaviourally targeted advertising
283
. This rule mirrors the one which is already in
place for (linear and non-linear) audiovisual media service providers and which provides that they
cannot process the personal data of minors they collect (or otherwise process) for commercial
purposes, such as direct marketing, profiling and behaviourally targeted advertising
284
.
The non-discrimination principle is also enshrined in different EU legislative instruments. These
instruments are the “Race Directive”
285
, the “Employment discrimination Directive”
286
, the “Gender
equality Directive”
287
, the “Gender Directive”
288
and the “Geo-blocking Regulation”
289
.
These instruments prohibit direct discrimination (i.e. situations where a person is treated less
favourably than another is, has been or would be treated in a comparable situation on the basis of a
protected ground) and indirect discriminations, unless objectively justified (i.e. situations where an
apparently neutral provision, criterion or practice would put persons characterised by a protected
ground at a particular disadvantage), on the basis of different protected grounds. These instruments
also have different scopes of application. On the basis of these instruments targeted advertising may
be prohibited if the targeting is directly based on areas under protection or if it is based on neutral
criteria which nonetheless have discriminatory effects on a group of people subject to protections.
The Race Directive applies, among others, within the field of employment, social security, education
and access to goods and services (including housing) and prohibits direct and indirect discrimination
based on racial or ethnic origin. The Employment discrimination Directive applies only within the
context of employment and prohibits direct and indirect discrimination based on the grounds of
religion or belief, disability, age and sexual orientation. The Gender equality Directive applies in the
context of social security and employment and prohibits direct and indirect discrimination based on
sex. Within the field of access to and supply of goods and services, advertising and media content is
explicitly excluded from the scope of the Gender Directive.
Finally, the Geo-blocking Regulation prohibits direct and indirect discrimination based on nationality
and place of residence in the internal market. In particular, traders cannot apply different general
conditions of access to their goods and services on the basis of these criteria. This Regulation does
however allow traders to apply different general conditions of access between Member States or
specific customers groups on a non-discriminatory basis
290
.
283
Article 28b (3) of AVMS Directive.
284
Article 6a of AVMS Directive.
285
Council of the European Union, 2000, Council Directive 2000/43/EC of 29 June 2000 implementing the principle of equal treatment
between persons irrespective of racial or ethnic origin, OJ L 180 of 19.7.2000.
286
Council of the European Union, 2000, Council Directive 2000/78/EC of 27 November 2000 establishing a general framework for equal
treatment in employment and occupation, OJ L 303 of 02.12.2000.
287
European Parliament and the Council of the European Union, 2006, Directive 2006/54/EC of the European Parliament and of the Council
of 5 July 2006 on the implementation of the principle of equal opportunities and equal treatment of men and women in matters of
employment and occupation (recast), OJ L 204 of 26.7.2006.
288
Council of the European Union, 2004, Council Directive 2004/113/EC of 13 December 2004 implementing the principle of equal treatment
between men and women in the access to and supply of goods and services, OJ L 373 of 21.12.2004.
289
European Parliament and the Council of the European Union, 2018, Regulation (EU) 2018/302 of the European Parliament and of the
Council of 28 February 2018 on addressing unjustified geo-blocking and other forms of discrimination based on customers' nationality,
place of residence or place of establishment within the internal market and amending Regulations (EC) No 2006/2004 and (EU) 2017/2394
and Directive 2009/22/EC, OJ L 60I of 02.03.2018.
290
Geo-Blocking Regulation, Article 4 (2).
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 70
The UCPD may also be relevant on a stand-alone basis or in combination with one of the legal
instruments mentioned above. Indeed, three kinds of commercial practices are considered to be unfair
under the UCPD and, therefore, prohibited
291
: (i) misleading or aggressive practices listed in Annex I
that must in all circumstances be regarded as unfair; (ii) misleading or aggressive practices prohibited
under semi-general clauses of Articles 6 to 9; and (iii) practices prohibited under the general clause of
Article 5(2). Various practices related to advertising and, in particular, targeted advertising may
therefore fall under the scope of the UCPD and should be assessed under this three step test. Once
again, it must be stressed that the practice of targeted advertising is not prohibited as such. A new
practice prohibited in all circumstances, relevant for advertising (and, targeted advertising), is however
added by the Directive on Better enforcement and modernisation of Union consumer protection rules
(cf. new point 11a of the Annex): consumers should be informed of “any paid advertisement or
payment specifically for achieving higher ranking of products within the search results”.
Furthermore, with regard to minors or children, there is no general prohibition of advertising (nor
targeted advertising). One exception to this is that there is a specific aggressive practice which is
prohibited in all circumstances and listed in Annex I: “including in an advertisement a direct exhortation
to children to buy advertised products or persuade their parents or other adults to buy advertised
products for them”. It is however very specific and, as a result, not very useful with regard to current
market practices. Reference must also be made to the average consumer test, applicable to commercial
practices aimed at specific groups of vulnerable consumers (such as children). In accordance with
Article 5(3) of the UCPD, the assessment must be carried out by the competent authorities from the
perspective of the average member of that group. The lack of specific rules protecting minors or
children under the UCPD may also be an issue within the Member States: since the UCPD is a full
harmonisation directive, Member States are not allowed to prohibit commercial practices addressed to
minors or children (as otherwise, they would not be compliant with the directive).
The European Commission also presented on 21 April 2021 its proposal to regulate the use of artificial
intelligence (AI) in Europe
292
. Once adopted, this new Regulation could also contain rules which will
affect online advertising to the extent that advertising is based on AI in scope. In particular, Article 5 of
the proposal foresees that the following practices should be prohibited:
• the placing on the market, putting into service or use of an AI system that deploys subliminal
techniques beyond a person’s consciousness in order to materially distort a person’s behaviour
in a manner that causes or is likely to cause that person or another person physical or
psychological harm;
• the placing on the market, putting into service or use of an AI system that exploits any of the
vulnerabilities of a specific group of persons due to their age, physical or mental disability, in
order to materially distort the behaviour of a person pertaining to that group in a manner that
causes or is likely to cause that person or another person physical or psychological harm.
291
UCPD, Article 5.
292
European Commission, 2021, Proposal for a regulation of the European Parliament and of the Council laying down harmonised rules on
artificial intelligence (artificial intelligence act) and amending certain Union legislative acts, 21.4.2021 COM(2021) 206 final 2021/0106
(COD).
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
71 PE 662.913
Box 14: Main issues and gaps: Prohibition of some practices
•
Under the UCPD, there are no general rules prohibiting some marketing or advertising practices
targeting or directed to minors (or children).
• Under the AVMS, the new rules only apply to video sharing platforms in scope and therefore
platforms that do not have as their principle or essential functionality the provision of
programmes, user-generated video or both to inform, entertain or educate the general public
will not be covered. The video sharing platforms in scope are prevented from using data
collected to protect children from processing it for commercial purposes, such as for direct
marketing, profiling and behaviourally targeted advertising.
• The scopes of non-discrimination Directives are different and advertisements with direct or
indirect discriminatory effects may be prohibited depending on the area. If applicable, the AVMS
Directive however requires that audiovisual commercial communications offered directly by
video sharing platforms shall not include or promote discrimination based on sex, racial or
ethnic origin, nationality, religion or belief, disability, age or sexual orientation
293
.
Source: Authors’ own elaboration.
3.2.5. Many players involved and key role played by the platforms
Since the modification of the CRD in 2019 by the Better Enforcement Directive, some platforms (i.e.
online marketplaces) are also subject to information obligations under consumer protection law. Under
this body of law, the obligations of platforms remain limited, as they mainly need to provide
information on the main parameters used for ranking of offers presented to consumers after a search
query (new Article 6a(1)(a) of the CRD).
As highlighted in the EDBP guidelines, targeting Internet users in general may involve many actors such
as targeters
294
, social media platforms, data brokers, data management providers, data analytics
companies…
295
. The main actors under GDPR are data controllers (i.e. entities defining alone or jointly
with other the purpose and the means of the processing) and the data processor (i.e. entities
processing data on behalf of a controller)
296
. Furthermore, data processing involves “joint controllers”
which requires a case by case analysis of the processing and the roles of the involved entities
297
.
Qualification of (joint) controller/processor is important to determine the stakeholder’s duties under
the GDPR. The data controller bears the majority of the obligations enshrined in the GDPR. The data
controller is for instance in charge of ensuring transparency requirements and must enable the data
subjects to exercise their rights. When processing is partly/wholly subject to joint control, Article 26
GDPR requires that the joint controllers determine their respective responsibilities for compliance with
the obligations under the GDPR (including transparency requirement and exercise of the data subject
rights) by means of an arrangement.
293
Revised AVMS Directive, Article 9(1)(c)(ii).
294
The EDPB uses the term targeter to designate natural or legal persons that use social media services in order to direct specific messages
at a set of social media users on the basis of specific parameters or criteria. See following footnote.
295
European Data Protection Board, 2020, Guidelines 8/2020 on the targeting of social media users, 2 September 2020, p. 9.
296
GDPR, Article 4(7) and Article (8).
297
European Data protection board, 2020, Guidelines 07/2020 on the concepts of controller and processor in the GDPR, 02 September 2020.
p. 22.
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 72
Finally, when a data controller/processor relationship exists, it must be governed by a contract
298
. This
requirement ensures, among others, respect of data security requirements and control of the data
controller when sub-processors are involved.
When many actors intervene at different stages of the processing, joint control may exist only with
regard to part of the processing operations necessary during the process of displaying targeted
content. The CJEU has, for instance considered that joint control exists between a website operator and
a social network for the personal data collection and transmission through use of a social media plugin
for advertising purpose
299
. A joint controller relationship exists however only for processing for which
both entities effectively determine the purposes and the means
300
. In its guidelines on targeting on
social media, the EDPB has stated that a joint control exists between the targeter and social media
platforms for some processing operations related to targeting of social media users, among others on
the basis of (i) data provided by the data subject to social media by means of criterions selected by the
targeter, (ii) observed personal data by means of tracking pixels or GPS location, (iii) inferred personal
data by means of interests on social media
301
. Additionally, when targeting is done through displaying
advertising on websites, the Article 29 Working Party has stated that joint control may exists between
the publisher and the ad network provider
302
.
Box 15: Main issues and gaps: Many players involved and key role played by the platforms
•
The qualification of data (joint) controllers/data processor can be complex in an ecosystem
where many actors may intervene at different stages of the processing.
• Targeters, publishers and ad network providers may not necessarily be aware of their joint
controller status under the GDPR when using web services in order to promote their goods and
services by means of targeted advertising. They may however be liable for any damage resulting
from an infringement to the Regulation
303
.
• In the absence of adequate qualification, joint controllers may not conclude arrangements
framing their respective responsibilities. Consequently, the essence of this arrangement cannot
be made public
304
which can impair the ability of the data subject to be informed and exercise
their rights.
Source: Authors’ own elaboration.
298
GDPR, Article 28.
299
CJEU, judgement Fashion ID ,29 July 2019, C-40/17, EU:C:2019:629, point 78 and 80.
300
CJEU, judgement Fashion ID ,29 July 2019, C-40/17, EU:C:2019:629, point 85.
301
European Data Protection Board, 2020, Guidelines 8/2020 on the targeting of social media users, 2 September 2020, pp. 13-14, 17-18 and
20.
302
Article 29 Data Protection Working Party, 2010, Opinion 2/2010 on online behavioural advertising, 22 June 2010, WP171, p. 11.
303
GDPR, Article 82.
304
See, for this requirement, GDPR, Article 26(2).
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
73 PE 662.913
3.2.6. Summary
Table 4: Summary table on gaps
Processing
of personal
data
Transparency
specific rules
Transparency
general rules
Opt-in/opt-out
Prohibition
of practices
Specific
P2B Regulation
v
v
ePrivacy (incl.
Proposal)
v
v
v (for tracking
devices and direct
marketing
communication)
AVMS
v
v
General
UCPD
v
v
v
Misleading and
Comparative
Advertising DIR
v
CRD
v
v
GDPR
v
v
v
v (depending on
the legal ground
of the processing)
Source: Authors’ own elaboration.
3.3. How far do existing enforcement measures address the problems
and what are the gaps?
3.3.1. Overview of the main enforcement measures
Existing legislation applicable to targeted advertising mostly relies on Member States to ensure the
enforcement of substantial provisions. This is the case in particular for the eCommerce Directive, the
P2B Regulation, the ePrivacy Directive, the UCPD, the CRD, the GDPR, and the AVMS Directive.
The eCommerce Directive
305
, the P2B Regulation
306
, the UPCD
307
, the Directive on misleading and
comparative advertising
308
, and the CRD
309
merely provide that Member States shall ensure the
enforcement of substantial provisions by using effective, proportionate and dissuasive measures
and/or sanctions. Depending on the Member State, there may be criminal, civil and/or administrative
sanctions. Since the penalties are not usually harmonised at the EU level, there may be differences
between the Member States. It must however be stressed that, following the Better enforcement
Directive companies that breach UPCD and the CRD in three or more EU countries will face maximum
fines of at least 4% of their annual turnover in the concerned countries or at least EUR 2m, when
information about the trader’s annual turnover is unavailable.
305
See notably Article 20, eCommerce Directive.
306
See notably Article 15(1) and (2), P2B Regulation.
307
See notably Article 13, UPCD. In the Better Enforcement Directive, some further guidance on enforcement was provided (e.g. on criteria
to impose penalties).
308
See notably Article 5, Directive on misleading and comparative advertising.
309
See notably Article 24, CRD. In the Better Enforcement Directive, some further guidance on enforcement was provided (e.g. on criteria to
impose penalties).
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 74
The DSA proposal
310
, the ePrivacy Directive (and proposed Regulation)
311
, the GDPR
312
, and the AVMS
Directive
313
contain more detailed rules on enforcement measures, as they require Member States to
grant national competent authorities with powers to investigate, and enforce the substantive rules
provided for in the texts (notably through fines and other sanctions).
In relation to the DMA proposal, enforcement would however be ensured by the Commission, which
would be granted the powers to adopt implementing acts, investigate cases, impose interim measures
and fines, etc.
314
.
Some additional enforcement mechanisms are in place at EU level. The Directive on representative
actions for the protection of the collective interests of consumers was adopted on 25 November 2020
and needs to be implemented by 25 June 2023
315
. It puts in place a procedural mechanism to allow so-
called qualified entities to bring representative actions before national courts or administrative
authorities on behalf of consumers where the infringer harms the collective interest of consumers. The
collective interest of a consumer refers to the general interest of consumers and, in particular for the
purposes of redress measures, the interests of a group of consumers.
Out of court dispute settlement is also very much encouraged and the Directive on Consumer ADR
316
and a Regulation on consumer ODR
317
aim to ensure that disputes between consumers and traders can
be submitted to entities offering impartial, transparent, effective and fair alternative dispute resolution
procedures. The consumer protection cooperation regulation
318
which applies when traders and
consumers are not established in the same country, may also be relevant, as it involves the
establishment of a cooperation network which may be useful in addressing cross-border
infringements.
3.3.2. Main issues and gaps
A key issue is that most consumers will not seek redress before competent courts when targeted
advertising infringes their rights as consumers, citizens or data subjects. Among the reasons for not
seeking judicial redress, are the complexity of the legal framework, the lack of knowledge about their
rights or the identity of the person responsible for the advertising, the low value of the case, the length
and the cost of the procedure, the symbolic outcome that may be expected (injunction deciding the
stopping of an advertising campaign, already finished many months before the judgement…), the lack
of evidence, etc.
310
See notably Articles 38 and 41, proposed DSA.
311
See notably Article 15, ePrivacy Directive.
312
See notably Article 58, GDPR.
313
See notably Article 28b, AVMS Directive.
314
See notably Articles 18, 26 and 37 of the proposed DMA.
315
European Parliament and the Council of the European Union, 2020, Directive (EU) 2020/1828 of 25 November 2020 on representative
actions for the protection of the collective interests of consumers and repealing Directive 2009/22/EC, OJ L 409 of 4.12.2020.
316
European Parliament and the Council of the European Union, 2013, Directive 2013/11/EU of the European Parliament and of the Council
of 21 May 2013 on alternative dispute resolution for consumer disputes and amending Regulation (EC) No 2006/2004 and Directive
2009/22/EC (Directive on consumer ADR), OJ L 165 of 18.6.2013.
317
European Parliament and the Council of the European Union, 2013, Regulation (EU) No 524/2013 of the European Parliament and of the
Council of 21 May 2013 on online dispute resolution for consumer disputes and amending Regulation (EC) No 2006/2004 and Directive
2009/22/EC (Regulation on consumer ODR), OJ L 165 of 18.6.2013.
318
European Parliament and the Council of the European Union, 2017, Regulation (EU) 2017/2394 of the European Parliament and of the
Council of 12 December 2017 on cooperation between national authorities responsible for the enforcement of consumer protection laws
and repealing Regulation (EC) No 2006/2004, OJ L 345 of 27.12.2017.
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
75 PE 662.913
In most cases, when court decisions are handed down in the context of advertising, the claim is not
introduced by a consumer but by another trader, because of unfair market practices (and relying on
the violation of consumer protection rights).
Some of these gaps are already taken into account – partially at least – by the current legal framework,
with the Injunction Directive, the new Representative Action Directive or the Directive on Consumer
ADR. This is however not specific to targeted advertising and a continuous assessment needs to be
made to ensure that access to justice is improved. The existence of specific enforcement bodies (such
as those under the GDPR, ePrivacy Directive or AVMS) could also help fill some gaps, provided they
have sufficient funding and investigation and enforcement powers. A particular challenge will certainly
be the coordination between these authorities when cases fall under the scope of several legal texts
(and, therefore, several enforcement bodies).
3.3.3. Summary on enforcement
Table 5: Summary table on enforcement
Specific
Enforcement
Body (in
addition to
traditional
courts)
MS shall
take
enforcement
means
Criteria
for
penalties
Possible
fines
calculated
on
turnover
Burden of
proof (and
risks) on the
provider
Civil
sanctions
Self-
regulatory
bodies
Specific
eCommerce
DIR
v
v
P2B
Regulation
v
v
ePrivacy (incl.
Proposal)
v
v
v
v
(proposal)
General
UCPD
v
v
v
v
Misleading
and
Comparative
Advertising
DIR
v
v
v
CRD
v
v
v
v
GDPR
v
v
v
v
AVMS DIR
v
v
v
Source: Authors’ own elaboration.
3.4. What are the new measures proposed by the Commission in DSA
and DMA proposals?
3.4.1. DSA Proposal
The rules on advertising contained in the proposed DSA are aimed at providing more transparency
surrounding advertising which in turn may help users to oppose targeted advertising by refusing to
be profiled on data protection grounds. This transparency also aims to enable scrutiny by authorities
and vetted researchers as to how advertisements are displayed and how they are targeted.
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 76
All online platforms (i.e. hosting service providers that communicate content to the public) would be
required
319
to ensure that the recipients of the service receive individualised information to enable
them to identify, for each specific advertisement displayed to each individual recipient, in a clear and
unambiguous manner and in real time that the information displayed is an advertisement; the (natural
or legal) person on whose behalf the advertisement is displayed; and meaningful information about
the main parameters used to determine the recipient to whom the advertisement is displayed (for
targeted advertising).
Although these transparency obligations are already present in EU leglislation, the current rules are
spread across different legal instruments and do not apply to all platforms. This is therefore a welcome
new element. However it may be useful to clarify the links between the DSA and the GDPR and the
ePrivacy Directive (and future Regulation), in particular whether and how data subjects need to provide
consent before being exposed to behavioural advertising. At the very least also, since data subjects are
always allowed to withdraw their consent, the DSA could also specify that the information should also
include information on how users can object to the further processing of their data for advertising
purposes. There are also no rules in the DSA on the collection and use of chidren’s personal data for
behavioural/targeted advertising purposes whereas a prohibition is included for video sharing
platforms in the recently revised AVMS Directive. For sake of coherence, this rule should also be
included in the future DSA. Also in terms of shortcomings, the DSA does not contain any rule on non-
discrimination whereas such rules are contained in various EU legal instruments, creating a patchwork
of rules, applying to different situations.
The very large online platforms (VLOPs) would also need to compile and make publicly available
through application programming interfaces (APIs) a repository containing certain information, while
also making sure that the repository does not contain any personal data. In particular, the repository
would have to specify whether the advertisement was intended to be displayed specifically to one or
more particular groups of recipients of the service and if so, the main parameters used; the total
number of recipients reached and, where applicable, aggregate numbers for the group or groups of
recipients to whom the advertisement was targeted specifically
320
. The Commission would have to
support and promote the development of voluntary industry standards to ensure the interoperability
of these repositories
321
.
This also addresses some of the concerns highlighted above and is particularly interesting because this
data will need to be made publicly available, meaning that end-users and business users (on top of
researchers and enforcers) will have access to this data. It must however be noted that only the very
large online platforms will be subject to these requirements. Of course, the ‘readilbillity’ of the data
may not be obvious to the layman, but the future European Board for Digital Services could play a role
in making this information more accessible to users.
The Commission would have to encourage and facilitate the development of codes of conduct at EU
level to support and complement the transparency obligations relating to advertisements with
transparency obligations that would go beyond the imposed rules
322
.
Although not specific to advertising, it is also interesting to recall that the rules on the liability of hosting
service providers would remain broadly the same as under the eCommerce Directive. For the purpose
of this study, we note the new addition to the rules on the liability of hosting service providers.
319
Article 24 of the proposed DSA.
320
Article 30 of the proposed DSA.
321
Article 34 of the proposed DSA.
322
Article 36 of the proposed DSA.
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
77 PE 662.913
Article 5.3 of the proposed DSA states that the exemption of Article 5.1 would not apply “with respect
to liability under consumer protection law of online platforms allowing consumers to conclude
distance contracts with traders, where such an online platform presents the specific item of information
or otherwise enables the specific transaction at issue in a way that would lead an average and
reasonably well-informed consumer to believe that the information, or the product or service that is
the object of the transaction, is provided either by the online platform itself or by a recipient of the
service who is acting under its authority or control”. Although this paragraph does not directly mention
advertising, it would not be unreasonable to consider to what extent it may be used in cases where the
platform leads the consumer to believe that the advertisement is arranged by the platform direcly and
not by a third party intermediary.
The proposal also introduces mechanisms for users to flag illegal content in Article 14, rules on internal
complaints handling in Article 17 and rules on out-of-court dispute settlment in Article 18. However, it
is not clear how these rules will apply in case of advertising that is in breach of EU or national legislation.
3.4.2. DMA Proposal
The DMA proposal applies to the gatekeeper of advertising services which also provides one of the
following digital services, the so-called Core Platforms Services: B2C intermediation services (including
marketplaces and app stores), search engines, social networks, video-sharing platforms, number-
independent interpersonal communication services, cloud computing or operating systems
323
.
A gatekeeper is defined on the basis of a cumulative three criteria test, namely: (i) significant impact on
the EU internal market; (ii) control of an important gateway for business users to reach end-users; and
(iii) entrenched and durable position
324
. The DMA proposal introduces a rebuttable presumption, for
each of the three criteria, in the form of a threshold. The first criterion (significant impact) will be
deemed fulfilled if the Core Platforms Services provider achieved an annual EEA turnover equal to or
above 6.5 billion EUR or a market capitalisation of at least 65 billion EUR and that Core Platforms
Services provider is currently active in at least three Member States. The second criterion (important
gateway) will be deemed met if the Core Platforms Services provider reached more than 45 million
monthly active end-users in the EU (around 10% of the EU population) as well as more than 10,000
active business users on an annualised basis. As for the third criterion (entrenched and durable
position), it will be fulfilled if the turnover and user thresholds set above are met for the last 3 financial
years
325
.
A digital platform which has been designated as gatekeeper for one or several Core Platforms Services
is subject to several obligations and prohibitions amongst which some are related to online advertising.
Two obligations aim to complement competition law and to improve the functioning of the Adtech
value chain:
• Provide advertisers and publishers with information concerning the price paid by the advertiser
and publisher and remuneration paid to the publisher
326
; and
323
DMA proposal, Article 2(2h).
324
DMA proposal, Article 3(1).
325
DMA proposal, Article 3(2).
326
DMA proposal, Article 5 (Article 5.g). Such lack of transparency is currently investigated in the Google AdTech case, Cases AT. 40 660 and
40 670.
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 78
• Provide advertisers and publishers, free of charge, access to the performance measuring tools
of the gatekeeper and the information necessary to carry out their own independent
verification of the ad inventory
327
.
One obligation aims to allow more privacy competition between substitute digital services and, in turn,
to prevent deep consumer profiling from becoming the industry standard:
• Submit to the Commission an independently audited description of consumer profiling
techniques used
328
.
327
DMA proposal, Article 6(1)(g).
328
DMA proposal, Article 13 and recital 61.
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
79 PE 662.913
WHAT SOLUTIONS HAVE BEEN APPLIED?
4.1. Comparison of national context
The main issues emerging in each country case study are quite homogenous. All country systems
generally tackle trademarks and copyright protection, market manipulation, unfair competition, youth
protection, consumer protection, data protection and protection of personal rights. These can
therefore be seen as global challenges that are the result of misleading and/or aggressive practices. In
all case studies, consumer protection laws apply equally to online advertising and are therefore the
main elements of consumer protection legislation, complemented in all cases by specific laws relating
to the online space, data, markets or vulnerable consumer groups. This is evidence that, first and
foremost, all countries see consumer protection as the key issue when it comes to regulating
online advertising. The major differences are in how each country tries to address these challenges,
with a similar set-up across the EU (and former EU) countries and some highly relevant practices in the
non-EU case studies.
As a consequence of the focus on consumer protection, the key legislation, outlined in Table 6, aims to
prohibit advertising that unduly constrains consumers’ freedom of choice. In addition to the legislation
in this table, the GDPR is the regulatory response to data protection concerns in the EU case studies
and the United Kingdom, where EU legislation still applies following the UK’s departure from the EU.
Unlike the EU, the US does not have a separate data protection authority or a single comprehensive
privacy law at the federal level, like the GDPR. However, the Federal Trade Commission (FTC), which
implements the relevant United States (US) legislation, has Section 5 which deals with unfair or
deceptive acts or practices in or affecting commerce, including data protection. Section 5 also is
complemented by privacy-related statutes to protect consumer privacy, including for example the
Children’s Online Privacy Protection statute. This approach makes up a mosaic of regulation that
mirrors many issues addressed by the GDPR
329
.
Table 6: Main piece of legislation in case study countries
Country Key legislation Year Revisions
Australia Competition and Consumer Act 2010 N/A
France Consumer Code and law on trust in digital
economy
1993 and
2004
2017
Germany Law against unfair competition 2004 2008, 2015
Ireland Consumer Protection Act (CPA) and the
Electronic Communications Networks and
Services, Privacy and Electronic
Communications Regulations
2007 and
2011
respectively
N/A
United Kingdom Unfair Trading Regulations 2008 2014, 2015
United States The Federal Trade Commission Act 1914 Numerous
(latest 2018)
Source: Authors’ own elaboration.
329
Federal Trade Commission, Children's Online Privacy Protection Rule ("COPPA"), Children's Online Privacy Protection Act of 1998, 15 U.S.C.
6501–6505.
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 80
The review of the legal framework of the case study countries shows that the issues faced are similar.
Perhaps the only exception is in Ireland, which has a far higher volume of claims passing to the
Competition and Consumer Protection Commission (CCPC), Ireland’s statutory body implementing the
main legislation applying to consumer protection. This has to do with the fact that a large number of
online platforms are based in Ireland. The Irish law under which most claims have been made is the
2011 Electronic Communications Networks and Services, Privacy and Electronic Communications
Regulations, outlined in the above table and expanded on in the annex case study
330
.
What can be seen when looking at the overall value of online advertising spending in the case study
countries is that the United States accounts for more than all the other case study countries combined,
and indeed is the largest market globally for online advertising spending. The EU country with the
largest spending in terms of total value is Germany. The picture when broken down per capita is more
nuanced. The United States is still the largest spender. However, Australia, despite spending 93% less
overall than the United States, appears second, with USD 396
331
per capita in 2020. In the EU, Ireland is
the highest by far in terms of per capita spending, partly reflecting the fact that this is where the
platforms are based. In terms of the breakdown by activity, Ireland shows the largest percentage of
total spending on social media spend in the EU, Germany in display advertising and France in search
spending. However, all case studies show quite similar percentages in this respect, roughly half of the
spend going on search, then 20-30% going on display and social media, respectively.
Table 7: Breakdown of online advertising spending in 2020 (USD)
Country
Search
value
Display value
Social media
advertising
value
Total value
Per capita
spending
Australia 4.5 bn (44%) 3.7 bn (36%) 2.1 bn (20%) 10.3 bn 396
France 3.1 bn (54%) 1.1 bn (20%) 1.5 bn (26%) 5.7 bn 85
Germany 4.7 bn (52%) 2.6 bn (29%) 1.7 bn (19%) 9 bn 107
Ireland 427 mn
(51%)
183 bn (22%) 229 mn (27%) 839 mn 168
United Kingdom 10.8 bn
(53%)
3.7 bn (18%) 5.8 bn (29%) 20.3 bn 299
United States 60.9 bn
(44%)
38 bn (27%) 39.7 bn (29%) 138.6 bn 419
Source: Table compiled by authors with data from Statista, available at: www.statista.com.
In terms of growth, all countries have seen significant increases in the costs of digital advertising, with
COVID-19 representing a small deviation in the general trend (see Table 7 for more details). For
example, Ireland saw the digital advertising spend in the Irish market for 2019 rise by 17% to reach EUR
673 million
332
. In France, 2018 saw digital becoming the leading medium for advertising sales (40%),
compared with 27% for television
333
.
330
Irish Statute Books, 2011, European Communities (Electronic Communications Networks and Services) (Privacy and Electronic
Communications) Regulations.
331
This corresponds roughly EUR 323 per capita [exchange rate 31.12.2020].
332
RTE, 2020, Irish digital advertising spend grew by 17% last year.
333
Report commissioned by the French Minister of Culture and the Secretary of State in charge of digital, 2020, Online advertising: A level
playing field.
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
81 PE 662.913
The Syndicat des Régies Internet (SRI) provided an estimate of the growth of the digital advertising
market in France at +7% for the whole of 2021
334
. The advertising market share of online advertising in
Germany has increased from 12% in 2019 to 40% in 2020
335
. From a consumer perspective, this growth
is reflected in the prices of goods and services, particularly those which make heavy use of digital
advertising (for example consumer electronics, hotels, and flights). As elsewhere, market manipulation
and the costs of misleading actions and aggressive commercial practices are a concern in the UK. In
2009, Consumer Focus estimated that misleading and aggressive practices cost EUR 3.8 billion (GBP 3.3
billion)
336
.
Box 16: Effect of COVID-19 on online advertising in France and Australia
The COVID-19 pandemic is having a large impact on various industries, and digital advertising is no
exception. Businesses have been assessing their marketing efforts,
which includes online
advertising. According to a recent report by the Interactive Advertising Bureau, almost 24% of media
buyers, planners, and brands paused ad spending until Q2 2020
337
. In France, Digital advertising
revenues were down by 8% in the first half of 2020
338
. Display, banners, video, special operations and
audio all experienced a 17% decline in revenue and search was down by 9%, however retail search
digital advertising has continued to grow at a rapid pace, especially due to the success of
eCommerce during the lockdown. The Australian Digital Advertising market experienced double-
digit decline in Q2 2020 due to the Impact of COVID-19. All online advertising categories showed
declines compared to quarter ending June 2019 with Search down 9% and Display 11%. Video has
outperformed the overall general display market, with no change in expenditure year on year,
increasing its share of the general display market by to 53%
339
. However, recent research from the
Consumer Policy Research Centre (CPRC) has also found that, after the adjustment in 2020, the
COVID-19 pandemic has resulted in consumers relying on digital markets at much higher levels
340
.
The market is therefore expected to recover fully in the long-
term and continue growing
exponentially.
Source: Authors’ own elaboration.
4.2. Design and implementation of responses
All case study countries except Australia have chosen to implement self-regulatory principles, to
support consumer protection legislation, when it comes to online advertising. As can be seen in
Table 8 below, whether a system is self-regulatory or implements mandatory regulation affects the
actors in the system. Most EU Members States, and many non-European Union countries, have a self-
regulatory organisation that is a member of the European Advertising Standards Alliance (EASA). EASA
co-ordinates the cross-border complaints system for its members (which include the Advertising
Standards Agency (UK))
341
.
334
Oliver Wyman, 2020 Report, Syndicat des Régies Internet (SRI).
335
Giersberg, F., 2020, Advertising market 2020 with clear corona effect.
336
Department for Business, Energy & industrial Strategy, 2018, Misleading and aggressive commercial practices: new private rights for
consumers.
337
Mordor Intelligence, 2021, Online advertising market – growth, trends, covid-19 impact and forcasts (2021 - 2026)
338
Amiot, E., 2020, Digital Advertising in France Weakened by the Pandemic, Syndicat des Régies Internet (SRI).
339
Le Roy, G., 2020, Australian Digital Advertising Market Experiences Double Digit Decline In Q2 2020 Due To Impact Of COVID-19.
340
Consumer Policy Research Centre, 2020, Unfair trading practices in digital markets – evidence and regulatory gaps, Research and policy
briefing.
341
Advertising Standards Agency, 2014, Scope of the Code: CAP Code.
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 82
There are two other notable points from Table 8 below. The first is that Germany has no dedicated
national authority. This is because the main legislation is enforced by the market participants in
the district court (Landgericht) via injunctions which can prohibit online advertising activities. This a
specific legal instrument in German civil law, which has legal effects for both process participants and
third parties. For example, through this instrument a ‘blacklist’ of unfair business actions, which impact
entrepreneurs and include both misleading and aggressive commercial practices, has been created
342
.
The rationale of this ‘blacklist’ is to protect consumers’ freedom of choice and consumers’ right to make
informed decisions.
The other notable example is the UK system which involves the Committee of Advertising Practice
(CAP), rather than one single industry body. This Committee develops and implements the UK Code of
Non-Broadcast Advertising and Direct Promotional Marketing (CAP Code) and is mainly supported by
the Advertising Standards Agency. However, it also includes member organisations who, through
contractual agreements with media publishers and carriers, agree to comply with the Code on behalf
of other businesses who may not be a member of the Advertising Standards Agency. In that sense, the
CAP is not just one ‘catch all’ industry body, but a mixture.
Table 8: Effect of COVID-19 on online advertising in France and Australia
Country
Self-
regulation
Designated
industry body(-ies)
National authority(-ies)
Australia
N
None
Australian Competition and Consumer
Commission (ACCC) and the Australian
Communications and Media Authority
(ACMA)
France
Y
Regulatory authority for
advertising (ARPP)
The National Competition Authority (NCA),
National Commission for Informatics and
Freedoms (NCIF), and the Audiovisual Council
(AC)
Germany
Y
German Advertising
Association (GAA)
None
Ireland
Y
Advertising Standards
Agency (ASAI)
Competition and Consumer Protection
Commission (CCPC)
United
Kingdom
Y
Committee of
Advertising Practice
(CAP)
National Trading Standards (NTS)
United States
Y
BBB National Programs
(BBBNP)
Federal Trade Commission (FTC)
Source: Authors’ own elaboration.
In contrast to the other case studies, in Australia there is no single designated industry body mirroring
the national authorities mandated to regulate online advertising in the country. This is because many
of the mandatory regulations, such as the ‘News Media Bargaining Code’, overseen by the ACCC and
the Australian Communications and Media Authority (ACMA), were only introduced in early 2021.
Furthermore, the lack of a self-regulatory system negates the demand for a designated industry
interlocutor for the national authorities. It is important to note that, in terms of impact, despite the
‘News Media Bargaining Code’ being less than one year old, it has already led to both Google and
Facebook negotiating voluntary commercial agreements with a number of large and small Australian
342
German Law, 2004, Act against Unfair Competition (UWG), Annex (to Section 3 Paragraph 3).
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
83 PE 662.913
news businesses. A short description of the general context of mandatory regulation in Australia,
including from the origins of the code, is outlined in the box below.
Box 17: The path of online advertising regulation in Australia
To tackle the issues arising from online advertising in both business-to-business (B2B) and business-
to-consumer (B2C) relations, the ACCC made a number of recommendations to the Australian
Government as part of the Final Report of the Digital Platforms Inquiry
343
. A number of these
recommendations have already been implemented by the Australian Government.
The Digital Platforms Inquiry which ran between December 2017 and July 2019 looked at the
effect that digital search engines, social media platforms and other digital content aggregation
platforms have on competition in media and advertising services markets
344
. The final report
published in 2019 contains recommendations reflecting the issues arising from the growth of digital
platforms
345
. The recommendation to address the imbalance in the bargaining relationship
between these platforms and news media businesses was a key driver behind the adoption of the
News Media Bargaining Code.
Another noteworthy inquiry is the inquiry into markets for the supply of digital platform
services, which commenced in February 2020 and will continue until March 2025. The first report
of this inquiry was released in September 2020 and provides an in-depth focus on online private
messaging services in Australia. The second report, which considers mobile app marketplaces, is to
be publicly released by the end of April 2021
346
. A separate ACCC inquiry into markets for the
supply digital advertising technology services and digital advertising agency services
released an interim report earlier this year, with a final report due to be publicly released in
September
347
.
Overall, many different regulatory measures have been adopted in Australia, some of which have
not become law yet, as there is typically a significant consultation period, and the results of inquiries
are carefully considered and often reflected in legislative proposals.
Source: Authors’ own elaboration.
In terms of self-regulation, all the case study countries using a system of self-regulation have some
form of code or guidelines for online advertising. In France for example, the French self-regulatory
authority’s (ARPP) code of conduct on online advertising is the main set of principles and includes
recommendations applicable to some digital format/techniques, such as social media. In the UK, the
12
th
edition of the CAP Code came into force in September 2010 and was released because the digital
remit of the Advertising Standards Agency (ASAI) was extended to cover online marketing
communications. In Ireland, the Advertising Standards Authority is the independent self-regulatory
body set up and financed by the advertising industry, and it publishes a Code of Standards
348
, which
includes rules on online behavioural advertising, which are binding for its members. More information
can be found in Box 18 below. In Germany, the Advertising Standards Council has a set of operating
principles, last updated in 2009
349
.
343
ACCC, 2019, Digital Platforms Inquiry: Final Report.
344
Ibid.
345
Ibid.
346
ACCC, 2021, Digital platform services inquiry 2020-2025.
347
ACCC, 2021, Digital advertising services inquiry.
348
Advertising Standards Authority for Ireland, 2016, Code, 7th edition.
349
German Advertising Standards Council, 2009, Working Principles.
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 84
Box 18: Ireland, an example of mixed regulatory systems
In Ireland, similar to France, Germany the United Kingdom and to some extent, the United States, a
designated industry body (the ASAI) and a national authority (CCPC) both enforce regulations on
online advertising. Both authorities use codified principles to seek action against traders and
advertisers. The ASAI Code more specifically targets marketing and also includes distinct rules for
online behavioural advertising. While consumers who have experienced damages can choose to
follow either the CCPC or the ASAI procedures, the self-regulatory procedure under the ASAI
Code provides quicker relief than the procedure under the Consumer Protection Act (CPA), with the
CCPC. Indeed, while a certain level of severity has to be reached for the CCPC to seek action against
a trader, the ASAI takes action as soon as an advertiser has breached their code. This implies any
consumer can start a procedure with the ASAI. The decisions by the ASAI are also made public, which
may cause negative publicity to advertisers and thus, create a dissuasive effect
350
.
The CCPC is a member of the EU Consumer Protection Cooperation (CPC) Network, which is
responsible for enforcing EU consumer protection laws in EU and EEA countries. Under the CPC
framework, any authority in a country where consumer protection law may have been breached can
ask its counterpart, in the country where the trader is based, to take action to stop this breach of law.
Authorities can also alert each other to possible breaches that could spread to other countries and,
with the European Commission’s support, authorities can also coordinate their approaches to
applying consumer protection law to tackle widespread infringements
351
.
Source: Authors’ own elaboration.
In the United States, the system can be seen as self-regulatory but underpinned by a strong regulator
with the ability to make case law (its role outlined in the box below). In this case, the FTC itself, in
consultation with industry, provides advice to online companies and commercial websites. In February
2009, the FTC released a revised set of principles for the self-regulation of online behavioural
advertising
352
. These guidelines are also currently under review and the role of the FTC more generally
is interesting from a best practice point of view.
350
Advertising Standards Agency for Ireland, 2020, Complaints Bulletins.
351
Competition and Consumer Protection Commission, 2019, Annual Report 2019.
352
Federal Trade Commission, 2009, Self-Regulation Principles for Online Behavioral Advertising.
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
85 PE 662.913
Box 19: The regulatory practice on online advertising in the United States
In the United States, the FTC’s role conducting investigations and contributions to case law differs
from the other national authorities’ roles and allows the FTC to react quickly to emerging market
practices. The American system differs from the other systems examined as the role of case law is
more prominent in upholding the consumer protection and privacy principles set in Section 5 of
the FTC Act. For example, the FTC was recently able to propose an order
353
on the use of online facial
recognition by companies which includes a new algorithmic remedy. To complement its
investigation activities, the FTC actively liaises with stakeholders, through guidelines and
workshops. The FTC is currently organising a workshop on dark commercial patterns, for which
stakeholders can share their comments ahead of time and which will result in the publication of a
report. In elaborating their guidelines, the FTC also adopts a collaborative approach with
stakeholders. The Endorsements and Testimonials in Advertising (known as the Endorsement
Guides) were last updated in 2009 and are currently under review
354
. As a part of the review, the FTC
initiated a notice and a consultation period during which stakeholders can address their comments
to the FTC, all of which are available online, on the ‘regulations.gov’ website. The procedure typically
takes a few years. The Commission will vote on whether to adopt the recommended changes.
The FTC has created the Consumer Sentinel Network, which provides a knowledge sharing
platform for its members on issues that relate to consumer protection through reports
355
.
Internationally, the FTC consults regularly with DG JUST about developments in consumer law and
enforcement cooperation. Although the US laws and legal system differs significantly from the EU,
the FTC and the EU share many of the same high-level goals in the area of consumer protection
including in relation to digital advertising. To that end, the FTC is very active in international
engagement in the OECD, the International Consumer Protection Enforcement Network (ICPEN),
and other international organisations, and engages at EU level or bilaterally with Member States
356
.
Source: Authors’ own elaboration.
4.3. Evaluation of EU Member State and third country experiences
Overall, there are several lessons concerning the institutional framework and enforcement that can be
learned at the EU-level from experiences at national level within and outside Europe.
The FTC as an authority shows many positive attributes. Although the research the FTC leads (including,
for example, the creation of the Consumer Sentinel Network), as well as the cases it deals with, are
specific to the US, its statutes can be much broader and higher level than EU laws. The FTC’s activity on
specific cases allows quick responses to a rapidly growing industry in which new issues arise quickly,
and its review process for guidelines is directed yet participatory, as evidenced by the review of the
Endorsement Guides.
In Ireland, the activity led by the industry body, the ASAI, does appear to complement the national
authority, CCPC, which deals predominantly with more severe breaches which have already caused
harm to consumers. The self-regulatory procedure under the ASAI Code provides quicker relief than
the procedure under the CPA, with the CCPC, and this dual system could be considered a best practice.
The UK industry body, the ASA, is also quite advanced with regards to self-regulation and dealing with
353
Federal Trade Commission, 2021, California Company Settles FTC Allegations It Deceived Consumers about use of Facial Recognition in Photo
Storage App, Press Release.
354
United States Federal Trade Commission, 2020, FTC Seeks Public Comment on its Endorsement Guides, Press Release.
355
United States Federal Trade Commission, Enforcement page: Consumer Sentinel Network.
356
United States Federal Trade Commission, About the FTC: Office of International Affairs.
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 86
specific cases. In 2019, the ASA issued 62 sanctions leading to compliance, and only 9 advertisers
needed to be referred to Trading Standards for further action to be taken
357
. In France, there is a self-
regulatory system with strong engagement from national authorities. Moreover, the competition
authority publishes opinions on online advertising, such as the one of 6 March 2018 on data
exploitation in the online advertising sector
358
. This quasi-regulatory activity by the French national
authorities could be seen as a form of ‘pro-active self-regulation’, which may be considered an EU best
practice most similar to the US model. In Germany, consumers have a positive view of the development
of the main legislation. This is perhaps due to a greater awareness of the debate around data and
privacy than in other countries. Consumers still perceive the current regulatory framework as a mixture
of paradoxical advantages, disadvantages and neutral or ambivalent elements
359
. This perception taps
into general consumer feelings in all the case study countries regarding what is known as the
‘personalisation-privacy’ paradox. However, it appears German consumers are perhaps more aware of
what this involves in terms of online advertising and ‘free’ online services.
In Australia, the adoption of a framework to curtail the market power of digital platform giants, such as
Google and Facebook, has attracted a strong interest from other policy makers and competition
regulators, including the UK’s Competition and Markets Authority (CMA)
360
. According to a recent
statement by the UK government, the UK’s Digital Secretary asked the new Digital Markets Unit to ‘look
specifically at how a [mandatory] code would govern the relationships between platforms and content
providers such as news publishers, including to ensure they are as fair and reasonable as possible
361
. It
is worth noting that this regulatory measure is comparable to Article 15 of the EU Copyright Directive,
which also aims to require payments from digital platforms to publishers
362
.
357
UK Advertising Standards Agency, 2019, Using technology for good – Annual Report 2019.
358
Autorité de la concurrence, 2018, Opinion n° 18-A-03 of 6 March 2018 on the use of data in the internet advertising sector advertising sector.
2018.
359
Dehling, T., Zhang, Y. & Sunyaev, A., 2019, ‘Consumer Perceptions of Online Behavioral Advertising’, Proceedings of the 21st IEEE Conference
on Business Informatics.
360
Lomas, N., 2021, After Facebook’s news flex, Australia passes bargaining code for platforms and publishers, Tech Crunch.
361
Ibid.
362
European Parliament and European Council, 2019, Directive (EU) 2019/790 of the European Parliament and of the Council of 17 April 2019 on
copyright and related rights in the Digital Single Market and amending Directives 96/9/EC and 2001/29/EC (Text with EEA relevance.).
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
87 PE 662.913
CONCLUSIONS AND RECOMMENDATIONS
In this section, we synthesise the insights from the previous sections to provide recommendations that
may address the gaps identified. We distinguish between recommendations to address challenges for
consumers and recommendations to address challenges for SMEs as publishers and advertisers, and to
support the development of a single market in online advertising.
5.1. Recommendations to better protect consumers
5.1.1. Ensuring that data use and sharing in digital advertising conforms with privacy
rules
The delivery of behavioural advertising typically involves the extensive use of consumer data that may
be collected without consumers’ knowledge. Consumers have little understanding of who is collecting
data and how or what data is being collected and processed. The use of dark patterns to leverage
consent for data collection and processing can further exacerbate this issue (see below). In addition,
advertising auctions are often based on the transfer of large amounts of personal data between a
variety of different actors. Usually there is little awareness and control of who the data is shared with
or whether it is used in contexts other than advertising. Both practices run counter to the rights
recognised in the GDPR and highlight the need for an improvement in the rules and enforcement.
The findings of this study and the interviews conducted suggest that while the GDPR has provided the
legal framework to address these issues, the fundamental problem with the GDPR is that the measures
are often only selectively enforced by National Data Protection Authorities. There is a need for more
consistent and adequate enforcement of the legal framework, to achieve the underlying objectives of
the regulation. However, National Data Protection Authorities are not always well equipped to
effectively and efficiently enforce the GDPR due to a lack of necessary resources. The European
Parliament Resolution on GDPR implementation
363
points to the lack and uneven enforcement of the
GDPR by National Data Protection Authorities across the EU which is due, in part, to lack of human,
technical and financial resources. According to the Resolution, weak enforcement is particularly
evident in cross-border complaints and the European Parliament has expressed concerns over the
functioning of the one-stop-shop mechanism regarding the role of the Irish and Luxembourg DPAs
which are responsible for dealing with issues relating to the largest tech companies which tend to be
registered in those two Member States.
A short term solution would be to improve the working of the coordination mechanism foreseen by
the GDPR. A more robust, but longer term, solution may be:
• To insert provisions in the GDPR that allow for the intervention of the European Commission as
a complement to the intervention of the authority of the country where the platform is
established when it comes to large digital companies operating across multiple countries,
similar to what is provided for in the current DSA proposal
364
.
363
European Parliament resolution of 25 March 2021 on the Commission evaluation report on the implementation of the General Data
Protection Regulation two years after its application, paras. 12-22. See also Brave (2020): Europe’s governments are failing the GDPR –
Brave’s 2020 report on the enforcement capacity of data protection authorities, which shows that as the number of complaints increases,
it becomes more difficult for the authority to act quickly and efficiently to address them. In addition, Brave shows that alongside a low
headcount, budget increases are also decelerating, which is equally problematic when acting against such financially strong companies
364
European Commission, 2020, Proposal of the Commission of 15 December 2020 for a Regulation of the European Parliament and of the
Council on a Single Market For Digital Services (Digital Services Act) and amending Directive 2000/31/EC, COM(2020) 825 final, 2020/0361
(COD),arts. 50-66.
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 88
5.1.2. Informing consumers about the fact that they are being targeted and improved
consent mechanisms
Consumers are often not aware that they are being targeted and thus influenced by online advertising.
In addition to affecting purchasing decisions, the ability to target messages to specific consumer
groups provides the potential for the manipulation of public opinion by certain players
365
. Therefore,
additional information requirements should be imposed on online platforms (and websites in general)
that display advertising, enabling the recipient to recognise advertising, to verify the sender of the
advertising and the parameters used to determine the recipient. Various existing legal instruments
cover aspects of these problems including the eCommerce and AVMS Directives, the amended CDS
and UCPD, and the GDPR. However, the rules are complex and may be subject to inconsistent
application and enforcement.
The current DSA proposal is a step in the right direction since it will strengthen the level of awareness
of consumers about online advertising. The proposed DSA could be further strengthened by ensuring
that the requirements apply to all intermediary services in scope (including search engines). It must be
noted however that the proposed DSA does not apply to all ‘information society services’ covered by
the eCommerce Directive, which means that many online websites/publishers will not be covered (see
Dealing with multiple actors below). In terms of the information to be conveyed to consumers it seems
debatable whether they should be informed about the value of their provided data in monetary terms.
Any individual consumer would most likely evaluate monetary amounts towards their own disposable
income or their wealth in general. Hence, consumers' impression of the "monetary" value of their data
would be mainly driven by their own financial status and not by their privacy preferences. The latter,
however, should be the crucial factor in the consent decision.
More promising could be to provide information on what data is specifically used in which targeting
method. In this way, consumers would be able to know how for instance their address, age, sex,
influences the displayed advertisements. Based on this information, consumers will be able to perform
a self-assessment of their individual privacy risk rather than relying on a pre-calculated monetary value.
How (format and location) to display mandatory information requirements could be specified in codes of
conduct.
Article 24 of the DSA proposal aims to address to some extent the need for clarity around the
parameters used for targeting, but further steps could be taken to ensure that all the challenges listed
are addressed. Specifically, the DSA could contain:
• A requirement to ensure that there is transparency concerning the fact that the consumer is
exposed to targeted advertising and on the main parameters used therefore (but not how they
are specifically processed in the targeting algorithm);
• A requirement that information about targeted advertising should be conveyed in a manner
that takes into account consumer biases and limitations in dealing with information overload.
To apply this principle in practice, different options could be tested by the platforms (A/B
testing) and those which are the most effective could be included in the codes of conduct
foreseen by Article 36 of the DSA proposal;
• A requirement to provide at least one option which is not based on targeting (similar to what
Article 29 of the proposed DSA foresees for the recommender systems used by the very large
online platforms); and
365
A concern expressed during an interview.
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
89 PE 662.913
• Moreover, the DSA should encourage opt-in setting to targeted advertising via the tools
available to support self- and co-regulation such as codes of conduct (Article 36 of the DSA
proposal).
5.1.3. Addressing “dark patterns” through GDPR guidelines
Dark patterns refer to designs and wording (e.g. in the context of consent forms) which aim to lead
consumers to take decisions which may not be in their best interests.
Some EU authorities have used the GDPR to sanction the use of dark patterns. For example, the French
Data Protection Authority fined Google for a “lack of transparency, inadequate information and lack of
valid consent in its ads personalisation tool
366
”. Some EDPB Guidelines also cover this issue to some
extent.
Further action may be appropriate to ensure the issue is addressed Europe wide. Potential solutions
could include:
• Defining design guidelines, among others, for cookie banners and consent forms; and
potentially as a longer-term solution (noting that EDPB guidelines are not binding);
• Establishing a higher legal standing for the EDPB guidelines in future revision of the GPPR, for
instance by explicitly providing that the DPA and national court should take the guidelines into
utmost account; and
• Providing a user-friendly tool to report websites that may not comply with the proposed design
guidelines
367
.
5.1.4. Clarifying that digital targeted advertising must not breach rules on discrimination
and improving algorithmic transparency
The availability of data is useful in displaying relevant ads. However, it can also enable advertisers to
intentionally engage in discrimination against certain customer groups e.g. based on age, race, gender
or disability. In addition, wherever a lot of data is evaluated by algorithms, unintentional discrimination
can occur in ad delivery if the training datasets for the ad algorithms are systematically biased with
respect to specific subgroups of the population.
Algorithmic systems are the cornerstone of platforms' content moderation, data analysis and
recommender as well as ranking systems. Incorporating an access obligation that refers to not only the
working algorithm but also grants access to data which was used to build and test the algorithm (so
called “training data”) is crucial for evaluating the compliance of these systems. Without the associated
training data, it is difficult to backward-engineer the construction of the algorithm and assess its
potential for, e.g., biases in content moderation or potential for data-based discrimination within
recommender systems.
These problems may largely be addressed by existing and planned legislation. The proposed AI Act will
be a horizontal legislation applicable to all the actors involved in online advertising. The AI Act may
tackle discrimination, as the current proposal suggests addressing algorithmic bias and requiring more
transparency, better documentation and traceability for (high-risk) AI systems
368
.
366
Global Data Hub, 2020, Beware 'dark patterns' – data protection regulators are watching.
367
In the US, users can report websites to their respective Attorney General if they feel they are subjected to a dark pattern. Available at:
https://darkpatternstipline.org/report/.
368
European Commission, 2021, Proposal for a regulation of the European Parliament and of the Council laying down harmonised rules on
artificial intelligence (artificial intelligence act) and amending certain Union legislative acts, 21.4.2021 COM(2021) 206 final 2021/0106
(COD), arts 7-19.
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 90
Moreover, the non-discrimination principle is enshrined in different EU legislative instruments like the
“Race Directive”
369
, the “Employment discrimination Directive”
370
, the “Gender equality Directive”
371
,
and the “Gender Directive”
372
, some of which may be applicable to digital advertising. In addition, the
GDPR includes an “explainability” rule regarding algorithms
373
and the AVMS Directive explicitly
requires non-discrimination in relation to video sharing platforms
374
. Meanwhile the DSA proposal
addresses illegal content (which should encompass content which breaches EU and national legislation
forbidding discrimination)
375
. However, the DSA could nonetheless further contribute to ensuring that
rules regarding discrimination are adhered to in the context of digital advertising through:
• Greater transparency concerning targeting itself and the main parameters thereof as well as
giving the possibility to change those parameters (see suggestions above); and
• Regular vetting of systems and training data by accredited researchers will offer new insights
on how to mitigate systemic risks and lead to lessened information asymmetries.
5.1.5. Ensure that minors are not subject to harmful targeted advertising which exploits
their vulnerabilities
User profiling can enable certain vulnerable consumer groups including minors to be identified and
shielded from inappropriate advertising. However, the same profiling techniques could also be used to
exploit such users’ vulnerabilities and display inappropriate advertisements.
The FTC in the US uses Section 5 complemented by privacy-related statutes to protect consumer privacy,
including the Children’s Online Privacy Protection statute. In December 2020, the FTC issued orders to 9
social media and video streaming companies under Section 6B of FTC act that allows them to collect
information from companies. The orders require the companies to provide data on how they collect, use,
and present personal information, their advertising and user engagement practices, and how their
practices affect children and teens.
In the EU, the AVMS Directive already includes obligations to protect minors from harmful content in
advertising, causing physical, mental or moral detriment. However, the provisions apply only to video
sharing platforms
376
. More generally, targeting advertising towards children is addressed indirectly by
the provisions of the GDPR that make clear that children cannot give consent
377
.
However, to clarify that minors (and potentially other vulnerable customer groups) should be
protected from targeted advertising:
369
Council of the European Union, 2000, Council Directive 2000/43/EC of 29 June 2000 implementing the principle of equal treatment
between persons irrespective of racial or ethnic origin, OJ L 180 of 19.7.2000.
370
Council of the European Union, 2000, Council Directive 2000/78/EC of 27 November 2000 establishing a general framework for equal
treatment in employment and occupation, OJ L 303 of 02.12.2000.
371
European Parliament and the Council of the European Union, 2006, Directive 2006/54/EC of the European Parliament and of the Council
of 5 July 2006 on the implementation of the principle of equal opportunities and equal treatment of men and women in matters of
employment and occupation (recast), OJ L 204 of 26.7.2006.
372
Council of the European Union, 2004, Council Directive 2004/113/EC of 13 December 2004 implementing the principle of equal treatment
between men and women in the access to and supply of goods and services, OJ L 373 of 21.12.2004.
373
GDPR, Article 22.
374
AVMSD, Article 28b(2) which refers to Article 9(1).
375
DSA proposal, Article 2(g) defines illegal content as “any information, which, in itself or by its reference to an activity, including the sale
of products or provision of services is not in compliance with Union law or the law of a Member State, irrespective of the precise subject
matter or nature of that law”.
376
AVMSD, Article 28b(2) which refers to Article 9(1).
377
GDPR, Article 8.
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
91 PE 662.913
• The DSA could include a similar provision to that related to commercial communication of the
AVMSD which is applicable to video sharing platforms, thereby extending the rules on the
protection of minors to all hosting platforms.
5.1.6. Dealing with multiple actors
Many actors are involved in targeted advertising: publishers, ad intermediaries, and advertisers. In
many cases, targeted advertising is pushed to the publisher’s website and to the end user without the
intervention of the publisher. The scope of the proposed DSA is limited to intermediation services- and
the rules on advertising are themselves limited to online platforms and VLOPS, which means that many
websites will not be covered, whereas the issues raised in this report go beyond these actors.
Furthermore, any legislative solution which is limited to certain platforms, needs to be mirrored up the
value chain since the ad intermediaries are often the engines of targeted advertising solutions.
Similarly, the notion of joint controller under the GDPR is not always crystal clear and not necessarily
apparent to the end-user. Possible solutions may be:
• To clarify that Article 5.3 of the DSA proposal could also apply to platforms which lead an
average and reasonably well-informed consumer to believe that advertising is provided by the
online platform itself or by a recipient of the service who is acting under its authority or control.
This would ensure that the platform has an incentive to comply with all the transparency rules
since otherwise they would be deprived of the liability exemption of Article 5.1; and
• To potentially consider new legislation at EU level which ensures that any ex-ante legislation
protecting consumers is also applicable to ad intermediaries.
5.1.7. Clarifying the redress mechanisms
Online advertising needs to comply with many EU and national rules. To date, the enforcement
mechanisms are different depending on the specific area that is
breached. The proposed DSA also sets
out a generic mechanism for users to flag illegal content and to seek redress. This may further
complicate matters for consumers. Potential solutions could be to:
• Address in more detail how sector specific rules will interact with the DSA, especially on
enforcement;
• Include among the list of tasks for the Digital Service Coordinators the need to provide
information to consumers on how to seek redress in relation to online advertising (among
other areas); and
• Envisage the adoption of a sector specific directive to address all the consumer protection
issues linked to online advertising, thereby ending a piecemeal approach which is not
necessarily conducive to consumer trust.
5.1.8. Facilitating the functioning of the internal market
The risk of legal uncertainty deriving from the potential application of national rules to advertising
services could be mitigated by enforcing the cooperation mechanisms between Member States so that
derogations remain the exception rather than the rule. Specifically, the condition of prior notification
of the measure laid down in Article 3(4b) of the eCommerce Directive, could be better harmonised and
clarified. This could be done for instance by:
• Amending the eCommerce Directive (possibly through the DSA) by introducing precise
deadlines and procedural conditions for the implementation of derogations by Member States
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 92
(e.g.: the Member State where the service is provided could not take any measure before the
expiration of a three-month deadline)
378
; and
• EU-wide codes of conduct which could be ‘vetted’ by the European Commission to define
which types of national restrictions would be compatible with the internal market clause.
5.2. Recommendations to better protect SMEs as publishers and
advertisers
5.2.1. Addressing exploitation by platforms which hold a dominant position in digital
advertising
Large platforms providing ad-tech services such as Google and Facebook allow access to a very high
proportion of EU Internet users. As such, smaller firms in particular may be reliant on these channels to
reach consumers, potentially allowing exploitation in the form of higher charges or other restrictions
in the advertising market.
The proposed DMA includes provisions which aim to shine a light on potential exploitation by requiring
digital gatekeepers to provide information on prices and performance regarding digital advertising to
advertisers
379
. Alongside approving the DMA provisions, potential exploitation of gatekeeper platforms
in digital advertising could be addressed by:
• Taking advantage of the additional information to pursue case by case enforcement under
competition law at the EU and national level.
5.2.2. Increase transparency concerning advertising auctions and the performance of
advertising
The distribution of ads via ad-auctions or RTB is marked by a lack of transparency towards both
advertisers and publishers. According to interviews, this can create a situation where advertisers cannot
understand why they win or lose auctions for specific keywords and how the final price for the
impression or conversion is determined. Likewise, publishers face similar uncertainty since they
observe only their personal ad revenue, but how much is pocketed by the intermediary remains
unknown. The same applies to performance data of the advertisement or the ad-inventory.
Advertisers largely depend on data that is supplied by the ad intermediaries and which has already
been edited or aggregated in some form. Access to raw data on ad performance is lacking. This creates
a situation in which it is difficult for advertisers to calculate their own unbiased returns on their
advertising expenses (Ad-ROI).
The insights of this study and the interviews conducted for this study suggest that the introduction of
obligations for B2B side performance data provision would be a promising step. Ideally, this data
should be unedited and in an agreed upon standardised format to maximise usability. The benefits of
this approach are two-fold. Firstly, access to raw data allows for a quality comparison of different
intermediary services based on actual success and conversion rates. This gives both advertisers and
publishers the possibility to make informed decisions with respect to their ad spending and ad
revenues.
378
Such recommendations are inspired by the mechanisms laid down in Article 3 of the AVMS Directive (as amended as amended by
European Parliament and the Council of the European Union, 2018, Directive (EU) 2018/1808 of the European Parliament and of the
Council of 14 November 2018 in view of changing market realities, OJ L 303/69 of 28.112018).
379
DMA proposal, Article 5(g) and Article 6(1)(g).
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
93 PE 662.913
Furthermore, this would most likely increase the competitive pressure on large intermediaries of
Facebook and Google, which is currently rather low, and would promote innovation in the long run.
Secondly, increased transparency with respect to calculable click-through and conversion rates would
strongly benefit the potential to detect ad-fraud. Currently, it is difficult for an advertiser to detect
whether they are targeted by fraudulent behaviour or for a publisher whether fraud takes place on their
websites. If access to raw data is granted, this change as ad-fraud would most likely lead to a significant
increase in click-through-rates (increasing an advertiser’s costs) while conversion-rates would drop
(absolute conversions remain constant). Hence, an unusually large discrepancy between click-through
and conversion rates could be a first indicator for present fraud.
The DMA proposal covers some of the following issues, at least to some degree. For example,
gatekeepers are required by Article 6(1)(g) of the DMA proposal to provide advertisers and publishers
with access to the performance measurement tools and information they need to independently
validate advertising inventory. In addition, gatekeepers are obligated by Article 5(g) of the DMA
proposal to “provide advertisers and publishers to which it supplies advertising services, upon their
request, with information concerning the price paid by the advertiser and publisher, as well as the
amount or remuneration paid to the publisher, for the publishing of a given ad and for each of the
relevant advertising services provided by the gatekeeper”
380
. However, the opacity of the ad auction
system is not entirely addressed by the latter provision. It remains unclear how the resulting prices are
calculated, which other factors are considered in the auction process and how they are weighted when
determining the auction winner. A potential solution could be:
• To add to the proposed transparency obligations in the DMA obligations regarding
transparency of the auction criteria used by the gatekeeper of the ad-tech core platform service
that include details of the price components (CPC, CPI, CPM) as well as other factors which are
taken into account in the auction process and their weighting.
5.2.3. Tackling bundling and tying by gatekeeper intermediaries of premium advertising
space
Ad-inventory of vertically integrated large intermediaries (e.g. Facebook, YouTube (Google)) is
considered very valuable from an advertiser’s perspective but is often exclusively marketed via their
own Ad-Network or Ad-Exchange. This applies for example to ad inventory on Facebook or on YouTube
which are only accessible via their respective networks. However, the non-availability of such premium
publishers’ inventories to smaller competing ad intermediation networks may act as an entry barrier
and limit their ability to compete. This concern has been also expressed in an interview conducted with
a European ad-network. This issue has been recently addressed by the current DMA proposal of the
European Commission
381
. According to Article 5(f) of the DMA proposal a gatekeeper shall “refrain from
requiring business users or end users to subscribe to or register with any other core platform
services”
382
. Essentially, a gatekeeper is prohibited from jointly marketing/selling multiple core services
for which it has been designated as enjoying a gatekeeper position.
380
European Commission, 2020, Proposal of the Commission of 15 December 2020 for a Regulation of the European Parliament and of the
Council on contestable and fair markets in the digital sector (Digital Markets Act), COM(2020) 842 final, 2020/0374 (COD).
381
Cabral, L., Haucap, J., Parker, G., Petropoulos, G., Valletti, T. and Van Alstyne, M., The EU Digital Markets Act, Publications Office of the
European Union, Luxembourg, 2021, ISBN 978-92-76-29788-8 (online), doi:10.2760/139337 (online), JRC122910.
382
European Commission, 2020, Proposal of the Commission of 15 December 2020 for a Regulation of the European Parliament and of the
Council on contestable and fair markets in the digital sector (Digital Markets Act), COM(2020) 842 final, 2020/0374 (COD).
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 94
Separating the intermediary services from these vertically integrated players and the ad-inventory of
their publisher’s sites could be beneficial in reducing entry barriers and could promote competition on
the intermediary level in the online advertising value chain. Furthermore, it could work against the
strong concentration tendencies observed in this market and lessen concerns over market power and
negative effects on advertising prices and consumer choice.
5.2.4. Addressing asymmetric access to consumer data
Due to their user-facing services, large providers of ad-tech services like Google and Facebook can
access extensive amounts of data while other companies lack such access. This leaves challenges to
large providers in a weaker competitive position. Concerns have been raised that these large providers
not only bundle access to the data with the purchase of advertising solutions and services, but may
also directly prohibit or introduce considerable obstacles to the use of the data via a competitors'
advertising services
383
. Although the prohibition on bundling and self-preferencing in the DMA
proposal provides some assistance in this regard
384
, further obligations could be established to address
this issue more vigorously. Other regulations, such as the Data Governance Act, do not yet seem to fully
address the data sharing issue. However, the forthcoming Data Act may include some data sharing
provisions. The P2B Regulation addresses the need to be transparent in terms and conditions on data
sharing. A recommendation could be:
• Further research on this issue may be helpful with a focus on the obstacles to the use of data
by third parties and which measures could be introduced to tackle barriers and foster
innovation.
5.2.5. Counteracting the monopolisation of the browser market
Google’s Privacy Sandbox as well as Apple’s ad tracking restriction efforts have been introduced with
reference to their ability to increase user privacy. While Google's measure aims to make third-party
cookies obsolete for behavioural advertising in the future, Apple's empowers consumers with more
control over their data. At first glance, these efforts appear to be a positive development, indicating
that the market is addressing consumers' needs for privacy and developing mechanisms to protect
them.
However, both companies operate platforms that are used by a large number of consumers. As a result,
other companies in the advertising chain are dependent on them. The restrictions will hinder these
players in carrying out their business, at least in the short term.
Google’s approach is viewed particularly critically. Google itself recently acknowledged that the current
implementation of their Privacy Sandbox technology FLoC (Federated Learning of Cohorts) which is
designed to deliver behavioural targeting without third-party cookies “might not be compatible with
European privacy law”
385
. In addition, some stakeholders and experts believe that Google’s approach is
likely to harm other players along the advertising value chain and will help to cement its current market
position, leading to lower levels of competition and creating greater dependencies. Although blocking
of third-party cookies is not a new approach, as Mozilla's Firefox and Apple's Safari browsers already
383
Autorité de la concurrence, 2018, Sector-specific investigation into online advertising; Srinivasan, D. (2020): Why Google Dominates
Advertising Markets – Competition Policy Should Lean on the Principles of Financial Market Regulation. Standford Technology Law
Review 24 (1): 55-175.
384
In particular, DMA proposal, Article5(e) and 5(f) as well as Article6(1)(d).
385
Schiff, A., 2021, Google Will Not Run FLoC Origin Tests In Europe Due To GDPR Concerns (At Least For Now), Adexchanger.
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
95 PE 662.913
block third-party cookies by default
386
, blocking them on Google Chrome may lead to distortions in the
market, as Google Chrome currently holds about 65% of the browser market
387
. In this context,
regulatory action may need to be considered.
5.3. Summary of conclusions
The following table summarises the problems discussed in this chapter and links the problems to
associated causes, as well as identifying potential solutions, in light of our analysis of the gaps
remaining after taking into account existing legislation, and proposals for the DSA and DMA.
In addition to the propositions linked to specific problems, it should be noted that there may also be
lessons to be learned from an institutional perspective from the experience of enforcing existing rules,
as shown in the case studies. Ireland, for example, demonstrates that the activities of the national
authority can be successfully complemented by self-regulatory procedures established by the industry
that helps to enable better and faster remedies. The establishment of a dual system could also be a
promising approach across Europe.
More generally, a greater level of harmonisation is needed at the EU level, especially with regard to the
enforcement of existing legal frameworks. In addition, national authorities should use similar concepts
and definitions for “advertising” and “commercial communication” and also “targeted advertising” in
order to maintain consistency.
386
Onlinemarketing.de, 2020, Welche Auswirkungen hat die Abschaffung von Third Party Cookies auf die Werbebranche?
387
Statcounter.com, 2021, Browser Market Share Worldwide – Statistic.
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 96
Table 9: Summary of conclusions
Problem Causes Objectives
Existing legal
instruments
Proposed solutions
(DMA/DSA)
Proposed further solutions
Consumers do not
know that their data
is being used to
target advertising.
Lack of transparency on
(1) ad targeting (and
person on whose behalf
the advertisement is
displayed) (2) main
parameters used for
targeting, (3) consent
given to use personal
data for ad targeting
More
transparency
and clearer
consent
mechanisms
(1) eCommerce and AVMS
Directives (2) CRD
amended and UCPD
amended (3) GDPR and e-
privacy : prior consent and
info
For (2) DSA Article 24;
but need to have clear
understanding on
which are the main
parameters
• Requirement to ensure transparency
concerning the existence of targeted
advertising (similar to ranking);
• How (format and location) to display
mandatory information requirements could
be specified in codes of conduct
•
Complementing Article 24 DSA with what is
foreseen in Article 29 on recommenders
systems for VLOPs, i.e. to give an option to
change the parameters used;
• Possibly, to encourage opt-in for targeted
advertising through self- or co-regulation
(provided in Article 36 DSA)
Algorithmic
discrimination
(e.g. based on
gender, disability)
Ability to intentionally
exploit personal data and
unintentional
algorithmic bias due to
biased training data
Limit
intentional and
unintentional
discrimination.
Artificial Intelligence Act
(AIA) proposes to address
algorithmic bias and to
provide transparency on
how the algorithm has
been trained for high risks
application; GDPR:
Explainability rule; AVMS:
non-discrimination on VSP
Not directly covered in
proposals
DSA covers all illegal
content (hence also
violations of EU and
national discrimination
rules) and has provision
regarding transparency
of profiling (Article 24)
Introduce further rules on the DSA by:
• Improving transparency (see above)
• Specifying in Recitals of DSA that all EU
instruments that prohibit certain forms of
advertising are in any event still applicable
as these are sector specific rules
• Also the obligations imposed in proposed
AIA should contribute to close the gap.
• Regular vetting of systems and training
data by accredited researchers will offer
new insights on how to mitigate systemic
risks and lead to lessened information
asymmetries.
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
97 PE 662.913
Problem Causes Objectives
Existing legal
instruments
Proposed solutions
(DMA/DSA)
Proposed further solutions
Privacy violations
Complex legal
framework; Problem of
enforcement
Clarity needed
on legal
framework and
better
enforcement
Provisions exist (GDPR and
e-privacy), but country of
origin principle/co-
ordination challenging.
Enforcement does not
always work
Not addressed in
proposals
•
Introduce in the GDPR the same
asymmetric enforcement mechanism as
foreseen in proposed DSA: Commission
may be in charge, next to the authority of
the country of establishment, for VLOPS.
Challenges for
consumers to seek
redress in case of
illegal advertising
Advertising can be illegal
because it is in breach of
different types of rules
(consumer protection,
privacy etc.)
Provide clarity
for consumers
on where to
seek redress
GDPR, national rules,
procedures in case of
cross-border disputes, ADR
Not addressed except
for in proposed DSA
rules on flagging,
internal complaints
handling, and out of
court dispute resolution
but application is not
clear in case of illegal
advertising
•
Address in more detail how sector specific
rules will interact with the DSA, especially
on enforcement
• Include among the list of tasks for the
Digital Service Coordinators the need to
provide information to consumers on how
to seek redress in relation to online
advertising (among other areas)
• Envisage the adoption of a sector specific
directive to address all the consumer
protection issues linked to online
advertising, thereby ending a piecemeal
approach which is not necessarily
conducive to consumer trust
Dark patterns
(1) Lack of guidelines
(2) Country of origin
principle impedes
enforcement
(3) Insufficient incentive
for local authority to
intervene
Freely given
consent
needed to
process data
Guidelines from EDPB, but
only soft law
Not addressed in
proposals
Amend the DSA:
• Develop design guidelines on consent
forms, improve enforcement (European
authority),
• Legal status of Guidelines - to improve
consideration by judge/national authority
• Interface with opt-out; provide tool to
enable consumers to report websites
infringing guidelines
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 98
Problem Causes Objectives
Existing legal
instruments
Proposed solutions
(DMA/DSA)
Proposed further solutions
Inappropriate
targeting of minors
Insufficient legal
framework
Protect minors
GDPR: children cannot
give consent - AVMSD for
VSPs: specific protection
for minors
Not addressed in
proposals
•
Include in DSA similar provisions as in
Article
28(b)(2) AVMS for VSP: prohibition to
target ads to minors
Multiple actors
involved (publishers,
intermediators,
advertisers), creates a
possible dilution of
responsibility
Linked to nature of the
market
Make sure that
the rights of
consumers and
legal
requirements
are respected
and endorsed
by all actors in
the value chain
Place more
responsibility
on platforms
GDPR (notion of joint
controller)
AVMS, DSA, CDR, UCPD
Any legal obligation
contained in the DSA
should be extended to
other
websites/publishers
(not in scope of DSA)
and to ad
intermediaries
• Make sure that the personal scope of
application of the rules fits by possibly
proposing legislation to mirror obligations
of the DSA for websites/publishers and ad
sector
• Clarify that Article 5.3 of the DSA should
also apply when consumers are led to
believe that the platform is organising the
advertising itself. This will provide an
incentive for platforms to conform with the
rules
Potential for
gatekeepers to
exploit market power
and to raise prices
and limit choice
Concentration of
intermediaries -
dependence by SMEs on
largest distributors
Potential to
intervene to
address
exploitative
conduct
Not addressed
Transparency on price
(Article 5g) and
performance in DMA for
gatekeeper platforms -
to be given to
advertiser
•
Adequately addressed in DMA
• Case by case intervention under
competition law, where needed
Ad fraud
Lack of information
about source of
clicks/revenues
Transparency
about source of
clicks/revenues
Not addressed
Access to performance
measuring tool in DMA
(Article 6.1g)
•
Adequately addressed in DMA
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
99 PE 662.913
Problem Causes Objectives
Existing legal
instruments
Proposed solutions
(DMA/DSA)
Proposed further solutions
Ad auction
algorithms could
disadvantage smaller
players / make it
more expensive to
launch new services
Lack of transparency in
ad auctions / profit
maximisation focus
Transparency in
the auction
system /
potential to
intervene if it
has the effect of
excluding SMEs
Not addressed
Not entirely covered by
the DMA
Price determination
process (including role
of quality) for auctions
not covered
Add to the proposed transparency obligations
in the DMA:
• Auction criteria used by the ad-tech
platform service that include details of the
price components (CPC, CPI, CPM) as well
as other factors which are taken into
account in the auction process and their
weighting.
• Obligations to provide B2B side
performance data
Asymmetric access to
data favours large
intermediaries over
competing ad
networks and
publishers
Large customer base and
data sets controlled by
gatekeeper platforms /
network effects
Level playing
field for
opportunities in
advertising
Not addressed
Data Governance Act
will not address data
sharing. DMA has some
data provisions for
gatekeepers (for
search). Forthcoming
Data Act may include
some data sharing
provisions. DMA self-
preferencing
prohibition (Article
6,1d) may help in some
cases. P2B Regulation
addresses need to be
transparent in T&C on
data sharing situation
•
To be further analysed
• DMA may help advertising linked to search,
but problem not wholly solved.
May be partly addressed if targeted
advertising declines in importance.
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 100
Problem Causes Objectives
Existing legal
instruments
Proposed solutions
(DMA/DSA)
Proposed further solutions
Lack of clarity on
functioning of
internal market
clause in relation to
advertising
Conditions for
derogations are wide and
subject to case by case
scrutiny
Create more
legal certainty
eCommerce Directive
Not addressed in
proposals
•
Amend eCommerce Directive (possibly
through DSA) by introducing more
effective procedural rules to derogation
procedure of Article 3
• Through EU-wide codes of conduct list
types of national derogations that could be
applied to cross-border advertising. These
codes could be ‘vetted’ by the European
Commission
Source: Authors’ own elaboration.
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
101 PE 662.913
REFERENCES
• ACCC, 2013, Google appeal upheld, media release. Available at: https://www.accc.gov.au/media-
release/google-appeal-upheld. [last accessed: 20.04.2021].
• ACCC, 2019, Digital Platforms Inquiry: Final Report. Available at:
https://www.accc.gov.au/system/files/Digital%20platforms%20inquiry%20%20final%20report.pdf.
[last accessed: 12.04.2021].
• ACCC, 2020, ACCC alleges Facebook misled consumers when promoting app to 'protect' users' data,
media release. Available at:
https://www.accc.gov.au/media-release/accc-alleges-facebook-misled-
consumers-when-promoting-app-to-protect-users-data.
• ACCC, 2020, ACCC alleges Google misled consumers about expanded use of personal data, media release.
Available at:
https://www.accc.gov.au/media-release/correction-accc-alleges-google-misled-
consumers-about-expanded-use-of-personal-data. [last accessed: 22.04.2021].
• ACCC, 2020, Trivago misled consumers about hotel room rates, media release. Available at:
https://www.accc.gov.au/media-release/trivago-misled-consumers-about-hotel-room-rates
.
[last accessed: 18.04.2021].
• ACCC, 2021, Digital advertising services inquiry. Available at:
https://www.accc.gov.au/focus-
areas/inquiries-ongoing/digital-advertising-services-inquiry. [last accessed: 21.04.2021].
• ACCC, 2021, Digital platform services inquiry 2020-2025. Available at: https://www.accc.gov.au/focus-
areas/inquiries-ongoing/digital-platform-services-inquiry-2020-2025. [last accessed: 20.04.2021].
• ACCC, 2021, Lack of competition in ad tech affecting publishers, advertisers and consumers. Available at:
https://www.accc.gov.au/media-release/lack-of-competition-in-ad-tech-affecting-publishers-
advertisers-and-consumers. [last accessed: 20.04.2021].
• ACCC, 2021, News media bargaining code. Available at: https://www.accc.gov.au/focus-areas/digital-
platforms/news-media-bargaining-code/final-legislation. [last accessed: 22.04.2021].
• AdNow, 2021, Website. Available at: https://adnow.com/about [last accessed: 26.05.2021].
• Advertising Standards Agency for Ireland, 2020, Complaints Bulletins. Available at:
https://www.asai.ie/complaints-bulletins/
. [last accessed: 12.04.2021].
• Advertising Standards Agency for Ireland, ASAI Code, 7
th
edition. Available at:
https://www.asai.ie/asaicode/
[last accessed: 19.04.2021].
• Advertising Standards Agency, 2014, CAP Code: Preface. Available at:
https://www.asa.org.uk/type/non_broadcast/code_folder/preface.html
. [last accessed: 13.04.2021].
• Advertising Standards Agency, 2014, Non-broadcast Code. Available at: https://www.asa.org.uk/codes-
and-rulings/advertising-codes/non-broadcast-code.html [last accessed: 13.04.2021].
• Advertising Standards Agency, 2014, Scope of the Code: CAP Code. Available at:
https://www.asa.org.uk/type/non_broadcast/code_folder/scope-of-the-code.html
. [last accessed:
19.04.2021].
• Advertising Standards Agency, The CAP Code: How the System Works. Available at:
https://www.asa.org.uk/type/non_broadcast/code_section/how-the-system-works.html
. [last
accessed: 20.04.2021].
• Advertising Standards Authority for Ireland, 2016, Code, 7th edition.
Available at: https://www.asai.ie/asaicode/
[last accessed: 20.04.2021].
• Advertising Standards Authority for Ireland, 2016, Code, 7th edition Section 18.1. Available at:
https://www.asai.ie/asaicode/section-18-online-behavioural-advertising/
. [last accessed: 21.04.2021].
• Advertising Standards Authority for Ireland, 2016, Code, 7th edition: Section 1.1(d). Available at:
https://www.asai.ie/asaicode/section-1-definitions/
.
• Advertising Standards Authority for Ireland, 2021, #InfluencerMarketing in 2021 and beyond, Webinar,
23
rd
February 2021. Available at: https://www.asai.ie/news/influencermarketing-in-2021-and-beyond-
2/. [last accessed: 27.04.2021].
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 102
• Aggarwal, G., Bursztein, E., Jackson, C., & Boneh, D., 2010, ‘An Analysis of Private Browsing Modes in
Modern Browsers’, Proc. of 19
th
Usenix Security Symposium (p. 79-93).
• Aguirre, E., Mahr, D., Grewal, D., de Ruyter, K. & M. Wetzels, 2015, ‘Unraveling the Personalization
Paradox: The Effect of Information Collection and Trust-Building Strategies on Online Advertisement
Effectiveness’, Journal of Retailing, 91 (1):34-49.
• Amiot, E., 2020, Digital Advertising in France Weakened by the Pandemic, Syndicat des Régies Internet
(SRI). Available at:
https://www.oliverwyman.com/our-expertise/insights/2020/sep/digital-advertising-
in-france-weakened-by-the-pandemic.html. [last accessed: 12.04.2021].
• Anderson, S. P., & Gans, J. S., 2011, ‘Platform siphoning: Ad-avoidance and media content’, American
Economic Journal: Microeconomics, 3(4), 1-34.
• Article 29 Data Protection Working Party, 2017, Guidelines on Data Protection Impact Assessment (DPIA)
and determining whether processing is “likely to result in a high risk” for the purposes of Regulation 2016/679,
4 October 2017, WP 248rev.01.
• Article 29 Data Protection Working Party, 2018, Guidelines on Automated individual decision-making and
Profiling for the purposes of Regulation 2016/679, 3 October 2018, WP251rev.01, p. 15.
• Article 29 Data Protection Working Party, 2010, Opinion 2/2010 on online behavioural advertising, 22 June
2010, WP171.
• Aseri, M., Dawande, M., Janakiraman, G., & S. Mookerjee, V., 2020, ‘Ad-Blockers: A Blessing or a Curse?’,
Information Systems Research, 31 (2) 627-646.
• Australian Government, 2018, Parliament passes News Media and Digital Platforms Mandatory Bargaining
Code, media release. Available at:
https://ministers.treasury.gov.au/ministers/josh-frydenberg-
2018/media-releases/parliament-passes-news-media-and-digital-platforms.
[last accessed: 20.04.2021].
• Autorité de la concurrence, 2010, No 10-A-29 Opinion.
• Autorité de régulation professionnelle de la publicité, 2015, Recommandations on digital commercial
communications. Available at:
https://www.arpp.org/nous-consulter/regles/regles-de-
deontologie/digital-advertising-and-marketing-communications-code/. [last accessed: 26.04.2021].
• Autorité de la concurrence, 2016, Decision No 16-D-25 of 23 November 2016 on practices implemented in
the online advertising sector online advertising.
Available at: https://www.autoritedelaconcurrence.fr/sites/default/files/commitments//16d25.pdf
. [last
accessed: 09.04.2021].
• Autorité de la concurrence, 2017, Decision 17-D-24 of 18 December 2017 concerning practices
implemented in the online search engine and online advertising intermediation sectors. Available at:
https://www.autoritedelaconcurrence.fr/fr/decision/relative-des-pratiques-mises-en-oeuvre-dans-le-
secteur-des-moteurs-de-recherche-en-ligne. [last accessed: 09.04.2021].
• Autorité de la concurrence, 2018, Opinion n° 18-A-03 of 6 March 2018 on the use of data in the internet
advertising sector advertising sector. 2018.
Available at: https://www.autoritedelaconcurrence.fr/sites/default/files/commitments//18a03.pdf
. [last
accessed: 09.04.2021].
• Autorité de la concurrence, 2018, Sector-specific investigation into online advertising. Available at:
https://www.autoritedelaconcurrence.fr/en/press-release/6-march-2018-sector-specific-investigation-
online-advertising. [last accessed: 12.04.2021].
• Autorité de la concurrence, 2019, Decision 19-D-26 of 20 December 2019 regarding practices employed in
the online search advertsing sector.
Available at: https://www.autoritedelaconcurrence.fr/en/decision/regarding-practices-implemented-
sector-online-search-advertising-sector [last accessed: 26.05.2021].
• Bbb National Programs, National Advertising Division, Information Page. Available at:
https://bbbprograms.org/programs/all-programs/national-advertising-division
. [last accessed:
16.04.2021].
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
103 PE 662.913
• BEUC, 2017, Proposal for a regulation on Privacy and electronic communication (ePrivacy) – BEUC position
paper, BEUC-X-2017-059, 09/06/2017.
• Blake, T., Nosko, C. & Tadelis, S., 2015, ‘Consumer Heterogeneity and Paid Search Effectiveness: A Large‐
Scale Field Experiment‘, Econometrica, 83: 155-174.
• Bleier, A., & Eisenbeiss, M., 2015, ‘Personalized online advertising effectiveness: The interplay of what,
when, and where’, Marketing Science, 34(5), 669-688.
• Bleier, A., & M. Eisenbeiss, 2015, ‘The Importance of Trust for Personalized Online Advertising’, Journal of
Retailing, 91 (3):390-409.
• Bösch, Ch., Erb, B., Kargl, F., Kopp, H. & S. Pfattheicher, 2016, Tales from the Dark Side: Privacy Dark
Strategies and Privacy Dark Patterns, Proceedings on Privacy Enhancing Technologies. 2016 (4): 237–254.
• Bourguignon, C., 2019, Les réglementations des professionnels de la santé revues à l’aune des libertés de
circulation économiques : l’exemple de la publicité à l’ère du numérique, R.D.T.I., 2019/1, n° 74.
• Bourguignon, C., 2020, La vente en ligne transfrontalière de médicaments au sein de l’Union européenne,
D.C.C.R., 2020/2, n° 127.
• Brave, 2020, Europe’s governments are failing the GDPR – Brave’s 2020 report on the enforcement capacity
of data protection authorities. Available at:
https://brave.com/wp-content/uploads/2020/04/Brave-
2020-DPA-Report.pdf.
• Brazzoni, A., 2020, 71% of Publishers Seeking to Diversify Revenue After COVID: Here Are the Models They're
Considering, lineup. Available at: https://www.lineup.com/newsroom/industry-analysis/publishers-
diversify-revenue. [last accessed: 12.04.2021].
• Brignull, H., 2021, What are dark patterns. Available at: www.darkpatterns.org. [last accessed:
20.05.2021].
• Briz, N., 2018, This is Your Digital Fingerprint, Mozilla Blog.
Available at: https://blog.mozilla.org/internetcitizen/2018/07/26/this-is-your-digital-fingerprint/
. [last
accessed: 26.05.2021].
• Bundeskartellamt, 2018, Online advertising. Series of papers on “Competition and Consumer Protection in
the Digital Economy”.
Available at:
https://www.bundeskartellamt.de/SharedDocs/Publikation/EN/Schriftenreihe_Digitales_III.pdf?__blob
=publicationFile&v=5. [last accessed: 12.04.2021].
• Buse, N., 2021, Onlinewerbung- was ist erlaubt, was ist verboten?, Buse Herz Grunst Rechtsanwälte.
Available at:
https://www.kanzlei-wirtschaftsrecht.berlin/aktuelles/onlinewerbung-was-ist-erlaubt-
was-ist-verboten/. [last accessed: 07.04.2021].
• Business-Wissen, 2008, Einheitliche Werberegeln für die ganze EU. Available at: https://www.business-
wissen.de/artikel/uwg-novelle-einheitliche-werberegeln-fuer-die-ganze-eu/.
[last accessed: 20.04.2021].
• Cai, Y., Yee, G. O. M., Gu, Y. X., & C.-H. Lung, 2020, ‘Threats to Online Advertising and Countermea-sures:
A Technical Survey’, Digit. Threat.: Res. Pract. 1, 2, Article 11.
• Cai, Y., Yee, G. O., Gu, Y. X., & Lung, C. H., 2020, ‘Threats to online advertising and countermeasures: A
technical survey’, Digital Threats: Research and Practice, 1(2), 1-27.
• Cabral, L., Haucap, J., Parker, G., Petropoulos, G., Valletti, T. and Van Alstyne, M., 2021, The EU Digital
Markets Act, Publications Office of the European Union, Luxembourg, ISBN 978-92-76-29788-8 (online).
• Chellappa, R. K., & R. G. Sin, 2005, ‘Personalization versus Privacy: An Empirical Examination of the Online
Consumer’s Dilemma’, Information Technology and Management, 6 (2):181-202.
• Chen, J., & Stallaert, J., 2014, ‘An economic analysis of online advertising using behavioral targeting’, Mis
Quarterly, 38(2), 429-A7.
• Clifford, D., 2014, ‘EU Data Protection Law and Targeted Advertising: Consent and the Cookie Monster -
Tracking the crumbs of online user behaviour’, JIPITEC, pp. 194 – 195.
Available at: https://www.jipitec.eu/issues/jipitec-5-3-2014/4095
. [last accessed: 20.02.2021].
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 104
• CMA, 2020, Online platforms and digital advertising: Market study final report.
Appendix M: intermediation in open display advertising.
Available at:
https://assets.publishing.service.gov.uk/media/5fe495c28fa8f56afaf406d4/Appendix_M_-
_intermediation_in_open_display_advertising_WEB.pdf. [last accessed: 20.04.2021].
• CMA, 2020, Online platforms and digital advertising: Market study final report. Available at:
https://assets.publishing.service.gov.uk/media/5fa557668fa8f5788db46efc/Final_report_Digital_ALT_
TEXT.pdf. [last accessed: 20.04.2021].
• CMA, 2021, Investigation into Google’s ‘Privacy Sandbox’ browser changes. Available at:
https://www.gov.uk/cma-cases/investigation-into-googles-privacy-sandbox-browser-changes [last
accessed: 26.05.2021].
• Competition and Consumer Protection Commission, 2016, Selling Online: What you need to know,
Business Guide. Available at:
https://www.ccpc.ie/business/wp-
content/uploads/sites/3/2016/08/CCPC-Guide-for-selling-online-final.pdf. [last accessed: 15.04.2021].
• Competition and Consumer Protection Commission, 2019, Annual Report 2019. Available at:
https://www.ccpc.ie/business/wp-content/uploads/sites/3/2020/07/CCPC-Annual-Report-2019.pdf
.
[last accessed: 22.04.2021].
• Competition and Consumer Protection Commission, 2020, Strategy Statement 2021-2023. Available at:
https://www.ccpc.ie/business/wp-content/uploads/sites/3/2020/12/CCPC-Strategy-Statement-2021-
2023.pdf. [last accessed: 20.04.2021].
• Competition and Consumer Protection Commission, 3 November 2020, Consumer warning about
irelandsstore .com, News. Available at:
https://www.ccpc.ie/business/consumer-warning-about-
irelandsstore-com/. [last accessed: 15.04.2021].
• Competition and Markets Authority, 2020, Online platforms and digital advertising: A Study. Available at:
https://www.gov.uk/cma-cases/online-platforms-and-digital-advertising-market-study
. [last accessed:
12.04.2021].
• Comstock, M., 2020, COVID-19 and Programmatic Advertising, PMG. Available at:
https://www.pmg.com/blog/covid-19-and-programmatic-advertising/
. [last accessed: 12.04.2021].
• Constantin, L., 2016, The AdGholas malvertising campaign infected thousands of computers per day.
Available at:
http://www.computerworld.com/article/3101823/security/the-adgholas-malvertising-
campaign-infected-thousands-of-computers-per-day.html.
• Consumer Focus, 2008, Waiting to be heard: Giving consumers the right of redress over Unfair Commercial
Practices.
• Consumer Policy Research Centre, 2020, Unfair trading practices in digital markets – evidence and
regulatory gaps, Research and policy briefing. Available at:
https://cprc.org.au/app/uploads/2021/03/Unfair-Trading-Practices-in-Digital-Markets.pdf
.
[last accessed: 12.04.2021].
• Council of the European Union, 2000, Council Directive 2000/43/EC of 29 June 2000 implementing the
principle of equal treatment between persons irrespective of racial or ethnic origin, OJ L 180 of 19.7.2000.
• Council of the European Union, 2000, Council Directive 2000/78/EC of 27 November 2000 establishing a
general framework for equal treatment in employment and occupation, OJ L 303 of 02.12.2000.
• Council of the European Union, 2004, Council Directive 2004/113/EC of 13 December 2004 implementing
the principle of equal treatment between men and women in the access to and supply of goods and services,
OJ L 373 of 21.12.2004.
• Criteo, 2021, State of Digital Advertising 2021: i budget e le priorità del marketing in un mondo nuovo.
Available at:
https://www.criteo.com/it/wp-content/uploads/sites/9/2021/02/Criteo-REPORT-State-of-
Digital-Advertising-2021-Global-IT.pdf. [last accessed: 20.04.2021].
• Criteo, 2021, Website. Available at: https://www.criteo.com/ [last accessed: 26.05.2021].
• Datanyze, 2020, Advertising Networks, Available at: https://www.datanyze.com/market-
share/advertising-networks--9. [last accessed: 12.04.2021].
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
105 PE 662.913
• Datanyze, 2021, Market Share Category – Ad Exchanges Market Share, Available at:
https://www.datanyze.com/market-share/ad-exchanges--39. [last accessed: 21.05.2021].
• Dave, V., Guha, S., & Y. Zhang, 2013, ‘Viceroi: Catching click-spam in search ad networks’, Proceedings of
the ACM SIGSAC Conference on Computer and Communications Security. ACM, 765–776.
• Dehling, T., Zhang, Y. & Sunyaev, A., 2019, ‘Consumer Perceptions of Online Behavioral Advertising’,
Proceedings of the 21st IEEE Conference on Business Informatics.
• Department for Business, Energy & industrial Strategy, 2018, Misleading and aggressive commercial
practices: new private rights for consumers. Available at:
https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/
721872/misleading-aggressive-commercial-practices-guidance.pdf [last accessed: 20.04.2021].
• Despotakis, S., Ravi, R., & Sayedi, A., 2019, First-price auctions in online display advertising. Available at:
SSRN: https://ssrn.com/abstract=3485410 or http://dx.doi.org/10.2139/ssrn.3485410
. [last accessed:
25.05.2021].
• Despotakis, S., Ravi, R., & Srinivasan, K., 2020, ‘The beneficial effects of ad blockers’, Management Science.
64 (4).
• DEUS Marketing GmbH, 2020, Welche Auswirkungen hat die Abschaffung von Third Party Cookies auf die
Werbebranche?, OnlineMarketing.de.
Available at: https://onlinemarketing.de/unternehmensnews/google-third-party-cookies-blockieren-
folgen. [last accessed: 13.04.2021].
• Dingledine, R., Mathewson, N., & Syverson, P., 2004, Tor: The Second-Generation Onion Router. Available
at:
http://citeseerx.ist.psu.edu/viewdoc/download;jsessionid=D21FDB1A9E5E9262AAF9F9FA185AC2A5?
doi=10.1.1.144.2492&rep=rep1&type=pdf. [last accessed: 20.05.2021].
• Duch-Brown, N. & B. Martens, 2016, The Economic Impact of Removing Geo-blocking Restrictions in the EU
Digital Single Market. Institute for Prospective Technological Studies Digital Economy Working Paper
2016/02. JRC Technical Reports. Available at:
https://ec.europa.eu/jrc/sites/jrcsh/files/JRC101100.pdf.[last accessed: 16.02.2021].
• Econsultancy, & Upland Adestra, 2019, 2019 Email Marketing Industry Census. Available at:
https://content.adestra.com/hubfs/2019_Reports/2019%20Email%20Marketing%20Census.pdf
. [last
accessed: 19.05.2021].
• Edelman, B., Ostrovsky, M., & Schwarz, M., 2007, ‘Internet advertising and the generalized second-price
auction: Selling billions of dollars worth of keywords’, American economic review, 97(1), 242-259.
• EDRi, 2019, Real Time Bidding: The auction for your attention. Available at:
https://edri.org/our-
work/real-time-bidding-the-auction-for-your-attention/. [last accessed: 19.05.2021].
• eMarketer.com, 2020, Ad Blocking Is Slowing Down, but Not Going Away. Available at:
https://www.emarketer.com/content/ad-blocking-growth-is-slowing-down-but-not-going-away
[last
accessed: 21.05.2021].
• European Advertising Standards Alliance, 2021, Members: Germany. Available at: https://www.easa-
alliance.org/members/europe/germany. [last accessed: 23.04.2021].
• European Commission, 2014, DG Justice Guidance Document concerning Directive 2011/83/EU of the
European Parliament and of the Council of 25 October 2011 on consumer rights, amending Council Directive
93/13/EEC and Directive 1999/44/EC of the European Parliament and of the Council and repealing Council
Directive 85/577/EEC and Directive 97/7/EC of the European Parliament and of the Council, June 2014.
Available at: https://ec.europa.eu/info/sites/default/files/crd_guidance_en_0.pdf [last accessed:
26.05.2021.
• European Commission, 2016, Commission Staff Working Document, Guidance on the
implementation/application of Directive 2005/29/EC on unfair commercial practices, SWD(2016) 163 final.
• European Commission, 2017, Proposal for a Regulation of the European Parliament and of the Council
concerning the respect for private life and the protection of personal data in electronic communications and
repealing Directive 2002/58/EC, COM(2017) 10 final - 2017/0003 (COD).
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 106
• European Commission, 2018, Consumer market study on online market segmentation through
personalised pricing/offers in the European Union. Available at:
https://ec.europa.eu/info/publications/consumer-market-study-online-market-segmentation-
through-personalised-pricing-offers-european-union_en. [last accessed: 21.05.2021].
• European Commission, 2019, Antitrust: Commission fines Google €1.49 billion for abusive practices in
online advertising.
Available at. https://ec.europa.eu/commission/presscorner/detail/en/IP_19_1770
. [last accessed:
12.04.2021].
• European Commission, 2020, Proposal of the Commission of 15 December 2020 for a Regulation of the
European Parliament and of the Council on a Single Market For Digital Services (Digital Services Act) and
amending Directive 2000/31/EC, COM(2020) 825 final, 2020/0361 (COD).
• European Commission, 2020, Proposal of the Commission of 15 December 2020 for a Regulation of the
European Parliament and of the Council on contestable and fair markets in the digital sector (Digital Markets
Act), COM(2020) 842 final, 2020/0374 (COD).
• European Commission, 2020, Communication from the Commission to the European Parliament and
the Council, New Consumer Agenda - Strengthening consumer resilience for sustainable recovery. Available
at:
https://eur-lex.europa.eu/legal-
content/EN/TXT/?uri=CELEX%3A52020DC0696&qid=1605887353618. [last accessed: 21.04.2021].
• European Commission, 2020, Guidelines of 7 December 2020 on ranking transparency pursuant to
Regulation (EU) 2019/1150 of the European Parliament and of the Council, OJ C 424/1 of 8.12.2020.
• European Commission, 2021, Proposal for a regulation of the European Parliament and of the Council
laying down harmonised rules on artificial intelligence (artificial intelligence act) and amending certain
Union legislative acts, COM(2021) 206 final 2021/0106 (COD).
• European Data Protection Board, 2019, Guidelines 2/2019 on the processing of personal data under Article
6(1)(b) GDPR in the context of the provision of online services to data subjects, 8 October 2019. Available at:
https://edpb.europa.eu/sites/default/files/files/file1/edpb_guidelines-art_6-1-b-
adopted_after_public_consultation_en.pdf . [last accessed: 25.05.2021].
• European Data protection Board, 2019, Opinion 5/2019 on the interplay between the ePrivacy Directive and
the GDPR, in particular regarding the competence, tasks and powers of data protection authorities, 12 March
2019. Available at
https://edpb.europa.eu/sites/default/files/files/file1/201905_edpb_opinion_eprivacydir_gdpr_interpl
ay_en_0.pdf [last accessed: 25.05.2021].
• European Data Protection Board, 2019, Guidelines2/2019 on the processing of personal data under Article
6(1)(b) GDPR in the context of the provision of online services to data subjects, 8 October2019. Available at:
https://edpb.europa.eu/sites/default/files/files/file1/edpb_guidelines-art_6-1-b-
adopted_after_public_consultation_en.pdf. [last accessed: 19.05.2021].
• European Data Protection Board, 2020, Guidelines 8/2020 on the targeting of social media users, 2
September 2020. Available at: https://edpb.europa.eu/our-work-tools/documents/public-
consultations/2020/guidelines-082020-targeting-social-media-users_en. [last accessed: 25.05.2021].
• European Data Protection Board, 2020,
Guidelines 05/2020 on consent under Regulation 2016/679, 4 May 2020, p. 7. Available at:
https://edpb.europa.eu/sites/default/files/files/file1/edpb_guidelines_202005_consent_en.pdf. [last
accessed: 18.05.2021].
• European Data Protection Board, 2020, Guidelines 07/2020 on the concepts of controller and processor in
the GDPR, 02 September 2020. p. 22. Available at:
https://edpb.europa.eu/sites/default/files/consultation/edpb_guidelines_202007_controllerprocessor
_en.pdf. [last accessed: 26.05.2021].
• European Data Protection Board, 2021, Guidelines 02/2021 on Virtual Voices Assistants, 9 March 2021,
Available at: https://edpb.europa.eu/our-work-tools/documents/public-
consultations/2021/guidelines-022021-virtual-voice-assistants_en. [last accessed: 25.05.2021].
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
107 PE 662.913
• European Data protection Board, 2021, Guidelines 01/2020 on processing personal data in the context of
connected vehicles and mobility related applications, adopted on 9 March 2021.
Available at: https://edpb.europa.eu/system/files/2021-
03/edpb_guidelines_202001_connected_vehicles_v2.0_adopted_en.pdf. [last accessed: 19.05.2021].
• European Parliament and the Council of the European Union, 1998, Directive 98/34/EC of the European
Parliament and of the Council of 22 June 1998 laying down a procedure for the provision of information in
the field of technical standards and regulations, OJ L 204 of 21.07.1998.
• European Parliament and the Council of the European Union, 1998, Directive 98/48/EC of the European
Parliament and of the Council of 20 July 1998 amending Directive 98/34/EC laying down a procedure for the
provision of information in the field of technical standards and regulations, OJ L 217 of 5.08.1998.
• European Parliament and the Council of the European Union, 2000, Directive 2000/31/EC of the European
Parliament and of the Council of 8 June 2000 on certain legal aspects of information society services, in
particular electronic commerce, in the Internal Market, OJ L 178 of 17.07.2000.
• European Parliament and the Council of the European Union, 2002, Directive 2002/58/EC of the European
Parliament and of the Council of 12 July 2002 concerning the processing of personal data and the protection
of privacy in the electronic communications sector, OJ L 201 of 31.7.2002.
• European Parliament and the Council of the European Union, 2003, Directive 2003/33/EC of the European
Parliament and of the Council of 26 May 2003 on the approximation of the laws, regulations and
administrative provisions of the Member States relating to the advertising and sponsorship of tobacco
products, OJ L 152 of 20.6.2003.
• European Parliament and the Council of the European Union, 2005, Directive 2005/29/EC of the European
Parliament and of the Council of 11 May 2005 concerning unfair business-to-consumer commercial practices
in the internal market and amending Council Directive 84/450/EEC, Directives 97/7/EC, 98/27/EC and
2002/65/EC of the European Parliament and of the Council and Regulation (EC) No 2006/2004 of the
European Parliament and of the Council (‘Unfair Commercial Practices Directive’), OJ L 149 of 11.6.2019.
• European Parliament and the Council of the European Union, 2005, Directive 2005/29 of the European
Parliament and of the Council of 11 May 2005 concerning unfair business-to-consumer commercial practices
in the internal market, OJ L 149 of 11.06.2005.
• European Parliament and the Council of the European Union, 2005, Directive 2005/36/CE of the European
Parliament and of the Council of 7 September 2005 on the recognition of professional qualifications, OJ L
255 of 30.9.2005.
• European Parliament and the Council of the European Union, 2005, Directive 2005/36/EC of the European
Parliament and of the Council of 7 September 2005 on the recognition of professional qualifications, OJ L
255 of 30.09.2005.
• European Parliament and the Council of the European Union, 2006, Directive 2006/54/EC of the European
Parliament and of the Council of 5 July 2006 on the implementation of the principle of equal opportunities
and equal treatment of men and women in matters of employment and occupation (recast), OJ L 204 of
26.7.2006.
• European Parliament and the Council of the European Union, 2006, Directive 2006/114/EC of the
European Parliament and of the Council of 12 December 2006 concerning misleading and comparative
advertising (codified version), OJ L 376 of 27.12.2006.
• European Parliament and the Council of the European Union, 2006, Directive 2006/123/EC of the
European Parliament and of the Council of 12 December 2006 on services in the internal market, OJ L 376 of
27.12.2006.
• European Parliament and the Council of the European Union, 2010, Directive 2010/13/EU of the European
Parliament and of the Council of 10 March 2010 on the coordination of certain provisions laid down by law,
regulation or administrative action in Member States concerning the provision of audiovisual media services
(Audiovisual Media Services Directive) (codified version), OJ L 095 of 15.4.2010.
• European Parliament and the Council of the European Union, 2011, Directive 2011/83 of the European
Parliament and of the Council of 25 October 2011 on consumer rights, OJ L 304 of 22.11.2011.
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 108
• European Parliament and the Council of the European Union, 2013, Directive 2013/55/EU of the European
Parliament and of the Council of 20 November 2013 amending Directive 2005/36/EC on the recognition of
professional qualifications and Regulation (EU) No 1024/2012 on administrative cooperation through the
Internal Market Information System (‘the IMI Regulation’), OJ L 354 of 28.12.2013.
• European Parliament and the Council of the European Union, 2013, Directive 2013/11/EU of the European
Parliament and of the Council of 21 May 2013 on alternative dispute resolution for consumer disputes and
amending Regulation (EC) No 2006/2004 and Directive 2009/22/EC (Directive on consumer ADR), OJ L 165
of 18.6.2013.
• European Parliament and the Council of the European Union, 2013, Regulation (EU) No 524/2013 of the
European Parliament and of the Council of 21 May 2013 on online dispute resolution for consumer disputes
and amending Regulation (EC) No 2006/2004 and Directive 2009/22/EC (Regulation on consumer ODR), OJ
L 165 of 18.6.2013.
• European Parliament and the Council of the European Union, 2015, Directive (EU) 2015/1535 of the
European Parliament and of the Council of 9 September 2015 laying down a procedure for the provision of
information in the field of technical regulations and of rules on Information Society services, OJ L 241,
17.9.2015
• European Parliament and the Council of the European Union, 2016, Regulation 2016/679 of the European
Parliament and of the Council of 27 April 2016 on the protection of natural persons with regard to the
processing of personal data and on the free movement of such data, OJ L 119 of 04.05.2016.
• European Parliament and the Council of the European Union, 2017, Regulation (EU) 2017/2394 of the
European Parliament and of the Council of 12 December 2017 on cooperation between national authorities
responsible for the enforcement of consumer protection laws and repealing Regulation (EC) No 2006/2004,
OJ L 345 of 27.12.2017.
• European Parliament and the Council of the European Union, 2018, Regulation (EU) 2018/302 of the
European Parliament and of the Council of 28 February 2018 on addressing unjustified geo-blocking and
other forms of discrimination based on customers' nationality, place of residence or place of establishment
within the internal market and amending Regulations (EC) No 2006/2004 and (EU) 2017/2394 and Directive
2009/22/EC, OJ L 60I of 02.03.2018.
• European Parliament and the Council of the European Union, 2018, Directive (EU) 2018/1808 of the
European Parliament and of the Council of 14 November 2018 in view of changing market realities, OJ L
303/69 of 28.11.2018
• European Parliament and the Council of the European Union, 2018, Directive (EU) 2018/958 of the
European Parliament and of the Council of 28 June 2018 on a proportionality test before adoption of new
regulation of professions, OJ L 173 of 9.7.2018.
• European Parliament and the Council of the European Union, 2019, Regulation (EU) 2019/1150 of the
European Parliament and of the Council of 20 June 2019 on promoting fairness and transparency for
business users of online intermediation services, OJ L 186 of 11.7.2019.
• European Parliament and the Council of the European Union, 2019, Directive (EU) 2019/790 of the
European Parliament and of the Council of 17 April 2019 on copyright and related rights in the Digital Single
Market and amending Directives 96/9/EC and 2001/29/EC (Text with EEA relevance.), OJ L 130 of 17.5.2019.
• European Parliament and the Council of the European Union, 2019, Directive 2019/2161 of the European
Parliament and of the Council of 27 November 2019 amending Council Directive 93/13/EEC and Directives
98/6/EC, 2005/29/EC and 2011/83/EU of the European Parliament and of the Council as regards the better
enforcement and modernisation of Union consumer protection rules, OJ L 328 of 18.12.2019.
• European Parliament and the Council of the European Union, 2020, Directive (EU) 2020/1828 of 25
November 2020 on representative actions for the protection of the collective interests of consumers and
repealing Directive 2009/22/EC, OJ L 409 of 4.12.2020.
• European Parliamentary Research Service, 2020, Digital Service Act: European Added Value Assessment.
Available at:
https://www.europarl.europa.eu/RegData/etudes/STUD/2020/654180/EPRS_STU(2020)654180_EN.pdf
.
[last accessed: 22.04.2021].
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
109 PE 662.913
• eyeo.com, 2020, Ad-blocking usage report 2019, 2019 US ad-blocking usage report. Available at:
https://eyeo.com/2019-us-ad-blocking-usage-report/
. [last accessed: 21.05.2021].
• Federal Trade Commission, 1998, Privacy Online: a report to Congress. Available at:
https://www.ftc.gov/sites/default/files/documents/reports/privacy-online-report-congress/priv-
23a.pdf. [last accessed: 19.04.2021].
• Federal Trade Commission, 2009, Self-Regulation Principles for Online Behavioral Advertising. Available at:
https://www.ftc.gov/sites/default/files/documents/reports/federal-trade-commission-staff-report-self-
regulatory-principles-online-behavioral-advertising/p085400behavadreport.pdf. [last accessed:
16.04.2021].
• Federal Trade Commission, 2013, Mobile App Developers: Start with Security. Available at:
https://www.ftc.gov/tips-advice/business-center/guidance/app-developers-start-security
. [last
accessed: 16.04.2021].
• Federal Trade Commission, 2013, Mobile Privacy Disclosures: Building Trust Through Transparency: A
Federal Trade Commission Staff Report. Available at:
https://www.ftc.gov/reports/mobile-privacy-
disclosures-building-trust-through-transparency-federal-trade-commission.
[last accessed: 08.04.2021].
• Federal Trade Commission, 2019, FTC Seeks to Examine the Privacy Practices of Broadband Providers, Press
Release. Available at:
https://www.ftc.gov/news-events/press-releases/2019/03/ftc-seeks-examine-
privacy-practices-broadband-providers. [last accessed 02.05.2021].
• Federal Trade Commission, 2020, Consumer Sentinel Network Data Book: Annual data book 2020.
Available at:
https://www.ftc.gov/system/files/documents/reports/consumer-sentinel-network-data-
book-2020/csn_annual_data_book_2020.pdf. [last accessed: 08.04.2021].
• Federal Trade Commission, 2020, Data on Refunds to Consumers, Cases and Proceedings. Available at:
https://www.ftc.gov/enforcement/cases-proceedings/refunds/data-refunds-consumers
. [last accessed:
15.04.2021].
• Federal Trade Commission, 2020, FTC Data Shows Record Surge in Online Shopping Complaints During
Pandemic, Press Release. Available at:
https://www.ftc.gov/news-events/press-releases/2020/07/ftc-
data-shows-record-surge-online-shopping-complaints-during. [last accessed: 08.04.2021].
• Federal Trade Commission, 2020, FTC Issues Orders to Nine Social Media and Video Streaming Services
Seeking Data About How They Collect, Use, and Present Information, Press Release. Available at:
https://www.ftc.gov/news-events/press-releases/2020/12/ftc-issues-orders-nine-social-media-video-
streaming-services. [last accessed: 23.04.2021].
• Federal Trade Commission, 2020, FTC Sends Report to Congress on Social Media Bots and Deceptive
Advertising, Press Releases. Available at:
https://www.ftc.gov/news-events/press-releases/2020/07/ftc-
sends-report-congress-social-media-bots-deceptive-advertising. [last accessed: 15.04.2021].
• Federal Trade Commission, 2021, Bringing Dark Patterns to Light: An FTC Workshop, 29 April 2021.
Available at:
https://www.ftc.gov/news-events/events-calendar/bringing-dark-patterns-light-ftc-
workshop. [last accessed 02.05.2021].
• Federal Trade Commission, 2021, California Company Settles FTC Allegations It Deceived Consumers about
use of Facial Recognition in Photo Storage App, Press Release. Available at:
https://www.ftc.gov/news-
events/press-releases/2021/01/california-company-settles-ftc-allegations-it-deceived-consumers. [last
accessed: 06.04.2021].
• Federal Trade Commission, 2021, Developer of Popular Women’s Fertility-Tracking App Settles FTC
Allegations that It Misled Consumers About the Disclosure of their Health Data, Press Release. Available at:
https://www.ftc.gov/news-events/press-releases/2021/01/developer-popular-womens-fertility-
tracking-app-settles-ftc. [last accessed: 06.04.2021].
• Federal Trade Commission, Children’s Online Privacy Protection Rule, 1998. Available at:
https://www.ftc.gov/enforcement/rules/rulemaking-regulatory-reform-proceedings/childrens-online-
privacy-protection-rule. [last accessed: 19.05.2021].
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 110
• Federal Trade Commission, Children's Online Privacy Protection Rule ("COPPA"), Children's Online Privacy
Protection Act of 1998, 15 U.S.C. 6501–6505.
Available at:
https://www.ftc.gov/enforcement/rules/rulemaking-regulatory-reform-proceedings/childrens-online-
privacy-protection-rule. [last accessed: 15.04.2021].
• Federal Trade Commission, Consumer Advice: Imposter Scams.
Available at: https://www.consumer.ftc.gov/features/feature-0037-imposter-scams
. [last accessed:
17.04.2021].
• Federal Trade Commission, Truth in Advertising, Media Resources.
Available at: https://www.ftc.gov/news-events/media-resources/truth-advertising
. [last accessed:
19.04.2021].
• Fernandez, S., Jenkins, P., & Vieira, B., 2020, Europe’s digital migration during COVID-19: Getting past the
broad trends and averages. Available at: https://www.mckinsey.com/business-functions/mckinsey-
digital/our-insights/europes-digital-migration-during-covid-19-getting-past-the-broad-trends-and-
averages. [last accessed: 20.04.2021].
• Forbrukerrådet, 2018, Deceived by Design - How tech companies use dark patterns to discourage us from
exercising our rights to privacy.
Available at:
https://fil.forbrukerradet.no/wp-content/uploads/2018/06/2018-06-27-deceived-by-
design-final.pdf. [last accessed: 20.04.2021].
• Ford, G. S., 2020, ‘COVID-19 and Broadband Speeds: A Multi-Country Analysis’, Phoenix Centre Policy
Bulletin 49.
• Fou, A., 2020, Out Of The Darkness (Of Ad Fraud) And Into The Light, Forbes. Available at:
https://www.forbes.com/sites/augustinefou/2020/07/05/out-of-the-darkness-of-ad-fraud-and-into-
the-light/. [last accessed: 14.02.2021].
• French Legal Statute, 2017, Decree No. 2017-159 of February 9, 2017 on digital advertising services.
Available at: https://www.legifrance.gouv.fr/loda/id/JORFTEXT000034024418/
. [last accessed:
08.04.2021].
• French Minister of Culture and the Secretary of State in charge of digital, 2020, Online advertising: A level
playing field.
Available at:
https://www.igf.finances.gouv.fr/files/live/sites/igf/files/contributed/IGF%20internet/2.RapportsPublic
s/2020/2020-M-039-04%20PUB%20internet-VPUBLIQUE.pdf. [last accessed: 23.04.2021].
• GDPR.EU, 2021, The EU’s GDPR only applies to personal data, which is any piece of information that relates
to an identifiable person. It’s crucial for any business with EU consumers to understand this concept for GDPR
compliance. Available at: https://gdpr.eu/eu-gdpr-personal-data/
. [last accessed: 20.04.2021].
• GDPR.eu, What is considered personal data under the EU GDPR? Available at: https://gdpr.eu/eu-gdpr-
personal-data/. [last accessed: 20.04.2021].
• Geradin, D. & D. Katsifis, 2019, ‘An EU competition law analysis of online display advertising in the
programmatic age’, European Competition Journal, 15:1, 55-96.
• German Advertising Standards Council, 2009, Working Principles.
Available at: https://werberat.de/sites/default/files/uploads/working_principles_2009.pdf
[last
accessed: 21.04.2021].
• German Law, 2004, Act against Unfair Competition (UWG), Annex (to Section 3 Paragraph 3). Available at:
https://www.gesetze-im-internet.de/uwg_2004/anhang.html
. [last accessed: 15.04.2021].
• Ghislain, S., 2018, Publicité et professions libérales : questions actuelles au regard de l’arrêt Vanderborght du
4 mai 2017 de la Cour de Justice, D.C.C.R., 2018/3-4, nos 120-121.
• Giersberg, F., 2020, Advertising market 2020 with clear corona effects. Available at:
https://www.vau.net/zaw/content/zaw-werbemarkt-2020-deutlichen-corona-auswirkungen
. [last
accessed: 16.04.2021].
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
111 PE 662.913
• Giersberg, F., 2020, ZAW: Werbemarkt 2020 mit deutlichen Corona-Auswirkungen. Available at:
https://www.vau.net/zaw/content/zaw-werbemarkt-2020-deutlichen-corona-auswirkungen
. [last
accessed: 07.04.2021].
• Global Data Hub, 2020, Beware 'dark patterns' – data protection regulators are watching. Available at:
https://globaldatahub.taylorwessing.com/article/beware-dark-patterns-data-protection-regulators-
are-watching [last accessed: 02.05.2021].
• Global Witness, 2021, Do people really want personalised ads online?. Available at:
https://www.globalwitness.org/en/blog/do-people-really-want-personalised-ads-online/
.
[last accessed: 29.04.2021].
• Goldfarb, A., 2014, ‘What is different about online advertising?’, Review of Industrial Organization, 44(2),
115-129.
• Google, 2021, Ad Manager.
Available at: https://support.google.com/admanager/answer/152039?hl=en
.
[last accessed: 29.04.2021].
• Google, 2021, Search Ads 360 overview - What’s Search Ads 360.
Available at: https://support.google.com/searchads/answer/1187512?hl=en.
[last accessed: 02.06.2021].
• Gordon et al., 2020, ‘Inefficiencies in Digital Advertising Markets’, Journal of Marketing, 85(1), 7-25.
• Grossman, G. M., & Shapiro, C., 1984, ‘Informative advertising with differentiated products’. The Review
of Economic Studies, 51(1), 63-81.
• Guha, S., Cheng, B., & Francis, P., 2011, ‘Privad: Practical privacy in online advertising’, Proceedings of the
8th USENIX Conference on Networked Systems Design and Implementation (pp. 169-182).
• GuptaMedia, 2021, Understanding the effects of COVID-19 on the Facebook Ads Marketplace, Available at:
https://www.guptamedia.com/blog/facebook-ads/understanding-the-effects-of-covid-19-on-the-
facebook-ads-marketplace/. [last accessed: 08.04.2021].
• Haider et al., 2018, ‘An ensemble learning based approach for impression fraud detection in mobile
advertising’, Journal of Network and Computer Applications, 112, 126-141.
• Ham, C., 2017, ‘Exploring how consumers cope with online behavioral advertising’, International Journal
of Advertising, 36 (4):632-658.
• Huber, D., 2021, Werbung im Internet: Ist das rechtlich zulässig?, IT-Recht Kanzlei zum Werberecht.
Available at: https://www.it-recht-kanzlei.de/werberecht-werbung-internet-serie.html?print=1
[last
accessed: 26.04.2021].
• IAB Europe, 2017, Research report: The economic contribution of online advertising in Europe, Available
at: https://iabeurope.eu/knowledgehub/policy/research-report-the-economic-contribution-of-online-
advertising-in-europe/. [last accessed: 20.05.2021].
• IAB Europe, 2020, IAB Europe AdEx Benchmark 2019 Study reveals European Digital Advertising Market
Exceeds €64bn in 2019. Available at:
https://iabeurope.eu/all-news/iab-europe-adex-benchmark-2019-
study-reveals-european-digital-advertising-market-exceeds-e64bn-in-2019/ . [last accessed: 26.05.2021].
• IAB Europe, 2020, IAB Europe Backs ‘Don’t Block Quality European Journalism’. Available at:
https://iabeurope.eu/blog/iab-europe-backs-dont-block-quality-european-journalism/
. [last accessed:
22.04.2021].
• IAB Europe, 2021, What would an Internet without targeted ads look like?
Available at:
https://iabeurope.eu/wp-content/uploads/2021/04/IAB-Europe_What-Would-an-
Internet-Without-Targeted-Ads-Look-Like_April-2021.pdf. [last accessed: 26.04.2021].
• ICC, 2018, ICC Advertising and Marketing Communications Code.
Available at: https://iccwbo.org/publication/icc-advertising-and-marketing-communications-code/.
[last accessed: 26.05.2021].
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 112
• ICO, Our work on adtech. Available at: https://ico.org.uk/about-the-ico/what-we-do/our-work-on-
adtech/.
• IMF, 2021, World Economic Outlook Database, Extraction of GDP Europe datapoint 2019. Available at:
https://www.imf.org/en/Publications/WEO/weo-database/2021/April/weo-
report?a=1&c=998,&s=NGDPD,PPPGDP,NGDPRPPPPC,PPPPC,&sy=2018&ey=2026&ssm=0&scsm=1&sc
c=0&ssd=1&ssc=0&sic=0&sort=country&ds=.&br=1.
• Ipsos, London Economics & Deloitte, 2018, Consumer market study on online market segmentation
through personalised pricing/offers in the European Union - Request for Specific Services 2016 85 02 for the
implementation of Framework Contract EAHC/2013/CP/04. Study prepared for the European Commission.
https://ec.europa.eu/info/sites/info/files/aid_development_cooperation_fundamental_rights/aid_and
_development_by_topic/documents/synthesis_report_online_personalisation_study_final_0.pdf.
[last accessed: 12.04.2021].
• Irish Statute Book, 1992, Patents Act 1992.
Available at: http://www.irishstatutebook.ie/eli/1992/act/1/enacted/en/html
.
[last accessed: 15.04.2021].
• Irish Statute Book, 1995, Consumer Credit Act 1995.
Available at: http://www.irishstatutebook.ie/eli/1995/act/24/enacted/en/html
.
[last accessed: 16.04.2021].
• Irish Statute Book, 1996, Trademarks Act 1996.
Available at: http://www.irishstatutebook.ie/eli/1996/act/6/enacted/en/html
.
[last accessed: 16.04.2021].
• Irish Statute Book, 2000, Copyright and Related Rights Act 2000. Available at:
http://www.irishstatutebook.ie/eli/2000/act/28/enacted/en/html
. [last accessed: 16.04.2021].
• Irish Statute Book, 2007, Consumer Protection Act 2007.
Available at: http://www.irishstatutebook.ie/eli/2007/act/19/enacted/en/html
.
[last accessed: 15.04.2021].
• Irish Statute Books, 2011, European Communities (Electronic Communications Networks and Services)
(Privacy and Electronic Communications) Regulations.
Available at: http://www.irishstatutebook.ie/eli/2011/si/336/made/en/pdf
. [last accessed: 15.04.2021].
• Johnson, J. P., 2013, ‘Targeted advertising and advertising avoidance’, The RAND Journal of Economics,
44(1), 128-144.
• Juniper Research, 2017, Ad Fraud to Cost Advertisers $19 billion in 2018, Representing 9% of Total Digital
Advertising Spend. Available at:
https://www.juniperresearch.com/press/press-releases/ad-fraud-to-
cost-advertisers-19-billion-in-2018. [last accessed: 23.04.2021].
• Kafka, P., 2010, Google Cuts Off AppNexus, and the Ad Tech World Shudders, AllThingsD.com. Available at:
http://allthingsd.com/20101130/google-cuts-off-appnexus-and-the-ad-tech-world-shudders/
.
[last accessed: 13.04.2021].
• Kaspersky (2020): Spam and Phishing in Q2 2020. Available at:
https://securelist.com/spam-and-
phishing-in-q2-2020/97987/ and https://media.kasperskycontenthub.com/wp-
content/uploads/sites/43/2020/08/04164340/sl_spam_report_q2_01-en-dolya-spama-v-mirovom-
pochtovom-trafike-q1-2020-q2-2020-gg.png. [Last accessed: 16.02.2021].
• Kaye, B. K., & Medoff, N., 2001, Just a Click Away: Advertising on the Internet. Massachusetts: Allyn and
Bacon as cited in Evans, D. S. (2009): ‘The Online Advertising Industry: Economics, Evolution, and
Privacy’, Journal of Economic Perspectives 23:3, 37–60.
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
113 PE 662.913
• Kincaid, J., 2010, Google (Temporarily) Blocks AppNexus From Its Ad Exchange, TechCrunch. Available at:
https://techcrunch.com/2010/11/30/google-temporarily-blocks-appnexus-from-its-ad-
exchange/?guccounter=1&guce_referrer=aHR0cHM6Ly93d3cuZ29vZ2xlLmRlLw&guce_referrer_sig=AQ
AAALba9123WWwNuONVTQm3ISs9bgpJNpSHPT7Lnt63zIiymC-fR7QFsANvMaETIlLYSpS-LnKIf_0-
_H0Gebf9LAb9nHxQNX-IKOIPpYer6om5VNM7PsgjLkpu3MvpvF6EuT366grt-LSvzhV6aS4QIFO--
oRA3x78Zr2mVNdoZ8cB. [last accessed: 13.04.2021].
• Kingaby, H., 2020, AI and Advertising: A Consumer Perspective. Available at: https://789468a2-16c4-4e12-
9cd3-063113f8ed96.filesusr.com/ugd/435e8c_3f6555abb25641be8b764f5093f1dd4f.pdf.
[last accessed: 13.04.2021].
• Köhler, H., Bornkamm, J., Henning-Bodewig, F., 2002, Vorschlag für eine Richtlinie zum Lauterkeitsrecht
und eine UWG-Reform | WRP 2002, 1317-1328.
• Le Roy, G., 2020, Australian Digital Advertising Market Experiences Double Digit Decline In Q2 2020 Due To
Impact Of COVID-19. Available at: https://iabaustralia.com.au/news/16338/
. [last accessed: 12.04.2021].
• Les Echos, 2017, Decree on the transparency of online advertising finally published, News Article. Available
at:
https://www.lesechos.fr/2017/02/le-decret-sur-la-transparence-de-la-pub-en-ligne-enfin-publie-
160395. [last accessed: 09.04.2021].
• LG Itzehoe, 23.11.2018, Az. 3 O 151/18; OLG Braunschweig, 08.01.2019, Az. 2 U 89/19.
• LG München I, 29.04.2019, Az. 4 HK O 14312/18.
• Li, C. & Hall, S., 2020, This is how COVID-19 is affecting the advertising industry. World Economic Forum.
Available at:
https://www.weforum.org/agenda/2020/06/coronavirus-advertising-marketing-covid19-
pandemic-business/. [last accessed: 12.04.2021].
• Liu, D., Chen, J., & Whinston, A. B., 2010, ‘Ex ante information and the design of keyword auctions’,
Information Systems Research, 21(1), 133-153.
• Lomas, N., 2021, After Facebook’s news flex, Australia passes bargaining code for platforms and publishers,
Tech Crunch. Available at
: https://techcrunch.com/2021/02/25/after-facebooks-news-flex-australia-
passes-bargaining-code-for-platforms-and-publishers/. [last accessed: 15.04.2021].
• Lu, X., Zhao, X., & Xue, L., 2016, ‘Is combining contextual and behavioral targeting strategies effective in
online advertising?’, ACM Transactions on Management Information Systems (TMIS), 7(1), 1-20.
• Luguri, J. and Strahilevitz, L. J., 2019, Shining a Light on Dark Patterns, University of Chicago, Public Law
Working Paper. No. 719.
• Lyons, K., 2021, Google reportedly ran secret ‘Project Bernanke’ that boosted its own ad-buying system over
competitors, The Verge. Available at:
https://www.theverge.com/2021/4/11/22378296/google-secret-
project-bernanke-ad-buying-competitors-texas. [last access: 19.04.2021].
• Malloy, M., McNamara, M., Cahn, A., & Barford, P., 2016, Ad blockers: Global prevalence and impact,
Proceedings of the 2016 Internet Measurement Conference (pp. 119-125).
• Mathews-Hunt, K., 2019, ‘CookieConsumer: Tracking online behavioural advertising in Australia’,
Computer Law & Security Review 32, pp 55-90.
• Meurer, M., & Stahl II, D. O., 1994, ‘Informative advertising and product match‘, International Journal of
Industrial Organization, 12(1), 1-19.
• Moore, R., Moore, M., Shanahan, K, Horky, A. & B. Mack, 2015, ‘Creepy Marketing: Three Dimensions of
Perceived Excessive Online Privacy Violation’, Marketing Management Journal, 25:42-53.
• Mordor Intelligence, 2021, Online advertising market – growth, trends, covid-19 impact and forcasts(2021
- 2026). Available at: https://www.mordorintelligence.com/industry-reports/online-advertising-market
.
[last accessed: 16.04.2021].
• Morton, F. M. S., & Dinielli D. C., 2020, Roadmap for Digital Advertising Monopolization Case Against
Google. Omidyar Network. Available at:
https://omidyar.com/wp-content/uploads/2020/09/Roadmap-
for-a-Case-Against-Google.pdf. [last accessed: 20.04.2021].
• Mughees, M. H., & Qian, Z., 2017, ‘Detecting anti ad-blockers in the wild’, Proceedings on Privacy
Enhancing Technologies, 2017 (3):127–142.
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 114
• Munz, M. and Hickman, T., 2016, Court confirms that IP addresses are personal data in some cases. White
& Case Technology Newsflash. Available at: https://www.whitecase.com/publications/alert/court-
confirms-ip-addresses-are-personal-data-some-cases. [last accessed: 20.04.2021].
• Nithyanand, R., Khattak, S., Javed, M., Vallina-Rodriguez, N., Falahrastegar, M., Powles, J. E., ... & Murdoch,
S. J., 2016, ‘Adblocking and counter blocking: A slice of the arms race’, Proceedings of the 6th USENIX
Workshop on Free and Open Communications on the Internet (FOCI '16).
• Office of the Australian Information Commissioner, Targeted advertising: Your Rights. Available at:
https://www.oaic.gov.au/privacy/your-privacy-rights/advertising-and-marketing/targeted-
advertising/. [last accessed: 19.04.2021].
• OLG Celle, 08.06.2017, Az. 13 U 53/17; KG Berlin, 11.10.2017, Az. 5 W 221/17.
• Oliver Wyman, 2020 Report, Syndicat des Régies Internet (SRI). Available at:
https://www.sri-
france.org/wp-content/uploads/2021/02/Obsepub_2020_ENG.pdf. [last accessed: 18.04.2021].
• Onlinemarketing.de, 2020, Welche Auswirkungen hat die Abschaffung von Third Party Cookies auf die
Werbebranche? Available at:
https://onlinemarketing.de/unternehmensnews/google-third-party-
cookies-blockieren-folgen. [last accessed: 20.04.2021].
• Ostrovsky, M., & Schwarz, M., 2011, ‘Reserve prices in internet advertising auctions: A field experiment’,
Proceedings of the 12th ACM conference on Electronic commerce (pp. 59-60).
• PLUM, 2019, Online advertising in the UK: A report commissioned by the Department for Digital, Culture,
Media and Sport.
Available at:
https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/
777996/Plum_DCMS_Online_Advertising_in_the_UK.pdf. [last accessed: 12.04.2021].
• Politou, E., Alepis, E., & Patsakis, C., 2018, ‘Forgetting personal data and revoking consent under the
GDPR: Challenges and proposed solutions’, Journal of cybersecurity, Vol. 4, p. 12.
• Privacy International, 2019, Why am I really seeing that ad? The answer might be Real Time Bidding (RTB).
Available at:
https://privacyinternational.org/explainer/2974/why-am-i-really-seeing-ad-answer-might-
be-real-time-bidding-rtb. [last accessed: 18.04.2021].
• Procee, R., Arnold, R., Marcus, J. S., Fina, D., Lennartz, J., Kazakova, S. & E. Lykogianni, 2020, Study on the
impacts of the extension of the scope of the Geoblocking Regulation to audiovisual and non-audiovisual
services giving access to copyright protected content – Final Report. Study for the European Commission.
Available at:
https://op.europa.eu/en/publication-detail/-/publication/4fc70542-223b-11eb-b57e-
01aa75ed71a1/language-en#. [last accessed: 16.02.2021].
• Roberts, K., & Southgate, D., 2020, Impact Of COVID-19 On Consumer Behaviour, Attitudes and
Expectations, Kantar, iab Europe.
Available at:
https://iabeurope.eu/wp-content/uploads/2020/05/Kantar_Covid-19-Barometer_IAB-
Europe-Webinar-Presentation_Final.pdf. [last accessed: 20.04.2021].
• RTE, 2020, Irish digital advertising spend grew by 17% last year.
Available at:
https://www.rte.ie/news/business/2020/0424/1134515-iab-pwc-online-adspend-
report/#:~:text=Today's%20report%20shows%20that%20search,digital%20adspend%20at%20%E2%
82%AC41m. [last accessed: 01.04.2021].
• Rybnik, R., 2020, How can publishers prepare for the end of the third-party cookies? Available at:
https://netlabe.com/how-can-publishers-prepare-for-the-end-of-the-third-party-cookies-
db91147f985e. [last accessed: 15.04.2021].
• Schiff, A., 2021, Google Will Not Run FLoC Origin Tests In Europe Due To GDPR Concerns (At Least For Now),
Adexchanger. Available at:
https://www.adexchanger.com/platforms/google-will-not-run-floc-origin-
tests-in-europe-due-to-gdpr-concerns/. [last accessed: 20.04.2021].
• Schürmann, K., 2020, E-Mail-Marketing ohne Einwilligung – was ist erlaubt?, Schürmann Rosenthal Dreyer
Rechtsanwälte. Available at: https://www.srd-rechtsanwaelte.de/blog/e-mail-marketing-einwilligung/
.
[last accessed: 26.04.2021].
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
115 PE 662.913
• Segura, J., 2017, Canada and the U.K. hit by Ramnit Trojan in new malvertising campaign.
Available at:
https://blog.malwarebytes.com/threat-analysis/exploits-threat-analysis/20 17/03/canada-
u-k-hit-ramnit-trojan-new-malvertising-campaign/. [last accessed: 21.04.2021].
• Simply Business, 2011, Online marketing and the CAP Code – What you need to do. Available at:
https://www.simplybusiness.co.uk/knowledge/articles/2011/03/2011-03-11-online-marketing-and-
the-cap-code-what-you-need-to-do/. [last accessed: 12.04.2021].
• Sluis, S., 2020, Google Chrome Will Drop Third-Party Cookies In 2 Years, Adexchanger. Available at:
https://www.adexchanger.com/online-advertising/google-chrome-will-drop-third-party-cookies-in-2-
years/. [last accessed: 15.04.2021].
• Smith, A., 2020, Using Artificial Intelligence and Algorithms, blog. Available at: https://www.ftc.gov/news-
events/blogs/business-blog/2020/04/using-artificial-intelligence-algorithms.
[last accessed: 23.04.2021].
• Srinivasan, D., 2020, ‘Why Google Dominates Advertising Markets – Competition Policy Should Lean on
the Principles of Financial Market Regulation’, Standford Technology Law Review 24 (1): 55-175.
• Statcounter.com, 2021, Browser Market Share Worldwide – Statistic.
Available at: https://gs.statcounter.com/browser-market-share
. [last accessed: 25.04.2021].
• Statista, 2020, Digital Advertising Europe - Statistic.
Available at: https://de.statista.com/outlook/dmo/digitale-werbung/europa
.
[last accessed: 09.04.2021].
• Statista, 2020, Digital Advertising Report 2020.
• Statista, 2020, Digital Advertising worldwide – Statistic.
Available at: https://de.statista.com/outlook/dmo/digitale-werbung/weltweit
.
[last accessed: 09.04.2021].
• Statista, 2020, Marktanteile der einzelnen Werbemedien im deutschen Bruttowerbemarkt im Jahr 2019.
Available at:
https://de.statista.com/statistik/daten/studie/154767/umfrage/mediasplit-im-deutschen-
werbemarkt/. [last accessed: 21.04.2021].
• Statista, 2021, Digital Markets, Digitale Werbung, Europa.
Available at: https://de.statista.com/outlook/dmo/digitale-werbung/europa
.
[last accessed: 09.04.2021].
• Statista, 2021, Digital Markets, Digitale Werbung, Weltweit.
Available at: https://de.statista.com/outlook/dmo/digitale-werbung/weltweit
.
[last accessed: 09.04.2021].
• Storey, G., Reisman, D., Mayer, J., & Narayanan, A., 2017, The future of ad blocking: An analytical framework
and new techniques, arXiv preprint arXiv:1705.08568.
• Strycharz, J.; van Noort, G.; Smit, E. & N. Helberger, 2019, Consumer View on Personalized Advertising:
Overview of Self-Reported Benefits and Concerns. In: Advances in Advertising Research X: Multiple
Touchpoints in Brand Communication, edited by E. Bigne & S. Rosengren, 53-66. Wiesbaden: Springer
Fachmedien Wiesbaden.
• Stuyck, J., and Straetmans, G., 2016, (ed.), Commercial practices, Gent, Larcier.
• Sullivan, G., 2014, Cornell ethics board did not pre-approve Facebook mood manipulation study,
Washington Post.
Available at:
https://www.washingtonpost.com/news/morning-mix/wp/2014/07/01/facebooks-
emotional-manipulation-study-was-even-worse-than-you-thought/. [last accessed: 09.04.2021].
• Sunderland, J., Herrera, F., Esteves, S., Godlovitch, I., Wiewiorra, L., Taş, S., Kroon, P., Stronzik, M.,
Baischew, D., Nett, L., Tenbrock, S., Strube Martins, S., de Streel, A., Kalliala, J., Huerta Bravo, J., Maxwell,
W., & Renda, A., 2021, Digital Markets Act: Impact assessment support study: annexes, Available at:
https://op.europa.eu/de/publication-detail/-/publication/2a69fd2a-3e8a-11eb-b27b-
01aa75ed71a1/language-de. [last accessed: 20.05.2021].
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 116
• Tech Crunch, 2019, Behavioural advertising is out of control, warns UK watchdog, Article. Available at:
https://techcrunch.com/2019/06/20/behavioural-advertising-is-out-of-control-warns-uk-
watchdog/.[last accessed: 23.03.2021].
• Teplitzky, O., Peifer, K.-N., & Leistner, M., 2020, Großkommentar zum Gesetz gegen den unlauteren
Wettbewerb mit Nebengesetzen, Kommentar, Article 575.
• Teplitzky, O., Peifer, K.-N., & Leistner, M., 2020, Großkommentar zum Gesetz gegen den unlauteren
Wettbewerb mit Nebengesetzen.
• The Australian Consumer Law, Schedule 2 of the Competition and Consumer Act 2010. Available at:
https://consumerlaw.gov.au/australian-consumer-law/legislation
. [last accessed: 15.04.2021].
• Thorpe, E. K., 2020, First-party data has empowered publishers to experiment with personalisation and
better advertising. Available at:
https://whatsnewinpublishing.com/first-party-data-has-empowered-
publishers-to-experiment-with-personalisation-and-better-advertising/. [last accessed: 15.04.2021].
• Thorpe, E. K., 2020, RIP third-party cookies: What Google’s ‘pivot to privacy’ means for publishers, Available
at: https://whatsnewinpublishing.com/rip-third-party-cookies-what-googles-pivot-to-privacy-means-
for-publishers/. [last accessed: 15.04.2021].
• Top Draw Inc., 2020, Online Advertising Costs in 2021, Available at:
https://www.topdraw.com/insights/is-online-advertising-
expensive/#:~:text=We%27re%20talking%20about%20the,advertising%20is%20%2422%20and%20u
p. [last accessed: 19.05.2021].
• Toubiana, V., Narayanan, A., Boneh, D., Nissenbaum, H., & Barocas, S., 2010, ‘Adnostic: Privacy preserving
targeted advertising’, Proceedings of the Network and Distributed System Security Symposium. Available
at: https://www.ndss-symposium.org/wp-content/uploads/2017/09/toub.pdf
. [last accessed:
19.04.2021].
• Truyens, M., Van Eecke, P., 2009, EU study on the legal analysis of a single market for the information society,
European Commission. Available at:
https://op.europa.eu/en/publication-detail/-
/publication/a856513e-ddd9-45e2-b3f1-6c9a0ea6c722. [last accessed: 16.02.2021].
• Tucker, C. E., 2013, ‘Social Networks, Personalized Advertising, and Privacy Controls’, Journal of Marketing
Research, 51 (5):546-562.
• Turow, J., King, J., Hoofnagle, C., Bleakley, A., & M. Hennessy, 2009, Americans Reject Tailored Advertising
and Three Activities That Enable It. Available at: SSRN: https://ssrn.com/abstract=1478214
. [last accessed:
16.04.2021].
• UK Advertising Standards Agency, 2019, Using technology for good – Annual Report 2019. Available at:
https://www.asa.org.uk/uploads/assets/5d2b832d-8d12-4f28-9e9f5df54f7b93c1/809de510-6285-
4c57-9c904b1c91f6d0ec/ASA-CAP-2019-Annual-Report.pdf [last accessed: 21.04.2021].
• UK Statute Book, 2008, The Consumer Protection from Unfair Trading Regulations 2008. Available at:
https://www.legislation.gov.uk/uksi/2008/1277/contents/made
[last accessed: 12.04.2021].
• UK Statute Book, 2014, The Consumer Protection (Amendment) Regulations 2014. Available at:
https://www.legislation.gov.uk/uksi/2014/870/made
. [last accessed: 12.04.2021].
• UK Statute Book, 2015, Consumer Rights Act 2015.
Available at: https://www.legislation.gov.uk/ukpga/2015/15/contents
. [last accessed: 12.04.2021].
• Ullah, I., Boreli, R., & Kanhere, S. S., 2020, Privacy in targeted advertising: A survey, arXiv preprint
arXiv:2009.06861.
• United States Federal Trade Commission, 2020, FTC Seeks Public Comment on its Endorsement Guides,
Press Release. Available at:
https://www.ftc.gov/news-events/press-releases/2020/02/ftc-seeks-public-
comment-its-endorsement-guides.[last accessed: 13.04.2021].
• United States Federal Trade Commission, About the FTC: Office of International Affairs. Available at:
https://www.ftc.gov/about-ftc/bureaus-offices/office-international-affairs
. [last accessed: 14.04.2021].
• United States Federal Trade Commission, Enforcement page: Consumer Sentinel Network. Available at:
https://www.ftc.gov/enforcement/consumer-sentinel-network
. [last accssed: 13.04.2021].
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
117 PE 662.913
• Varian, H. R., 2009, ‘Online ad auctions’, American Economic Review: Papers & Proceedings, 99(2), 430-34.
• Verdoodt, V. & Feci, N., 2017, Children and targeted advertising in the proposed ePrivacy (Regulation)
colouring book: Time to use the whole box of crayons.
Available at: https://www.law.kuleuven.be/citip/blog/children-and-targeted-advertising-in-the-
proposed-eprivacy-regulation-colouring-book-time-to-use-the-whole-box-of-crayons/. [last accessed:
25.05.2021].
• W3Tech, 2021, Market share yearly trends for advertising networks,
Available at: https://w3techs.com/technologies/history_overview/advertising/ms/y. [last accessed:
21.05.2021].
• Wachter, S., 2019, Affinity Profiling and Discrimination by Association in Online Behavioural Advertising,
Berkeley Technology Law Journal, Vol. 35, No. 2.
• Wang, J., Zhang, W., & Yuan, S., 2016, Display advertising with real-time bidding (RTB) and behavioural
targeting. ArxivPreprint Arxiv:1610.03013(2016).
• Warren, T., 2020, Facebook hits back at Apple with second critical newspaper ad, The Verge. Available at:
https://www.theverge.com/2020/12/17/22180102/facebook-new-newspaper-ad-apple-ios-14-
privacy-prompt. [last accessed: 23.04.2021].
• Weber, M., 2020, Werbemarkt in Deutschland um 1,9 Prozent gewachsen, W&V. Available at:
https://www.wuv.de/marketing/werbemarkt_in_deutschland_um_1_9_prozent_gewachsen.
[last
accessed: 07.04.2021].
• Wettbewerbszentrale, 2010, Synopse UWG – Richtlinie unlautere Geschäftspraktiken (UGP-RL). Available
at: https://www.wettbewerbszentrale.de/media/getlivedoc.aspx?id=33555
. [last accessed: 15.04.2021].
• WFA, 2016, WFA issues first advice for combatting ad fraud.
Available at:
https://wfanet.org/knowledge/item/2016/06/06/WFA-issues-first-advice-for-combatting-
ad-fraud. [last accessed: 23.04.2021].
• WFA, 2020, Alcohol marketers seek to prevent minors from seeing alcohol ads.
Available at:
https://wfanet.org/knowledge/item/2020/04/06/Alcohol-marketers-seek-to-prevent-
minors-from-seeing-alcohol-ads. [last accessed: 23.04.2021].
• WFA, 2021, WFA global position on advertiser access to data in the digital advertising market, Available at:
https://wfanet.org/knowledge/data-insight/item/2021/04/09/WFA-global-position-on-advertiser-
access-to-data-in-the-digital-advertising-market. [last accessed: 20.05.2021].
• Which?, 2019, Control, Alt or Delete? The future of consumer data.
Available at:
https://www.which.co.uk/policy/digital/2659/control-alt-or-delete-the-future-of-
consumer-data-main-report.
• Which?, 2021, Consumer Protection from Unfair Trading Regulations 2008.
Available at:
https://www.which.co.uk/consumer-rights/regulation/consumer-protection-from-unfair-
trading-regulations-2008-asO0C3p6VZQR [last accessed: 30.03.2021].
• White & Case LLP, 2016, Court confirms that IP addresses are personal data in some cases. Available at:
https://www.whitecase.com/publications/alert/court-confirms-ip-addresses-are-personal-data-some-
cases. [last accessed: 20.04.2021].
• Widawska, I., 2021, One Year With COVID-19: Programmatic’s Performance Prior To And During The Course
Of The Pandemic. Available at:
https://yieldbird.com/one-year-with-covid-19-programmatic-
performance/. [last accessed: 30.04.2021].
• Wilde Beuger Solmecke Rechtsanwälte, 2021, Rechtsanwälte, Irreführende Werbung. Available at:
https://www.wbs-law.de/wettbewerbsrecht/werbung-rechtssicher/irrefuehrende-werbung/
.
• Wordstream, 2020, Conversion Rate Benchmarks: Find Out How Your Conversion Rate Compares.
Available at: https://www.wordstream.com/blog/ws/2019/08/19/conversion-rate-benchmarks
. [last
accessed: 26.05.2021].
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 118
• Zarras, A., Kapravelos, A., Stringhini, G., Holz, T., Kruegel, C., & Vigna, G., 2014, ‘The dark alleys of madison
avenue: Understanding malicious advertisements’, Proceedings of the 2014 Conference on Internet
Measurement Conference (pp. 373-380).
• Zenith, 2020, Internet advertising spending worldwide from 2007 to 2022, by format, Statista. Available at:
https://www.statista.com/statistics/276671/global-internet-advertising-expenditure-by-type/
.
[last accessed: 09.04.20201].
• Zhu, X., Tao, H., Wu, Z. Cao, J., Kalish, K. & J. Kayne, 2017, Ad Fraud Taxonomy and Prevention Mechanisms,
in: Fraud Prevention in Online Digital Advertising, SpringerBriefs in Computer Science 19-23.
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
119 PE 662.913
ANNEX 1: LEGISLATIVE MEASURES AND PROPOSALS
Specific measures applicable to advertising in the digital environment
a) eCommerce Directive
1. Nature and scope
The eCommerce Directive
388
aims at harmonising certain legal aspects of information society services
i.e. those that are necessary to establish an adequate level of confidence for users to make use of online
services and to ensure the proper functioning of the internal market
389
(minimum harmonisation).
Commercial communications, including unsolicited commercial communications by electronic mail
are among these harmonised areas
390
.
The eCommerce Directive applies to “information society services”. This refers to “any service normally
provided for remuneration, at a distance, by electronic means and at the individual request of a
recipient of services”
391
, regardless of whether the recipient is a consumer or a trader. The eCommerce
applies both to B2B (professional to professional) and B2C (professional to consumer) relationships.
The concept of “commercial communications” is defined broadly as “any form of communication
designed to promote, directly or indirectly, the goods, services or image of a company, organisation or
person pursuing a commercial, industrial or craft activity or exercising a regulated profession”
392
.
2. Substantial measures (relevant for targeted advertising)
Article 6 of the eCommerce Directive creates various identification and transparency requirements for
commercial communications which are part of, or constitute, an information society service. The
commercial communication must be clearly identifiable as such
393
. The person on whose behalf the
commercial communication is made must be clearly identifiable
394
. Furthermore, the eCommerce
Directive requires promotional offers, such as rebates, and promotional competitions or games, to be
clearly identifiable and the associated conditions easily accessible and presented clearly and
unambiguously
395
.
Article 7 contains additional requirements regarding unsolicited commercial communications sent by
electronic mails. The commercial communication shall be clearly and unambiguously identifiable as
such as soon as it is received by the recipient
396
. Moreover, the service provider using such way to
388
European Parliament and the Council of the European Union, 2000, Directive 2000/31/EC of the European Parliament and of the Council
of 8 June 2000 on certain legal aspects of information society services, in particular electronic commerce, in the Internal Market, OJ L 178
of 17.07.2000 (hereafter the “eCommerce Directive”).
389
eCommerce Directive, Recital 10.
390
eCommerce Directive, Recitals 29 to 31.
391
eCommerce Directive, Article 2(a) referring to the definition of “information society services” laid down in Article 1(2) of Directive 98/34/EC
(European Parliament and the Council of the European Union, 1998, Directive 98/34/EC of the European Parliament and of the Council of
22 June 1998 laying down a procedure for the provision of information in the field of technical standards and regulations, OJ L 204 of
21.07.1998), as amended by Directive 98/48/EC (European Parliament and the Council of the European Union, 1998, Directive 98/48/EC
of the European Parliament and of the Council of 20 July 1998 amending Directive 98/34/EC laying down a procedure for the provision
of information in the field of technical standards and regulations, OJ L 217 of 5.08.1998). The Directive 98/34/EC is no longer in force and
has been replaced by Directive (EU) 2015/1535 (European Parliament and the Council of the European Union, 2015, Directive (EU)
2015/1535 of the European Parliament and of the Council of 9 September 2015 laying down a procedure for the provision of information
in the field of technical regulations and of rules on Information Society services, OJ L 241 of 17.09.2015). The definition of the concept of
“information society services” is now provided in Article 1(b) of the new Directive, but remains identical.
392
eCommerce Directive, Article 2(f).
393
eCommerce Directive, Article 6(a).
394
eCommerce Directive, Article 6(b).
395
eCommerce Directive, Article 6(c) and (d).
396
eCommerce Directive, Article 7(a).
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 120
communicate shall consult regularly the opt-out registers, in which the recipients (natural persons)
who do not wish to receive such communications can register, and respect such a choice
397
. The
ePrivacy Directive (see below) has however consecrated an opt-in principle.
Article 8 of the eCommerce Directive is on the use of commercial communications by regulated
professions (such as dentists, pharmacists, lawyers, etc.)
398
. It creates a positive obligation for the
Member States. They have to authorise the regulated professions to use the commercial
communications which are part of, or constitute, an information society service. Such use shall
nevertheless comply with the professional rules regarding, in particular, the independence, dignity and
honor of the profession, professional secrecy and fairness towards clients and other members of the
profession
399
. The establishment of codes of conduct at a European level is also promoted
400
.
In addition, the internal market clause of Article 3 of the eCommerce Directive provides that
information society service providers are subject to the law of the Member State in which it is
established
401
, for the areas that are in the “coordinated field”. This “coordinated field” covers
requirements with which the service provider has to comply in respect of “the pursuit of the activity of
an information society service, such as requirements […] regarding the quality or content of the service
including those applicable to advertising […]”
402
. As a consequence, therefore the service provider
needs to comply with the law applicable to advertising according to legislation of the Member State
where the service provider is established, and not with law of the Member State where the service is
proposed (country of destination).
3. Enforcement and penalties
According to the directive, infringements to the principles laid down in the Directive must be
sanctioned according to rules in the Member States. Sanctions must however be effective,
proportionate and dissuasive
403
.
b) P2B Regulation
1. Nature and scope
The Regulation on platform to business relations (hereafter the “P2B Regulation”)
404
deals with the
relationships between online intermediation services or online search engines and professional users
of those online services. The P2B Regulation aims at addressing “potential frictions” that can exist in
397
eCommerce Directive, Article 7(b).
398
Article 2(g) of the eCommerce Directive refers to the definition provided several directives that have been replaced by Directive
2005/36/EC (European Parliament and the Council of the European Union, 2005, Directive 2005/36/EC of the European Parliament and of
the Council of 7 September 2005 on the recognition of professional qualifications, OJ L 255 of 30.09.2005); Article 3, paragraph 1(a) of this
new directive defines the “regulated profession” as follows : “a professional activity or group of professional activities, access to which,
the pursuit of which, or one of the modes of pursuit of which is subject, directly or indirectly, by virtue of legislative, regulatory or
administrative provisions to the possession of specific professional qualifications; in particular, the use of a professional title limited by
legislative, regulatory or administrative provisions to holders of a given professional qualification shall constitute a mode of pursuit.
Where the first sentence of this definition does not apply, a profession referred to in paragraph 2 shall be treated as a regulated
profession”.
399
eCommerce Directive, Article 8(1).
400
eCommerce Directive, Article 8(2).
401
eCommerce Directive, Article 3(1).
402
eCommerce Directive, Article 2(h).
403
eCommerce Directive, Article 20.
404
European Parliament and the Council of the European Union, 2019, Regulation (EU) 2019/1150 of the European Parliament and of the
Council of 20 June 2019 on promoting fairness and transparency for business users of online intermediation services, OJ L 186 of
11.7.2019.
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
121 PE 662.913
those relationships
405
. It establishes, at a European level, a set of mandatory rules aiming “to ensure a
fair, predictable, sustainable and trusted online business environment within the internal market”, so
that professional users of the online intermediation services are granted “appropriate transparency, as
well as effective redress possibilities
406
.
The P2B Regulation applies to online intermediation services and online search engines provided to
business users and corporate website users
407
.
“Online intermediation services” are defined as follows: (1) they are information society services
408
; (2)
those services “allow business users to offer goods or services to consumers, with a view to facilitating
the initiating of direct transactions between those business users and consumers, irrespective of where
those transactions are ultimately concluded” and (3) “they are provided to business users on the basis
of contractual relationships between the provider of those services and business users which offer
goods or services to consumers”
409
. The “business user”, under the P2B Regulation, is “any private
individual acting in a commercial or professional capacity who, or any legal person which, through
online intermediation services offers goods or services to consumers for purposes relating to its trade,
business, craft or profession”
410
.
The “online search engine” is the “digital service that allows users to input queries in order to perform
searches of, in principle, all websites, or all websites in a particular language, on the basis of a query on
any subject in the form of a keyword, voice request, phrase or other input, and returns results in any
format in which information related to the requested content can be found”
411
. The websites
concerned are particularly those of corporate website users. A “corporate website user” is defined as
“any natural or legal person which uses an online interface, meaning any software, including a website
or a part thereof and applications, including mobile applications, to offer goods or services to
consumers for purposes relating to its trade, business, craft or profession”
412
.
The relationships considered are business (platform) to business (professional users) relationships as
long as the business users concerned target consumers through their use of the online platform
services.
2. Substantial measures (relevant for targeted advertising)
Both online intermediation services and online search engine must inform their professional users of
the main parameters determining ranking
413
and the importance of those main parameters (including
on the possibility to influence ranking against remuneration)
414 415
.
405
P2B Regulation, Recital 2.
406
P2B Regulation, Recital 7.
407
P2B Regulation, Article 1, paragraph 2.
408
As defined in point (b) of Article 1(1) of Directive (EU) 2015/1535.
409
P2B Regulation, Article 2(2).
410
P2B Regulation, Article 2(1).
411
P2B Regulation, Article 2(5).
412
P2B Regulation, Article 2(7).
413
The concept of “ranking” refers to “the relative prominence given to the goods or services offered through online intermediation services,
or the relevance given to search results by online search engines, as presented, organised or communicated by the providers of online
intermediation services or by providers of online search engines, respectively, irrespective of the technological means used for such
presentation, organisation or communication” (P2B Regulation, Article 2(8)).
414
As explained in European Commission, 2020, Guidelines of 7 December 2020 on ranking transparency pursuant to Regulation (EU)
2019/1150 of the European Parliament and of the Council, OJ C 424/1 of 8.12.2020
415
P2B Regulation, Article 5.
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 122
Under the P2B Regulation, they must also inform their professional users about any differentiated
treatment which they may give, in relation to goods or services offered to consumers, through their
services
416
.
Finally, terms and conditions of online intermediation services must contain a description of technical
and contractual access, if any, to personal data and/or other data provided by users when using the
service or generated through the use of the services
417
.
The drawing up of codes of conduct by providers of online intermediation services together with
business users and by online search engines shall be encouraged in order to ensure the proper
application of the P2B Regulation
418
.
3. Enforcement and penalties
The Member States must ensure accurate and effective enforcement of the P2B Regulation
419
. They are
not obliged neither to create new enforcement bodies (they can entrust existing authorities for
ensuring such an implementation) nor to provide for ex officio enforcement or to impose fines
420
. Each
Member State determines which measures shall apply to infringements to the P2B Regulation. Those
measures must be effective, proportionate and dissuasive
421
.
c) DSA Proposal
1. Nature and scope
The Commission’s proposed Digital Services Act
422
(DSA) takes the form of a Regulation, which means
that it will be directly applicable in the Member States’ legal order. The DSA covers technical
intermediaries (such as internet access providers), hosting service providers (which include advertising
servers), online platforms (that disseminate content to the public) as well as very large online platforms
(VLOP) that have more than 45 million monthly active users in the EU
423
.
2. Substantial measures (relevant for targeted advertising)
The proposed DSA contains rules which differ according to the size of the online platform.
The rules are aimed to make sure that users receive more information to help them make use of their
rights as data subjects. This added transparency is also intended to enable the scrutiny by authorities
and vetted researchers on how advertisements are displayed and how they are targeted.
In addition to the requirements resulting from Article 6 of the eCommerce Directive, all online
platforms would be required
424
to ensure that the recipients of the service receive individualised
information so that they can identify, for each specific advertisement displayed to each individual
recipient, in a clear and unambiguous manner and in real time:
• that the information displayed is an advertisement;
416
P2B Regulation, Article 7.
417
P2B Regulation, Article 9.
418
P2B Regulation, Article 17.
419
P2B Regulation, Article 15(1).
420
P2B Regulation, Recital 46.
421
P2B Regulation, Article 15(2).
422
European Commission, 2020, Proposal of the Commission of 15 December 2020 for a Regulation of the European Parliament and of the
Council on a Single Market For Digital Services (Digital Services Act) and amending Directive 2000/31/EC, COM(2020) 825 final, 2020/0361
(COD)..
423
Article 1 of the proposed DSA.
424
Article 24 of the proposed DSA.
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
123 PE 662.913
• the (natural or legal) person on whose behalf the advertisement is displayed; and
• meaningful information about the main parameters used to determine the recipient to whom
the advertisement is displayed (for targeted advertising).
VLOPs would have to comply with additional rules and in particular would have to compile and make
publicly available through application programming interfaces (APIs) a repository containing certain
information, while also making sure that the repository does not contain any personal data. The
repository would have to contain at least the content of the advertisement; the person on whose behalf
the advertisement is displayed; the period during which the advertisement was displayed; whether the
advertisement was intended to be displayed specifically to one or more particular groups of recipients
of the service and if so, the main parameters used; the total number of recipients reached and, where
applicable, aggregate numbers for the group or groups of recipients to whom the advertisement was
targeted specifically
425
. The Commission would have to support and promote the development of
voluntary industry standards to ensure the interoperability of these repositories
426
.
The Commission would have to encourage and facilitate the development of codes of conduct at EU
level to support and complement the transparency obligations relating to advertisements with
transparency obligations that would go beyond the imposed rules
427
.
3. Enforcement and penalties
The Member States where the provider is mainly established would be in charge of enforcement.
428
Member States would have to designate one or more competent authorities who would be responsible
for the application and enforcement of the rules. One of these competent authorities would have to be
designated as the Digital Service Coordinator (DSC) at the national level. DSCs would be given a set of
far-reaching investigation and enforcement powers (including power to request information, on-site
inspections, power to accept commitments from providers and to make them binding, to order the
cessation of infringements and to impose fines)
429
.
For VLOPs, the proposal sets-out an enhanced supervision system, whereby the DSC of the country of
establishment could be asked either by the Commission or by a newly created European Board for
Digital Services to investigate a suspected infringement. Where infringements persist, the Commission
could itself intervene and the DSC would be removed from the case. In case of non-compliance with
the regulation, interim measures or binding commitments, the Commission can adopt a non-
compliance decision ordering the VLOPs to comply and may fine or impose periodic payments.
The level of fine is different according to whether VLOPs are involved or not.
• For VLOPs, the Commission would be able to impose fines up to 6% of the company total
turnover in the preceding financial year. Periodic penalty payments up to 5% of the company
average daily turnover in the preceding financial year per day may also be imposed under
conditions.
• For other providers, Member States would have to determine the rules and levels of fines but
the proposals sets that fines imposed by Member States could not exceed those specified for
VLOPs
430
.
425
Article 30 of the proposed DSA.
426
Article 34 of the proposed DSA.
427
Article 36 of the proposed DSA.
428
Article 38.1 of the proposed DSA.
429
Article 38 and 41 of the proposed DSA.
430
Article 42 of the proposed DSA.
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 124
d) DMA Proposal
1. Nature and scope
The Commission’s proposed Digital Markets Act (DMA) would also take the form of a Regulation
431
.
It
would apply to Core Platform Services (CPS) designated by the Commission as gatekeepers, which are
offered to business user or end-users that are located in the EU, irrespective of the place of
residence/establishment of the gatekeeper
432
. These CPS would be listed in an exhaustive manner in
the proposed DMA and include for instance:
• online intermediation services, as defined in the Platform to Business Regulation (see above);
• online search engines, also defined in the Platform to Business Regulation;
• online social networking services;
• video-sharing platform services (as defined in the Audiovisual Media Services Directive (see
below);
• number-independent interpersonal communication services, as defined in the European
Electronic Communications Code; and
• advertising services, including advertising intermediation services, as long as they are offered
by providers of the above services.
2. Substantial measures (relevant for targeted advertising)
Under the proposed DMA, gatekeeper platforms would have to submit to the Commission an
independently audited description of any consumer profiling techniques they use
433
. Also, they would
not be allowed to combine personal data sourced from these core platform services with personal data
from any other services offered by the gatekeeper or with personal data from third-party
services, unless the end user has been presented with the specific choice
434
. This could limit the ability
of gatekeeper platforms to engage in targeted and personalised advertising which draw on data from
multiple sources.
The other rules are mainly aimed at providing more transparency in the relationship between business
users i.e. between the gatekeepers and the advertisers and publishers:
• Gatekeepers would also have to provide information to advertisers and publishers (if they ask
for it) on the price paid by each of them, as well as the amount or remuneration paid to the
publisher, for the publishing of a given ad and for each of the relevant advertising services
provided by the gatekeeper
435
;
• Subject to regulatory dialogue, gatekeepers could also be asked to provide advertisers and
publishers, upon their request and free of charge, with access to the performance measuring
tools of the gatekeeper and the information necessary for advertisers and publishers to carry
out their own independent verification of the ad inventory
436
; and
431
European Commission, 2020, Proposal of the Commission of 15 December 2020 for a Regulation of the European Parliament and of the
Council on contestable and fair markets in the digital sector (Digital Markets Act), COM(2020) 842 final, 2020/0374 (COD).
432
Articles 1 and 2 of the proposed DMA.
433
Article 13 of the proposed DMA.
434
Article 5(a) of the proposed DMA.
435
Article 5(g) of the proposed DMA.
436
Article 6(g) of the proposed DMA.
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
125 PE 662.913
• Subject to regulatory dialogue, gatekeepers should also refrain from the ability to engage in
self-preferencing in relation for instance to the ranking of their products and services compared
to products and services of third parties
437
.
3. Enforcement and penalties
The European Commission would be the competent regulatory body to enforce the DMA
438
. The
Commission would have the power to develop, through delegated and implementing acts, further
aspects of the DMA, such as the list of gatekeepers’ obligations
439
. It would have the power to request
information from all undertakings, conduct on-site inspections and order interim measures. The
Commission could also make binding commitments proposed by gatekeepers. In case of non-
compliance, the Commission could order the gatekeeper to comply with its obligations and impose
fines of up to 10% of the gatekeeper’s total annual (global) turnover
440
. Other infringements (e.g. failure
to provide access to databases or algorithms, failure to notify that the provider meets the thresholds
to be presumed as gatekeeper) could lead to a fine of up to 1% of the gatekeeper’s total annual
turnover
441
. Further, the Commission could impose periodic penalty payments of up to 5% of the
average daily turnover
442
.
e) ePrivacy (and Proposal)
1. Nature and scope
The ePrivacy Directive
443
applies to the processing of personal data in the electronic communication
sector but it also contains rules on use of cookies, which might be necessary for personalised and
targeted advertising. These rules on cookies protect the terminal equipment of end users. A new e-
privacy Regulation
444
is currently being negotiated between the European Parliament and the Council.
The Council reached a general agreement on the text on 10 February 2021, which means that
negotiations between the European Parliament and the Council can now start.
2. Substantial measures (relevant for targeted advertising)
The main provision of interest is the one which provides that the storing of information, or the gaining
of access to information already stored on a device (e.g. third party cookies, accessing pictures on a
mobile phone) unless the user has given his consent; or when this storage is necessary for the
transmission of a communication, or to provide an information society service requested by the user
445
.
With regard to unsolicited communications (by email, for instance), ePrivacy Directive sets forth the
opt-in principle (prior consent of the users), with some exceptions
446
.
437
Article 6(d) of the proposed DMA.
438
Article 18 of the proposed DMA.
439
Article 37 of the proposed DMA.
440
Article 26 of the proposed DMA.
441
Article 26 of the proposed DMA.
442
Article 27 of the proposed DMA.
443
European Parliament and the Council of the European Union, 2002, Directive 2002/58/EC of the European Parliament and of the Council
of 12 July 2002 concerning the processing of personal data and the protection of privacy in the electronic communications sector
(Directive on privacy and electronic communications), OJ L 201 of 31.7.2002.
444
European Commission, 2017, Proposal for a Regulation of the European Parliament and of the Council concerning the respect for private
life and the protection of personal data in electronic communications and repealing Directive 2002/58/EC, COM(2017) 10 final -
2017/0003 (COD).
445
Article 15.3 of the ePrivacy Directive.
446
Article 13 of the ePrivacy Directive.
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 126
The regulation proposed by the European Commission foresees that the use of the processing and
storage capabilities of a device and the collection of information from that device (i.e. device
fingerprinting) is also prohibited (with the same exceptions except that first party web-audience
measuring would be allowed). Under the proposed regulation, software permitting electronic
communications (e.g. operating systems, browsers and other applications) would have to require
users, upon installation, to choose whether they want to prevent third parties from storing information
on their device (e.g. third party cookies); or processing information stored on their device
447
.
3. Enforcement and penalties
Enforcement is left to the Member States which need to specify the rules rules on penalties, including
criminal sanctions where appropriate. They must also ensure, without prejudice to any judicial remedy
which might be available, that the competent national authority and, where relevant, other national
bodies have the power to order the cessation of the infringements. Member States must ensure that
the competent national authority and, where relevant, other national bodies have the necessary
investigative powers and resources, including the power to obtain any relevant information they might
need to monitor and enforce national provisions adopted pursuant to this Directive
448
.
Under the proposed regulation, the authorities in charge would be the same as under the GDRP
449
.
Fines would also be set which may do up to 4% of the total world annual turnover of the company
450
.
General measures applicable to advertising
a) Unfair Commercial Practices Directive
1. Nature and scope
The Unfair Commercial Practice Directive
451
(UCPD) is a full/maximal harmonisation directive
452
;
therefore, the Member States cannot adopt national rules having a higher or lower consumer
protection level. The directive was recently amended by the Better Enforcement Directive
453
.
The UCPD applies to unfair business-to-consumer commercial practices. The concept of “commercial
practices” is defined broadly: advertising and marketing related to products are expressly included,
regardless of whether these activities are carried out online or offline
454
. Personal scope is limited to
B2C relationships (from a trader to a consumer)
455
.
447
Article 8 of the proposed ePrivacy Regulation.
448
Article 15a of ePrivacy Directive.
449
Article 18 of the proposed ePrivacy Regulation.
450
Article 23 of the proposed ePrivacy Regulation.
451
European Parliament and the Council of the European Union, 2005, Directive 2005/29 of the European Parliament and of the Council of
11 May 2005 concerning unfair business-to-consumer commercial practices in the internal market, OJ L 149 of 11.06.2005.. On this
Directive, please refer to European Commission, 2016, Commission Staff Working Document, Guidance on the implementation/application
of Directive 2005/29/EC on unfair commercial practices, SWD(2016) 163 final; J. Stuyck, J., and Straetmans, G., 2016,
(ed.), Commercial
practices, Gent, Larcier.
452
UCPD, Article 4.
453
European Parliament and the Council of the European Union, 2019, Directive 2019/2161 of the European Parliament and of the Council
of 27 November 2019 amending Council Directive 93/13/EEC and Directives 98/6/EC, 2005/29/EC and 2011/83/EU of the European
Parliament and of the Council as regards the better enforcement and modernisation of Union consumer protection rules, OJ L 328 of
18.12.2019.
454
UCPD, Article 3 and Article 2(d) (definition of the “commercial practice”).
455
UCPD, Article 2(a) and (b).
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
127 PE 662.913
2. Substantial measures
Three kinds of commercial practices are considered as unfair under UCPD and, therefore, prohibited
456
:
(i) misleading or aggressive practices listed in Annex I that must in all circumstances be regarded as
unfair; (ii) misleading or aggressive practices prohibited under semi-general clauses of Articles 6 to 9;
and (iii) practices prohibited under the general clause of Article 5(2). Various practices related to
advertising and, in particular, targeted advertising, may fall under the scope of the UCPD and should
therefore be assessed under this three step test. Among others, direct exhortation to children, included
in an advertisement, to buy a product, or persistent and unwanted solicitations by email must be
considered as unfair commercial practices under all circumstances
457
. Misleading omissions may also be
qualified as unfair practices (under the semi-general clause), especially when information requirements
prescribed by Union Law in the field of advertising or advertising are violated
458
. Following the
modifications introduced by the Better Enforcement Directive, the main parameters used for ranking
of product and their importance are considered as material information under UPCD if the consumers
have the possibility to search products on the basis of a query
459
.
Self-regulation and codes of conducts (especially relevant in the context of advertising)
460
are also
promoted.
3. Sanctions and enforcement
Adequate and effective means should be implemented by the Member States to ensure that the
practices carried out on the market comply with the rules of the UCPD. Special attention is paid to the
burden of proof of factual claims: the trader may be requested to provide the evidence and, in case he
fails to, the claim should be considered as inaccurate
461
. Effective, proportionate and dissuasive
penalties should also be laid down
462
. Following the Better Enforcement Directive, criteria to be taken
into account for the imposition of penalties are listed, and the possibility to impose fines calculated on
trader’s annual turnover is expressly mentioned.
b) Directive on misleading and comparative advertising
1. Nature and scope
Under the Directive on misleading and comparative advertising
463
, traders are protected against
misleading advertising
464
(from other traders), while the practice of comparative advertising
465
(in B2B
or B2C relationships) is permitted, under specific requirements.
456
UCPD, Article 5.
457
UCPD, Annex I.
458
UCPD, Article 7(5) and Annex II (where a non-exhaustive list of the Union Law instruments in relation to commercial communication is
provided).
459
UCPD, new Article 7(4a). This paragraph does not apply to providers of online search engines, as defined in Article 2 (6) of the P2B
Regulation. Cf. also new point 11a of the Annex, related to paid advertisement or payment specifically for achieving higher ranking of
products within the search results.
460
See for instance the ICC Code (2018).
461
UCPD, Article 12.
462
UCPD, Article 13.
463
European Parliament and the Council of the European Union, 2006, Directive 2006/114/EC of the European Parliament and of the Council
of 12 December 2006 concerning misleading and comparative advertising (codified version), OJ L 376 of 27.12.2006.
464
Directive on misleading and comparative advertising, Article 2(b).
465
Directive on misleading and comparative advertising, Article 2(c).
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 128
2. Substantial measures
In order to determine whether advertising is misleading, three main criteria, related to the
characteristics of goods and services, the price and the advertiser, are laid down in Article 3.
Comparative advertising is authorised, provided three conditions (e.g. it compares goods or services
meeting the same needs), and five prohibitions (e.g. it is not misleading or it does not create confusion
among traders), are respected
466
.
3. Enforcement and penalties
Some key principles, also laid down in the UCPD, can be found in the Directive on misleading and
comparative advertising: adequate and effective means to be taken by the Member States
467
, evidence
to be provided by the advertiser
468
. Voluntary controls, by self-regulatory bodies, are also encouraged.
c) Consumer Rights Directive
1. Nature and scope
As for the UCPD, the Consumer Rights Directive (CRD)
469
is a Directive of maximal harmonisation
470
.
Hence, Member States are not allowed to adopt national rules that have a higher or lower level of
consumer protection. The CRD was also amended in 2019, by the Better Enforcement Directive
471
.
The CRD applies to the conclusion of sales and services contracts, between consumers and professional
traders
472
. Advertising and marketing materials may be included in its scope, when they lead to the
conclusion of contracts between consumers and traders.
2. Substantial measures
The CRD ensures that consumers are provided with a minimum set of information before being bound
by a contract with a professional trader. Among other things, the trader must provide consumers with
information about the main characteristics of the goods or services, the identity of the trader, the total
price of the goods or services, etc.
473
. There are further transparency requirements in the context of off-
premises and distance contracts, including contracts concluded online
474
. Since some information
requirements must be fulfilled before the conclusion of the agreement, they could be relevant for
advertising materials, when such materials lead to the conclusion of contracts between consumers and
professional traders.
Following the modifications introduced by the Better Enforcement Directive, the CRD makes it
mandatory to inform consumers when prices were personalised by means of automated decision
466
Directive on misleading and comparative advertising, Article 4.
467
Article 5 of the Directive.
468
Article 7 of the Directive.
469
European Parliament and the Council of the European Union, 2011, Directive 2011/83 of the European Parliament and of the Council of
25 October 2011 on consumer rights, OJ L 304 of 22.11.2011.. On this Directive, please refer to European Commission, 2014, “DG Justice
Guidance Document concerning Directive 2011/83/EU of the European Parliament and of the Council of 25 October 2011 on consumer
rights, amending Council Directive 93/13/EEC and Directive 1999/44/EC of the European Parliament and of the Council and repealing
Council Directive 85/577/EEC and Directive 97/7/EC of the European Parliament and of the Council”, June 2014.
470
CRD, Article 4.
471
European Parliament and the Council of the European Union, 2019, Directive 2019/2161 of the European Parliament and of the Council
of 27 November 2019 amending Council Directive 93/13/EEC and Directives 98/6/EC, 2005/29/EC and 2011/83/EU of the European
Parliament and of the Council as regards the better enforcement and modernisation of Union consumer protection rules, OJ L 328 of
18.12.2019.
472
CRD, Article 3.
473
CRD, Article 5.
474
CRD, Article 6.
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
129 PE 662.913
making, before the conclusion of a distance contract
475
. Since prices may be automatically personalised
in advertising materials displayed to consumers, this modification of the CRD may be of relevance. In
addition, if the contract is concluded on online market places, the provider must inform consumers
about the main parameters used for the ranking of offers and their relative importance
476
. Where the
ranking of offers is notably based on advertising, this provision of the CRD may prove relevant as well.
3. Enforcement and penalties
As for the UPCD, adequate and effective means should be implemented by the Member States to
ensure the compliance with the Directive
477
. The burden of proof is upon the traders, regarding their
information obligations
478
. In addition, the Directive states that effective, proportionate and dissuasive
penalties should be laid down in national laws
479
. Following the modifications brought by the Better
Enforcement Directive, criteria to take into account for the imposition of penalties are now listed in the
CRD, and the possibility to impose fines calculated on trader’s annual turnover is detailed
480
.
d) GDPR
1. Nature and scope
The General Data Protection Regulation
481
(GDPR) applies to the (wholly or partly automated
processing of personal data
482
. Both processing and personal data are broadly defined concepts that
encompass a large scope of operations, including profiling
483
, which imply use of information relating
to identified or identifiable natural person
484
. Personal scope of the Regulation is not limited to B2C
relationships. The Regulation applies to relationship between data controllers (i.e. any entities
determining purposes and means data processing operations), processors and data subjects
485
. Hence
this Regulation applies to advertising practices as long as they imply use of personal data.
2. Substantial measures
Any data processing operation must comply with a set of principles enshrined in Article 5 of the
Regulation which include transparency fairness and purpose limitation. Controllers must be able to
demonstrate compliance with these principles
486
and ensure that processing operation are by design
compliant with these principles
487
. Data processing must also rely on a lawful ground enshrined in the
GDPR. In this context, and depending on intrusiveness of the processing, legitimate interest of the
controllers or consent of the data subject seems to be the only available grounds for processing related
475
CRD, new Article 6, paragraph 1, (ea).
476
CRD, new Article 6a, paragraph 1, (a).
477
CRD, Article 23.
478
CRD, Article 6.
479
CRD, Article 24.
480
CRD, new Article 24.
481
European Parliament and the Council of the European Union, 2016, Regulation 2016/679 of the European Parliament and of the Council
of 27 April 2016 on the protection of natural persons with regard to the processing of personal data and on the free movement of such
data, OJ L 119 of 04.05.2016.
482
GDPR, Article 2 (1).
483
GDPR, Article 4( 4) (definition of “profiling”).
484
GDPR, Article 4(1) (definition of “personal data”) and Article 4(2) (definition of “processing”). Also see GDPR, Recital 26 on the broad
definition of “personal data”.
485
GDPR, Article 4(7) (definition of “controller”) and Article 4, 8) (definition of “controller”). The GDPR also applies to joint controllers scenarios
(i.e. when two or more entities jointly determine the purposes and means of processing operations). See GDPR, Article 26.
486
GDPR, Article 5(2).
487
GDPR, Article 25.
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 130
to targeted advertising
488
. Use of sensitive data, even inferred
489
, such as health data, or political
opinion related data also requires explicit consent. Requirements for valid consent are however
demanding
490
. In addition to information right
491
, and depending on the lawful ground for processing,
controllers must grant data subjects rights which empower users facing online advertising. These rights
include the right to withdraw consent, the right to erasure and the right to object to processing for
marketing purposes
492
. Among their various obligations, controllers must assess the risks to the rights
and freedoms carried by their processing activities and may be required to conduct a data protection
impact assessment (DPIA)
493
. DPIA may be necessary, for instance when advertising practices target
vulnerable data subjects such as children
494
and imply profiling and combination of datasets from
different sources
495
. Finally, the GDPR entails data subjects with the right not to be subject to fully
automated decision to present adverting with far reaching effects if it, for instance take profit of the
data subject’s vulnerabilities
496
.
3. Enforcement and penalties
Member States are required to establish at least one independent supervisory authority to monitor the
application of the GDPR
497
. National supervisory authorities have the power to adopt corrective
measures including ban on processing activities and imposing effective proportionate and dissuasive
administrative fines
498
. Administrative fines may be calculated on the basis of the annual turnover of
data controllers/processors
499
. Criteria to be taken into account for the imposition of penalties are also
listed
500
. Additionally, the Regulation ensures that data subjects can obtain compensation for any
damage caused by an infringement to the GDPR requirements
501
. Member States shall lay down
effective, proportionate and dissuasive penalties for additional infringement such as those which may
not be subject to administrative fines
502
.
e) AVMS
1. Nature and scope
The Audiovisual Media Services Directive as amended in 2018
503
is a minimum harmonisation directive,
meaning that the Member States are allowed to introduce more detailed or stricter provisions than the
minimum set of rules it contains.
488
See European Data Protection Board, 2020, Guidelines 8/2020 on the targeting of social media users, 2 September 2020; European Data
Protection Board, 2019, Guidelines2/2019 on the processing of personal data under Article 6(1)(b) GDPR in the context of the provision
of online services to data subjects, 8 October2019.
489
S. Wachter, S.,2020, ‘Affinity Profiling and Discrimination by Association in Online Behavioural Advertising’, pp. 17-22.
490
GDPR, Article 4.11) (definition of “consent”).
491
GDPR, Article 13 and Article 14.
492
GDPR, Article 7(3), Article 17 and Article 21(1).
493
GDPR, Article 35.
494
GDPR, Recitals 38 and 75.
495
See Article 29 Data Protection Working Party, 2017, Guidelines on Data Protection Impact Assessment (DPIA) and determining whether
processing is “likely to result in a high risk” for the purposes of Regulation 2016/679, 4 October 2017, WP 248rev.01.
496
GDPR, Article 22; Article 29 Data Protection Working Party, 2018, Guidelines on Automated individual decision-making and Profiling for
the purposes of Regulation 2016/679, 3 October 2018, WP251rev.01.
497
GDPR, Article 51(1).
498
GDPR, Article 58(2).
499
GDPR, Article 83(4), Article 83(5) and Article 83(6).
500
GDPR, Article 83(2) and Article 83(3).
501
GDPR, Article 82.
502
GDPR, Article 84.
503
European Parliament and the Council of the European Union, 2010, Directive 2010/13/EU of the European Parliament and of the Council
of 10 March 2010 on the coordination of certain provisions laid down by law, regulation or administrative action in Member States
concerning the provision of audiovisual media services (Audiovisual Media Services Directive) (codified version), OJ L 095 of 15.4.2010.
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
131 PE 662.913
It was amended in 2018 and now includes rules that need to be complied with by video-sharing
platforms (VSPs). VSPs are defined as providers of a service (provided for remuneration) or a dissociable
section of a service; whose principal purpose “or an essential functionality” is to provide programmes,
user-generated videos or both to inform, entertain or educate the public; where no editorial
responsibility is exercised by the provider who however determines its organisation, including by
automatic means or algorithms (in particular by displaying, tagging and sequencing); by an electronic
communication network.
2. Substantial measures
On advertising, there are two sets of rules
504
.
First, for commercial communications that marketed, sold or arranged by the platforms themselves,
they need to respect the basic qualitative standards that apply to linear and linear broadcasters
505
i.e.
they cannot:
• use surreptitious, use subliminal techniques (or must be readily recognisable as such);
• prejudice respect for human dignity, discriminate (or promote discrimination) on the basis of
sex, racial or ethnic origin, nationality, religion or belief, disability, age or sexual orientation, or
encourage behaviours prejudicial to health or safety or to the protection of the environment;
• promote under prescription medicines or medical treatments, cigarettes (electronic cigarettes
and refill containers) and other tobacco products, or promote alcohol by targeting specifically
minors or by encouraging immoderate consumption; and
• harm minors physically, morally or mentally, show them in dangerous situations, directly
exhort them to buy or hire a product/service by exploiting their inexperience or credulity, or
directly encourage them to persuade their parents or others to buy the product/service, or
exploit the trust they have in parents, teachers, etc.
Second, for other commercial communications (uploaded by users), VSPs need to take appropriate
measures to make sure that users comply with the standards listed above. These measures can be for
instance to include these standards in the service terms and conditions, having a functionality for
uploaders to declare whether videos contain commercial communications (as far as they know or can
be reasonably expected to know) and clearly informing viewers where content includes commercial
communications, provided the unloader has declared it or there is knowledge. The measures must be
proportionate and practicable taking into account the size of the VSP and the nature of the service.
3. Enforcement and penalties
According to the AVMS Directive, the assessment of whether the measures are appropriate needs to
be carried out by the national regulatory authority of the Member State where the VSP is established
506
.
Availability of out of court redress needs to be ensured. For the rest (penalties and other sanctions),
Member States should decide.
504
Article 28b of AVMS Directive.
505
Article 9(1) of AVMS Directive.
506
Article 28b of AVMS Directive.
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 132
Tables
1. Nature and scope
Nature
Personal scope
Material Scope
Specific
eCommerce DIR
Directive (min
harmonisation)
B2B and B2C
(Society information service
provider to the recipient of an
information society service)
Commercial communications
which are part of, or constitute, an
information society service
P2B Regulation
Regulation
Platform to Business (only when
business users target
consumers)
Practices such as ranking,
differentiated treatment and
access to data
DSA Proposal
Regulation
Technical intermediaries (ISPs)
Hosting service providers
Online platforms
Very large online platforms
Intermediary services, information
society services
DMA Proposal
Regulation
Gatekeepers
Core Platform service, information
society service, digital sector,
online intermediation services,
etc.
ePrivacy (incl.
Proposal)
Directive (min
harmonisation)
Regulation
Terminal equipment of end-
users
Processing of personal data
General
UCPD
Directive (full
harmonisation)
B2C
Commercial practices (incl.
marketing and advertising) -
broad
Misleading and
Comparative
Advertising DIR
Directive
B2C and B2B (depending on
provisions)
Misleading and Comparative
Advertising
CRD
Directive (full
harmonisation)
B2C
Conclusion of B2C contracts (incl.
obligations of information before
conclusion of B2C contracts) -
broad
GDPR
Regulation
Relationships between data
controllers/processor and data
subjects
Processing of personal data (both
broad concepts)
AVMS
Directive (min
harmonisation)
Video sharing platforms – end
users
Advertising (content rules)
Source: Authors’ own elaboration.
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
133 PE 662.913
2. Substantial measures
Information
duties
Prohibition
of
misleading
practices
Prohibition
of
aggressive
practices
Opt-in
Opt-out
Code of
conducts
(or self/co
regulation
promoted)
Others
Specific
eCommerce
DIR
v
v
v
P2B
Regulation
v
v
DSA
Proposal
v
v
DMA
Proposal
v
ePrivacy
(incl.
Proposal)
v
v
General
UCPD
v
v
v
Misleading
and
Comparative
Advertising
DIR
v
CRD
v
GDPR
v
v
(if
processing
based on
consent)
v
(opposition
right + right
to withdraw
consent)
Regulatory
body at EU
level
AVMS
v
v
Source: Authors’ own elaboration.
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 134
3. Enforcement and penalties
Specific
Enforcement
Body (in
addition to
traditional
courts)
MS shall take
enforcement
means
Criteria
for
penalties
Possible
fines
calculated
on
turnover
Burden of
proof
(and
risks) on
the
provider
Civil
sanctions
Self-
regulatory
bodies
Specific
eCommerce
DIR
v
v
P2B
Regulation
v
v
DSA Proposal
v
v
v
v
v
DMA Proposal
v
(Commission
to enforce)
v
v
ePrivacy (incl.
Proposal)
v
v
v
v
(proposal)
General
UCPD
v
v
v
v
Misleading
and
Comparative
Advertising
DIR
v
v
v
CRD
v
v
v
v
GDPR
v
v
v
v
AVMS
v
v
v
Source: Authors’ own elaboration.
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
135 PE 662.913
ANNEX 2: CASE STUDIES
Australia
1. National context
Consumer protection seems to be at the forefront of the need for a regulatory response to online
advertising in Australia. Country-specific issues identified in academic literature
507
concern
predominantly industry transparency, disclosure, consent processes and compliance practices.
The Australian Consumer Law (ACL) is part of the Competition and Consumer Act 2010, which aims
to protect consumers and ensure fair trading in Australia. It applies in the same way to all sectors and
in all Australian jurisdictions, i.e., all consumers in Australia enjoy the same rights and all businesses
have the same obligations, irrespective of which state or territory they engaged in transactions. The
ACL covers general standards of business conduct, prohibits unfair trading practices, regulates specific
types of business-to-consumer transactions, provides basic consumer guarantees for goods and
services, and regulates the safety of consumer products and product-related services
508
.
To tackle the issues arising from online advertising in both business-to-business (B2B) and business-to-
consumer (B2C) relations, the ACCC made a number of recommendations to the Australian
Government as part of the Final Report of the Digital Platforms Inquiry
509
. A number of these
recommendations have already been implemented by the Australian Government.
Most notably, the News Media Bargaining Code
510
, overseen by the ACCC and the Australian
Communications and Media Authority (ACMA), is a mandatory legislative code of practice aiming to
address the bargaining power imbalance between news media businesses and major digital platforms
such as Google and Facebook. It promotes fair and transparent negotiating practices in reaching
commercial arrangements with Australian news media businesses. The Code constitutes a significant
microeconomic reform highlighting the importance of well-resourced media diversity in Australia
511
.
While the Australian Government has not yet ‘designated’ any digital platforms to participate in the
scheme, the passing of legislation in February 2021 has already encouraged both Google and Facebook
to successfully negotiate voluntary commercial agreements with a number of large and small
Australian news businesses.
Other measures relevant to online advertising include continued inquiries by the ACCC of core
competition issues arising from online advertising
512
or information and privacy issues
513
addressed by
the Australian Information Commissioner (OAIC).
507
Mathews-Hunt, K., 2019, ‘CookieConsumer: Tracking online behavioural advertising in Australia’, Computer Law & Security Review 32, pp
55-90.
508
The Australian Consumer Law, Schedule 2 of the Competition and Consumer Act 2010.
509
ACCC, 2019, Digital Platforms Inquiry: Final Report.
510
ACCC, 2021, News media bargaining code.
511
Australian Government, 2018, Parliament passes News Media and Digital Platforms Mandatory Bargaining Code, media release.
512
ACCC, 2021, Lack of competition in ad tech affecting publishers, advertisers and consumers.
513
Office of the Australian Information Commissioner, Targeted advertising: Your Rights.
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 136
2. Design and implementation of response
Several market sector inquiries were launched by the ACCC to better understand the specific issues
pertaining to online advertising. The Digital Platforms Inquiry between December 2017 and July
2019 looked at the effect that digital search engines, social media platforms and other digital content
aggregation platforms have on competition in media and advertising services markets
514
. The final
report published in 2019 contains recommendations reflecting the issues arising from the growth of
digital platforms
515
. The recommendation to address the imbalance in the bargaining relationship
between these platforms and news media businesses recently materialised in the adoption of the News
Media Bargaining Code.
Another noteworthy inquiry is the inquiry into markets for the supply of digital platform services,
which commenced in February 2020 and will continue until March 2025. The first report of this inquiry
was released in September 2020 and provides an in-depth focus on online private messaging services
in Australia. The second report, which considers mobile app marketplaces, is to be publicly released by
the end of April 2021
516
. A separate ACCC inquiry into markets for the supply digital advertising
technology services and digital advertising agency services released an interim report earlier this
year, with a final report due to be publicly released in September
517
.
Overall, many different regulatory measures have been adopted in Australia, some of which have not
become law yet, since industry players are generally hesitant before any significant change. The
Australian government was quite positive and accepted vast majority of proposals made by the ACCC’s
Digital Platforms Inquiry. The cooperation between the key regulators and Australian Government
Departments, i.e. the ACCC and the ACMA, the Office of the Australian Information Commissioner, the
Treasury and the Department of Infrastructure, Transport, Regional Development and
Communications (DITRDC) as well as cooperation with overseas regulators went smoothly.
In the case of the Bargaining Code, it was a challenge arriving at regulatory settings that struck an
appropriate balance between the interest of different stakeholder groups, including digital platforms,
news businesses and consumers – including around complex and contested issues such as commercial
negotiations and data sharing of digital platforms. Even though the existing legislation had to be
adjusted before adopting the Code, it took less than a year to adopt it in an accelerated procedure,
since the Government requested an urgent response to the commercial challenges being faced by
Australian news businesses
518
.
While the speed of adopting the Bargaining Code turned out to be a challenging factor due to the need
to get all the details right, other measures have taken longer to adopt, often because many
stakeholders are usually involved. Bringing the necessary coalition of stakeholders affected by the
measures can be time-consuming, but also helpful
519
.
In its work on the Digital Platforms Inquiry and the Bargaining Code, the ACCC held a large number of
meetings with the most affected stakeholders such as digital platforms, media outlets, other industry
stakeholders and consumer groups to develop an accurate understanding of all relevant policy issues
514
ACCC, 2019, Digital Platforms Inquiry: Final Report.
515
Ibid.
516
ACCC, 2021, Digital platform services inquiry 2020-2025.
517
ACCC, 2021, Digital advertising services inquiry.
518
Interview with Andrew Sudol and David Abkiewicz, Directors of the Digital Platforms Branch of the Australian Competition & Consumer
Commission (ACCC), 30/03/2021.
519
Ibid.
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
137 PE 662.913
and ensure the lowest possible burden when it comes to complying with the recommended regulatory
measures
520
.
The aim of many recommendations of the Digital Platforms Inquiry was to address privacy concerns,
improve consumer protection, facilitate consumer choices, and boost competition, due to the ACCC’s
finding that there is significant overlap and interaction between these goals. For example,
strengthened privacy and data protection laws can also empower consumers to make more informed
choices about how their data is processed. This, in turn, is likely to increase competition between digital
platforms regarding the privacy dimension of their services. It may also encourage the emergence of
alternative business models that generate value for, and from, consumers in other ways
521
.
3. Analysis and Evaluation
Before any legislation is passed into law in Australia, a regulatory impact statement must be issued.
While measures aimed at promoting competition have a clear objective, it is hard to assess the results
of the News Media Bargaining Code measure as it is too early in the process. While none of the other
DPI recommendations have been made into law yet, when it comes to digital platforms issues, there is
still robust public debate regarding data use and privacy.
The most relevant Australian consumer law court cases related to issues of online advertising
concerned alleged misleading conduct of digital platforms. In a case against Google from 2013, the
Federal Court decided that search engines are not liable for misleading conduct of advertisers. The
decision was subsequently upheld by the High Court
522
. In another case, the ACCC successfully argued
against misleading conduct through algorithmic ranking of hotel room rates by Trivago which
constituted a breach of the ACL
523
. Two cases concerning alleged misleading conduct relating to user
data collection against Facebook
524
and Google
525
respectively are still pending before the Federal
Court.
520
Ibid.
521
Ibid.
522
ACCC, 2013, Google appeal upheld, media release.
523
ACCC, 2020, Trivago misled consumers about hotel room rates, media release.
524
ACCC, 2020, ACCC alleges Google misled consumers about expanded use of personal data, media release.
525
Ibid.
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 138
Germany
1. National context
The main issues emerging from online advertising in Germany are trademarks and copyrights
protection, market manipulation, unfair competition, youth protection, consumer protection,
data protection and protection of personal rights
526
. These challenges are the result of misleading
and aggressive practices, such as email ads, “key words ads”, pop-up advertisement, exit-intent pop-
ups, layer-ads, interstitials, re-marketing and ads on social media
527
. The increasing amount
528
of
omnipresent, personalised and misleading advertising, particularly via social media, is seen to bring
considerable damages to consumers
529
. The most important legislation on online advertising is the law
against unfair competition (Gesetz gegen den unlauteren Wettbewerb, UWG). The UWG prohibits
advertising that unduly constrains consumers’ freedom of choice. This is the case if as a result of an
aggressive business act by the entrepreneur the consumer is led to undertake a business act, which it
would have not undertaken otherwise. In particular, an undue influence is considered the misuse by
the entrepreneur of its position of power to exert pressure on the consumer and, as a result, constrain
its ability to make an informed and thoughtful decision
530
. The General Data Protection Regulation
(GDPR) is the relevant regulatory response as regards aspects related to data protection. To be
considered are also youth protection regulations and copyright and trademark laws, the Telemedia Act
and the penal law.
2. Design and implementation of response
The current version of the UWG, initially elaborated in 2004, is the result of two major reforms in
2008 and 2015
531
. The initial version integrated, as a result of the active participation of the civil society
and academia in the course of the elaboration of the regulation
532
, several national laws into a unified
legal framework, codified important case law, considerably increasing consumer protection by
introducing a ‘black list’ and led to a high degree of liberalisation of competition law
533
. Significantly,
the UWG of 2008 harmonised Germany’s legislation with EU law (most significantly Directive
2005/29/EG) and strengthened the EU single market
534
.
The UWG is not being implemented by authorities ex officio. Rather, it is enforced by the market
participants in court via prohibitory injunctions. The latter is a rather unusual legal instrument in
the German civil law in so far as it has legal effects for both process participants and third parties. The
exclusively competent court of first instance is the district court (Landgericht). A more cost-effective
alternative is the arbitration for dispute settlements at the chambers for industry and commerce.
However, only traders who are in direct competition with the defendant and associations promoting
commercial and industrial interests are entitled to apply.
526
Buse, N., 2021, Onlinewerbung- was ist erlaubt, was ist verboten?, Buse Herz Grunst Rechtsanwälte.
527
Ibid.
528
Weber, M., 2020, Werbemarkt in Deutschland um 1,9 Prozent gewachsen, W&V.
529
Interview, Federal Ministry of Justice and Consumer Protection, 30.03.2021.
530
Ibid. (Fn. 1).
531
Teplitzky, O., Peifer, K.-N., & Leistner, M., 2020, Großkommentar zum Gesetz gegen den unlauteren Wettbewerb mit Nebengesetzen,
Kommentar, Article 575.
532
Köhler, H., Bornkamm, J., Henning-Bodewig, F., 2002, Vorschlag für eine Richtlinie zum Lauterkeitsrecht und eine UWG-Reform | WRP 2002,
1317-1328.
533
Wilde Beuger Solmecke Rechtsanwälte, 2021, Rechtsanwälte, Irreführende Werbung.
534
Business-Wissen, 2008, Einheitliche Werberegeln für die ganze EU.
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
139 PE 662.913
Moreover, the German system is characterised by self-regulation. Two advertising standards
organisations – the German Advertising Standards Council and the Wettbewerbszentrale – deal with
issues of social responsibility and unfair commercial practices by applying the UWG
535
.
According to Article 2(1) no. 6, the UWG targets all natural and legal persons that undertake a business
act within the framework of their commercial, trading or professional activities and all persons that act
in the name of or on behalf of these persons. Consequently, the UWG has a broad scope of application
and targets advertising networks, advertising firms and agencies, publishers and publishing platforms.
The scope of this liability is not clearly determined by the UWG and is subject to case law
536
. The
overall objective of the UWG is to protect consumers from undue business acts and the public interest
in an undistorted competition (Article 1 UWG). The GDPR, which is directly applicable in Germany,
focuses on aspects related to data protection and applies to targeted email advertising (Recital 47)
537
.
Following its overall objective, the UWG exclusively targets entrepreneurs and provides a ‘black-list’ of
undue business actions in the attachment to Article 3(3) UWG, which includes both misleading and
aggressive commercial practices
538
. The rationale of this ‘black-list’ is to protect the consumers’
freedom of choice and its right to make an informed and thoughtful decision.
3. Analysis and Evaluation
The UWG has mainly implemented Directive 2005/29/EG, which sets EU-wide minimum standards, into
national law. Going beyond the scope of application of the directive, Article 3(1) UWG covers C2B
(consumer to business) relations, the point of reference being the perceptible impairment of consumer
interests. In contrast, Article 5 of Directive 2005/29/EG more generally prohibits undue business
practices
539
. Moreover, Article 7(2) no. 3 UWG already sees an unreasonable harassment in cases where
businesses send consumers email ads without the latter’s prior explicit consent, whereas the EU
requirements do not foresee such an ‘opt-in’ (Annex I, point 26)
540
. In response to misleading or
aggressive commercial practices, the UWG foresees remedy claims and prohibitory injunctions (Article
8), claims for damages (Article 9) and claims concerning the profits skimmed as a result of undue
business practices. Moreover, as a result of data protection violations, the consumer has non-material
damage claims, which result from the general right of privacy.
Consumers viewed the development of the UWG favourably
541
. This can be attributed to the high
quality of the regulatory response and consumers’ high willingness to pay “free of charge” services with
private data
542
. While the regulatory response has had a positive impact on consumer protection and
established a level playing field for market competitors
543
, the regulatory response is generally seen
as being insufficient, in particular, given the surge of online advertising during the COVID-19
pandemic. The advertising market share of online advertising has increased from 12% in 2019
544
to 40%
in 2020
545
.
535
European Advertising Standards Alliance, 2021, Members: Germany.
536
Huber, D., 2021, Werbung im Internet: Ist das rechtlich zulässig?, IT-Recht Kanzlei zum Werberecht.
537
Schürmann, K., 2020, E-Mail-Marketing ohne Einwilligung – was ist erlaubt?, Schürmann Rosenthal Dreyer Rechtsanwälte.
538
Gesetz gegen den unlauteren Wettbewerb (UWG), Annex toArticle 3(3).
539
Wettbewerbszentrale, 2010, Synopse UWG – Richtlinie unlautere Geschäftspraktiken (UGP-RL).
540
Ibid, p. 18.
541
Interview, Verbraucherzentrale Bundesverband, 06.04.2021.
542
Interview, Federal Ministry of Justice and Consumer Protection, 30.03.2021.
543
Interview, Verbraucherzentrale Bundesverband, 06.04.2021.
544
Statista, 2020, Marktanteile der einzelnen Werbemedien im deutschen Bruttowerbemarkt im Jahr 2019.
545
Giersberg, F., 2020, ZAW: Werbemarkt 2020 mit deutlichen Corona-Auswirkungen.
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 140
The German Advertising Association demands new solutions to the rising dominance of online
advertisement at national and European level
546
. There is extensive case law on online advertising.
Most recent cases have dealt with ads on the social media platform “Instagram”. In particular, courts
have decided that the identification of advertising via the hashtag “#ad” is insufficient if it appears at
the end of a long list of hashtags
547
, that for “business-like acts” in the sense of the UWG the objective
external effect is decisive
548
and that the identification of advertising is dispensable if the advertising
account is public (blue checkmark) and has a high number of followers
549
.
546
Ibid.
547
OLG Celle, 08.06.2017, Az. 13 U 53/17; KG Berlin, 11.10.2017, Az. 5 W 221/17.
548
LG Itzehoe, 23.11.2018, Az. 3 O 151/18; OLG Braunschweig, 08.01.2019, Az. 2 U 89/19.
549
LG München I, 29.04.2019, Az. 4 HK O 14312/18.
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
141 PE 662.913
France
1. National context
In 2018, with 40% of advertising investments, digital is now the leading medium for advertising sales
in France, compared to 27% for television
550
. As mentioned by a recent report commissioned by the
French Minister of Culture and the Secretary of State in charge of digital, a significant transfer of
advertisement’s investments from television and traditional media to digital and online platforms
occurred in the last years. Oliver Wyman who was commissioned by the Syndicat des Régies Internet
(SRI) to provide an estimate of the size of the digital advertising market in France estimates the growth
of the digital advertising market in France at +7% for the whole of 2021
551
.
There is no specific legal framework for online advertising in France. The law that applies is
therefore that of advertising in general. Main rules applicable to online advertising and are also
applicable to traditional/offline advertising. It has to be noted however that self-regulation is
prevalent in the field of online advertising in France as it has proven to adapt quickly to a very fast-
changing market. In terms of regulation, Decree No. 2017-159 of Feb. 9, 2017
552
extends a law of
1993 “against corruption and in favour of transparency for advertising” (so-called ‘Sapin Law’) to online
advertising (i.e. advertisements conveyed on any devices connected to the internet) to
apply transparency obligations that currently apply to traditional advertising. In terms of self-
regulation, the ARPP code of conduct on online advertising, mentioned above, contains
recommendations applicable to some digital format/techniques, such as social media. For example,
on social media networks, the ARPP recommends adding an explicit notice, if the commercial nature
of the advert is not obvious
553
.
According to the media, the extension of transparency obligation set up in the “Sapin Law” to online
advertising (Decree No. 2017-159 of Feb. 9, 2017) has been welcomed by French advertisers
554
. An
Article from Les Echos indeed states that by introducing regulation into the free space that is the
Internet, “the publication of the decree now injects rigidity into online transactions. But perhaps it will
succeed in lightening an increasingly suffocating atmosphere of mistrust [in the online advertising field]”
555
.
2. Design and implementation of response
The actors involved in designing and implementing the French response to online advertising are not
specifically and solely working on online advertisement. These actors include the French competition
authority which publishes opinions on competition matters linked to online advertisement.
Additionally, the Commission Nationale de l'Informatique et des Libertés (CNIL) and the Conseil
Supérieur de l’Audiovisuel (CSA) are important actors as they organise consultation and publish
recommendations on the topic. Finally, the professional regulatory authority for advertising (ARPP)
may intervene during the elaboration of methodologies linked to online advertisement, such as digital
carbon footprint.
550
Report commissioned by the French Minister of Culture and the Secretary of State in charge of digital, 2020, Online advertising: A level
playing field.
551
Amiot, E., 2020, Digital Advertising in France Weakened by the Pandemic, Syndicat des Régies Internet (SRI).
552
French Legal Statute, 2017, Decree No. 2017-159 of February 9, 2017 on digital advertising services.
553
Autorité de régulation professionnelle de la publicité, 2015, Recommandations on digital commercial communications.
554
Les Echos, 2017, Decree on the transparency of online advertising finally published, News Article.
555
Ibid.
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 142
All French media are members of the ARPP, which allow all important stakeholders in the field of online
advertisement to propose good practices and elaborate recommendations, in a consultative manner.
3. Analysis and Evaluation
To regulate online advertising effectively and restore the balance between actors in the sector, the
report commissioned by the French Minister of Culture and the Secretary of State in charge of digital
highlights that it would be necessary to first “align the constraints between the traditional and digital
sectors on the one hand, and between online platforms and publishers on the other”
556
. To ensure a
level playing field between these actors, the report proposes to harmonise the legal framework
governing advertising on audiovisual media (linear or not) and display advertising on the Internet.
The report also found that traditional competition law is “complex to implement and sometimes too
slow in view of the speed with which the digital market is evolving”
557
. To solve this issue at national
level, it was suggested that the existing arsenal of competitive regulation be expanded and renewed
to deal effectively with the structuring nature of the major platforms, and thus act effectively on their
sources of market power and anti-competitive behaviour in the online advertising market
558
.
In France, there are no court cases specifically addressing online advertising, but the French
competition law authority took several decisions on the matter. For example, as concerns search
engine and advertising intermediation markets, in December 2017, the competition authority
rejected the complaint lodged by the company 1PlusV against Google for tying sale and discriminatory
treatment practices
559
. The French authority dismissed the complaint as 1PlusV failed to bring evidence
that:
• Google was using tying practices;
• there was no commercial partnership between the parties; and
• 1PlusV did not comply with the conditions to access Google’s services.
Another example of decision taken by the French competition authority concerns the online
advertising market, linked to search. In November 2016, the authority dismissed a complaint lodged
by the company IAH against Google’s AdWords referencing service
560
. Finally, the French competition
authority also has a consultative role and it published opinions on online advertising, such as the one
of 6 March 2018 on data exploitation in the online advertising sector
561
.
556
Report commissioned by the French Minister of Culture and the Secretary of State in charge of digital, 2020, Online advertising: A level
playing field.
557
Ibid.
558
Ibid.
559
Autorité de la concurrence, 2017, Decision 17-D-24 of 18 December 2017 concerning practices implemented in the online search engine and
online advertising intermediation sectors.
560
Autorité de la concurrence, 2016, Decision No 16-D-25 of 23 November 2016 on practices implemented in the online advertising sector online
advertising.
561
Autorité de la concurrence, 2018, Opinion n° 18-A-03 of 6 March 2018 on the use of data in the internet advertising sector advertising sector.
2018.
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
143 PE 662.913
Ireland
1. National context
In Ireland, consumer protection appears to be one of the most regulated issues relating to online
advertising. Privacy and data protection are also heavily regulated. Other issues which have arisen with
Online Behavioural Advertising are privacy, non-discrimination and group level protection
562
. The
review of the legal framework on online advertising and the interview show that issues Ireland faces
are similar to the issues faced by other Member States, with perhaps the exception of the volume of
claims the CCPC has to deal with, which has to do with the fact that a large number of online
platforms are based in Ireland. The law on online advertising is governed by the 2011 Electronic
Communications Networks and Services, Privacy and Electronic Communications Regulations
563
.
The Consumer Protection Act 2007
564
(CPA) is the main Irish law that covers advertising. The
Consumer Protection Act 2007 integrates the EU Directive on Unfair Commercial Practices
565
in Ireland.
The act includes rules on advertising in general, which also apply to online advertising. The main
aspects related to online advertising regulated by the act are misleading advertising and false
advertising directive. Other EU law applies in Ireland to legal advertising, namely the misleading
advertising directive
566
and the General Data Protection Regulation
567
. Finally, some sector specific
regulations also apply to online targeted advertising: The Consumer Credit Act 1995
568
; the Copyright
and Related Rights Act 2000
569
; the Trademarks Act 1996
570
; and the Patents Act 1992
571
.
In Ireland, self-regulation is the principal method of regulation of advertising. The Advertising
Standards Authority, an independent self-regulatory body set up and financed by the advertising
industry, publishes a Code of Standards
572
, which includes rules on online behavioural advertising,
binding for its members.
To tackle misleading advertising (including online advertising), the Competition and Consumer
Protection Commission (CCPC) prohibits a set of practices. The CPA established a national enforcement
body, the CCPC, which is responsible for dealing with citizens’ complaints. Citizens can also complain
to the Advertising Standards Authority for Ireland (ASAI), for any practice which breaches the ASAI
Code. An innovative area in which the code regulates is online behavioural advertising (OBA)
573
.
562
Wachter, S., 2019, ‘Affinity Profiling and Discrimination by Association in Online Behavioural Advertising’, Berkeley Technology Law Journal,
Vol. 35, No. 2.
563
Irish Statute Books, 2011, European Communities (Electronic Communications Networks and Services) (Privacy and Electronic
Communications) Regulations.
564
Irish Statute Book, 2007, Consumer Protection Act 2007.
565
European Parliament and European Council, 2005, Directive 2005/29/EC of the European Parliament and of the Council of 11 May 2005
concerning unfair business-to-consumer commercial practices in the internal market and amending Council Directive 84/450/EEC, Directives
97/7/EC, 98/27/EC and 2002/65/EC of the European Parliament and of the Council and Regulation (EC) No 2006/2004 of the European
Parliament and of the Council (‘Unfair Commercial Practices Directive’), Official Journal of the European Union.
566
European Parliament and European Council, 2006, Directive 2006/114/EC of the European Parliament and of the Council of 12 December 2006
concerning misleading and comparative advertising, Official Journal of the European Union.
567
European Parliament and European Council, 2016, Regulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016
on the protection of natural persons with regard to the processing of personal data and on the free movement of such data, and repealing
Directive 95/46/EC (General Data Protection Regulation), Official Journal of the European Union.
568
Irish Statute Book, 1995, Consumer Credit Act 1995.
569
Irish Statute Book, 2000, Copyright and Related Rights Act 2000.
570
Irish Statute Book, 1996, Trademarks Act 1996.
571
Irish Statute Book, 1992, Patents Act 1992.
572
Advertising Standards Agency for Ireland, ASAI Code, 7th edition.
573
Advertising Standards Authority for Ireland, 2016, Code, 7th edition Section 18.1.
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 144
The CCPC’s activity appears to be uncontroversial in the media and the legislative framework on
consumer protection provides a sufficient framework for the CCPC to take action if necessary.
2. Design and implementation of response
The ASAI Code was drawn up by the Board of ASAI after a consultation with key stakeholders including
industry representatives, the public, consumer representatives and the CCPC. Its current version came
into effect on the 1
st
of March 2016. The CCPC internally monitors certain areas where they suspect
consumer detriment might occur. The CCPC publishes a strategy paper
574
every three years, in which
the objectives from the previous strategy are reviewed, and new ones are determined. The CCPC
measures the number of Compliance Notices, Fixed payment Notices to bring about compliance and
end damaging practices. The CCPC also leads a number of investigations in the area of consumer
protection law
575
. To empower consumers directly, before any grief has been experienced, the CCPC
uses its website, as a channel to equip consumers to make informed choices. The number of visits on
the website is monitored, as well as the time spent on the website. The ASAI also monitors the respect
of the ASAI Code and the effectivity of its activity through complaints bulletins. The bulletins are
published periodically on the ASAI website by the Complaints Committee. They provide a summary of
the ASAI’s work on complaints and how it ensures ads in Ireland are lawful. It is a way for the
organisation to monitor its activity, in terms of the range and areas of complaints they deal with.
For complaints to the CCPC, consumers will start a procedure after they have made a commercial
decision. The CCPC will then enquire in order to decide whether the level of severity is important
enough to pursue further action against the advertiser. In case further action is sought, the CCPC can
either use fixed payment notices, prohibition notices or take enforcing action against the trader,
including compensation. Consumers can also decide to take action with the ASAI. The ASAI procedure
does not require a commercial decision to have been made. In that sense, it is sufficient for an ad to
breach the ASAI code for consumers to take action. If a breach is identified, the ASAI will contact the
advertiser to have them withdraw the ad. The decision is then made public.
Both the CCPC and the ASAI organise events where stakeholders including industry representatives
interact. For example, the ASAI recently organised the “#InfluencerMarketing in 2021 and beyond
576
”
webinar which included industry representatives and the CCPC. The CCPC also communicates with
businesses directly and regularly publishes guidance to help them deal comply with the law. For
example in 2016, the CCPC published the guide on how to sell online
577
to communicate with
consumers through warnings on their website. For example in 2020, they published a warning about a
website which engaged online sales. They advised consumers to exercise caution if buying goods from
it and set up a consumer helpline to address any issues
578
.
In both the ASAI code and the CPA, the rules on prohibition on unfair commercial practices target
traders and advertiser, including online traders and advertisers. Under the Consumer Protection Act
traders are defined as “a person who is acting for purposes related to the person’s trade, business or
profession, or a person acting on behalf of a trader. Those who promote goods or services on behalf of
a business may be considered a trader under consumer protection law”.
574
See Competition and Consumer Protection Commission, 2020, Strategy Statement 2021-2023.
575
Competition and Consumer Protection Commission, 2019, Annual Report 2019.
576
See: Advertising Standards Authority for Ireland, 2021, #InfluencerMarketing in 2021 and beyond, Webinar, 23rd February 2021.
577
See: Competition and Consumer Protection Commission, 2016, Selling Online: What you need to know, Business Guide.
578
See: Competition and Consumer Protection Commission, 2020, Consumer warning about irelandsstore .com, News.
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
145 PE 662.913
The ASAI Code targets advertisers, who are defined as “anyone disseminating marketing
communications, including promoters and direct marketers. References to advertisers should be
interpreted as including intermediaries and agencies unless the context indicates otherwise”
579
.
Regarding the definition of advertisers, the ASAI held a webinar
580
, during which it updated the
definition of advertiser to include influencers.
3. Analysis and Evaluation
The regulatory response seems to have a positive effect on unfair commercial practices, as a significant
number on complaints have been dealt with by the ASAI as shown by their complaints bulletins. For
example, the latest complaints bulletin contains 8 case reports, all of which were found in breach of the
ASAI Code. In that sense, the activity led by the ASAI seems to complement the CCPC, which deals
predominantly with more severe breaches which have already caused harm to consumers. The ASAI
also engages in preventive measures.
579
See for example, Section 1.1(d) of the ASAI Code.
580
See: Advertising Standards Authority for Ireland, 2021, #InfluencerMarketing in 2021 and beyond, Webinar, 23rd February 2021.
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 146
United Kingdom
1. National context
Consumer protection appears to be the dominant issue emerging from the debate around online
behavioural advertising (OBA) or targeted advertising in the UK. UK consumers are increasingly privacy-
conscious, with many consumers seeking to limit personalised targeting by setting privacy settings in
their browsers or online services
581
. Another aspect to consider is the costs of digital advertising
(amounting to around EUR 16.11 billion – GBP 14 billion – in the UK in 2019)
582
that are reflected in the
prices of goods and services, particularly those which make use of heavy digital advertising (for
example consumer electronics, hotels, and flights). Indicatively, Facebook (which is paid for indirectly
through advertising revenues) saw its average revenue per user increase from less than GBP 5
583
in 2011
to over GBP 50
584
in 2019
585
. Market manipulation and the costs of misleading actions and aggressive
commercial practices are another concern in the UK. In 2009, Consumer Focus estimated that
misleading and aggressive practices cost GBP 3.3 billion
586
.
The Consumer Protection from Unfair Trading Regulations 2008
587
, as amended in 2014 with the
Consumer Protection (Amendment) Regulations
588
, and the Consumer Rights Act 2015
589
form the
basis of the UK’s consumer protection legal framework. The 2008 Regulations are the instrument
implementing the Unfair Commercial Practices Directive into domestic law. In addition, The UK Code
of Non-Broadcast Advertising and Direct Promotional Marketing (CAP Code)
590
applies. While the
Unfair Trading Regulations 2008 made it a criminal offence to use misleading or aggressive commercial
practices, consumers had no private right of redress against a trader. The 2014 Amendment addressed
this, introducing standard remedies and an entitlement to seek damages when a consumer has fallen
victim to these practices. Moreover, the CAP Code supplements the gaps in the existing legal
framework, prohibiting both misleading and harmful advertising, as well as promoting social
responsibility around advertising of alcohol, motoring, lotteries, and gambling.
Generally speaking, the narrative around the Unfair Trading Regulations tends to be positive, in that
they are seen as protecting consumers from unfair or misleading trading practices
591
. That said, the UK’s
data protection regulator is concerned about the programmatic advertising process known as
real-time bidding (RTB), which makes up a large chunk of online behavioural advertising
592
. It
concludes that systematic profiling of web users via invasive tracking technologies such as cookies is
in breach of U.K. and pan-EU privacy laws.
581
PLUM, 2019, Online advertising in the UK, A report commissioned by the Department for Digital, Culture, Media and Sport.
582
Competition and Markets Authority, 2020, Online platforms and digital advertising: A Study.
583
This corresponds roughly to EUR 6 [exchange rate 31.12.2011].
584
This corresponds roughly to EUR 59 [exchange rate 31.12.2019].
585
Ibid.
586
Department for Business, Energy & industrial Strategy, 2018, Misleading and aggressive commercial practices: new private rights for
consumers. The value corresponds roughly to EUR 4.0 billion [exchange rate 31.12.2020].
587
See: UK Statute Book, 2008, The Consumer Protection from Unfair Trading Regulations 2008.
588
See: UK Statute Book, 2014, The Consumer Protection (Amendment) Regulations 2014.
589
See: UK Statute Book, 2015, Consumer Rights Act 2015. .
590
See: Advertising Standards Agency, 2014, Non-broadcast Code.
591
Which?, 2021, Consumer Protection from Unfair Trading Regulations 2008.
592
Tech Crunch, 2019, Behavioural advertising is out of control, warns UK watchdog, Article.
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
147 PE 662.913
2. Design and implementation of response
593
The 12
th
edition of the CAP Code came into force in September 2010. It was released because the digital
remit of the Advertising Standards Agency (ASA) was extended to cover online marketing
communications. Before this, Consumer Focus had in 2009 commissioned research into consumers’
experience of unfair commercial practices generally
594
. The Committee of Advertising Practice (CAP) is
the self-regulatory body that creates, revises and enforces the Code – it is responsible for monitoring.
The main actor involved was the Advertising Standards Agency and its CAP Committee. CAP's members
include organisations that represent the advertising, sales promotion, direct marketing and media
businesses. Through their membership of CAP member organisations, or through contractual
agreements with media publishers and carriers, those businesses agree to comply with the Code so
that marketing communications are legal, decent, honest and truthful and consumer confidence is
maintained
595
. One of CAP’s members is The Incorporated Society of British Advertisers (ISBA), which
has over 400 members representing nearly all the UK’s major advertisers.
It is the Advertising Standards Agency who is the UK's independent regulator of advertising across all
media
596
. They apply the advertising Codes, written by the CAP. The ASA is self-regulated and is a
non-statutory organisation, and so cannot interpret or enforce legislation. In addition to the CAP, the
system receives support through three industry panels – the General Media Panel (GMP), the Sales
Promotion and Direct Response Panel, and the Online Publications Media Panel. The CAP Code applies
to all actors advertising or promoting goods or services in 8 categories of activities, including print
(newspapers, magazines, brochures); materials in public places (posters, etc.); moving images (cinema,
video, etc.) non-broadcast electronic media in paid-for-space (banner, pop ups, online advertisements,
display, etc.); marketing databases which use consumer personal information; non-broadcast media,
advertorials and advertisements; and other marketing communications in non-paid-for space online
(e.g. own website)
597
.
The purpose of the CAP Code is to ensure that statements made about products or services are accurate
and not misleading. The CAP Committee itself says that compliance with the Code should mean that
“marketing communications are legal, decent, honest and truthful and consumer confidence is
maintained”
598
. The CAP’s purpose is therefore consumer-protection orientated. This is in line with the
aim of the Unfair Trading Regulations 2008, which is to give consumers simple, standardised remedies
against traders that have breached requirements. It aims to ensure that the previously relatively loosely
regulated online marketing field is more strictly regulated, although specific details with regards to
how it aims to change the business model are not clear. The response is focused on maintenance of
consumer confidence through proper treatment of data, protection from unfair practices and
standardised remedies against traders that have breached requirements.
3. Analysis and Evaluation
The value of self-regulation (inherent in the CAP Code) as an alternative to statutory control is
recognised in European legislation, including those on misleading and comparative advertising
(Directives 2005/29/EC and 2006/114/EC). Self-regulation is accepted by the Department for Business,
Innovation and Skills and Trading Standards as a first line of control in protecting consumers and the
593
Given that the 2008 Consumer Protection from Unfair Trading Regulations merely implements the EU’s Unfair Commercial Practices
Directive, the case study will mainly consider the ‘additional’ regulatory response, namely the CAP Code. Although the UK is no longer a
member of the EU, the substance of the implementing national legislation remains the same even after the end of the UK-EU transition
period.
594
Consumer Focus, 2008, Waiting to be heard: Giving consumers the right of redress over Unfair Commercial Practices.
595
See: Advertising Standards Agency, 2014, CAP Code: Preface.
596
See: Advertising Standards Agency, 2021, Website, https://www.asa.org.uk/.
597
Ibid.
598
Simply Business, 2011, Online marketing and the CAP Code – What you need to do.
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 148
industry
599
. Although ASA is non-statutory and so cannot enforce legislation, it can refer cases to
Trading Standards, which can then impose fines and criminal sanctions. It is important to note that
with its self-regulatory system, ASA depends upon the fact that practitioners in every sphere share an
interest in seeing that marketing communications are welcomed and trusted by their audience: unless
they are accepted and believed, marketing communications cannot succeed. If they are offensive or
misleading, they discredit everyone associated with them and the industry as a whole
600
.
Given that it goes beyond the scope of the Unfair Commercial Practices Directive, the CAP appears to
be effective in filling in gaps left by the legislation. In 2019, ASA resolved 34,717 complaints relating
to 24,886 advertisements, 70% of which were potentially misleading. The same year, 62 sanctions were
applied leading to compliance, and only 9 advertisers needed to be referred to Trading Standards for
further action to be taken
601
.
599
See: Advertising Standards Agency, 2014, CAP Code: Preface.
600
See: Advertising Standards Agency, The CAP Code: How the System Works.
601
Advertising Standards Agency, 2019, Using technology for good – Annual Report 2019.
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
149 PE 662.913
United States
1. National context
Consumer protection laws apply equally to online advertising and recent issues include space-
constrained screens and social media platforms. Privacy protection is also an important issue in the
United States, and is the object of some of the Federal Trade Commission’s regulations
602
. In 2020, there
were 3 top complaints related to online advertising (excluding identity theft)
603
: imposter scams where
someone gets tricked into sending money to someone else
604
; online shopping and negative reviews
(a lot of this was attributable to COVID-19, e.g., unreceived items, undisclosed costs, non-delivery, no
guarantees)
605
; and internet services.
The 1914 Federal Trade Commission Act empowers the Federal Trade Commission (FTC) to act in the
interest of consumers to prevent deceptive and unfair acts or practices. The FTC is also in charge of
enforcing the Children’s Online Privacy Protection Act
606
which applies to online advertising. The 1946
Lanham Act also contains some provisions on false advertising. The FTC also enforces a number of other
statutes and rules that relate to digital advertising. Examples include:
• The Restoring Online Shoppers’ Confidence Act, which prohibits any post-transaction third-
party seller from charging any financial account in an internet transaction without consent
and demands clearly displayed terms and conditions;
• The Mail, Internet, or Telephone Order Merchandise Rule, which requires sellers to reasonably
predict shipping time for products; and
• The Consumer Review Fairness Act, which protects consumers’ ability to share opinions about
a business’s products, services, or conduct, in any forum, including online.
The FTC provides advice to online companies and commercial websites. It published a report
607
in 1998
with guidance to companies that collect personal information from users. In February 2009, the FTC
released a revised set of principles for the self-regulation of online behavioural advertising
608
.
In the same area, the Obama administration issued the Consumer Privacy Bill of Rights which is
enforceable by the FTC and includes 7 principles intended to protect consumers’ rights and privacy.
In conjunction to federal law and guidelines, industry has also issued policies to which they voluntarily
adhere. The “Do not Track” initiative for example, was created by companies including Google,
Yahoo, Microsoft and AOL. It aims to provide consumers with opt-outs for behaviour-based
marketing.
Most cases the FTC handles are by virtue of Section 5 of the FTC Act. However, the FTC also relies on
more detailed guidelines (currently under review) to investigate cases. The FTC also issues significant
guidance and FAQs on a range of consumer protection related issues.
602
Federal Trade Commission, 2013, Mobile Privacy Disclosures: Building Trust Through Transparency: A Federal Trade Commission Staff Report.
603
See: Federal Trade Commission, 2020, Consumer Sentinel Network Data Book: Annual data book 2020.
604
See: Federal Trade Commission, Consumer Advice: Imposter Scams.
605
See: Federal Trade Commission, 2020, FTC Data Shows Record Surge in Online Shopping Complaints During Pandemic, Press Release.
606
Federal Trade Commission, Children's Online Privacy Protection Rule ("COPPA"), Children's Online Privacy Protection Act of 1998, 15 U.S.C.
6501–6505.
607
Federal Trade Commission, 1998, Privacy Online: a report to Congress.
608
Federal Trade Commission, 2009, Self-Regulation Principles for Online Behavioral Advertising.
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.913 150
An important part of the FTC’s activity relating to digital advertising revolves around privacy protection.
Unlike the EU, the US does not have a separate data protection authority or a single comprehensive
privacy law at the federal level like the GDPR. The FTC uses Section 5 complemented by privacy-related
statutes to protect consumer privacy, including the Children’s Online Privacy Protection statute
609
.
During the pandemic, children’s time spent online has drastically increased with online learning which
led to a range of new issues such as privacy issues in the educational technology area. Issues the FTC
deals with also include Artificial Intelligence
610
and dark commercial patterns
611
. To kick off awareness
around the issue, the FTC held a workshop in April 2021 to bring together consumer advocates industry
professionals and will publish a report
612
. The FTC is able to use Section 6B of FTC Act, to issue orders
which allow them to collect information from companies
613
.
2. Design and implementation of response
There is no set time period for the FTC guidelines review, contrary to rule making procedures (e.g. the
negative option rule making), for which some very specified procedures are set out. The procedure
typically takes a few years. The current revision of the guidelines seeks to update them to take
account of new marketing techniques. Through its responses to specific cases, the FTC provides
responses to issues which signal to the market how they interpret the law from the FTC Act and Section
5 in particular, which can sometimes become laws. The FTC can conduct investigations upon a
complaint by the public or by a corporation and formal complaints are addressed by an administrative
judge who may issue a cease-and-desist order or issue relief
614
. Complaints may also be brought before
the largest Industry body, BBB National Programmes, which has a National Advertising Division
615
.
Bringing enforcement cases is a privileged way for the FTC to tackle new consumer issues that arise
and allow the law to keep up with the most recent market practices.
The FTC also seeks to influence market practices in the area of privacy and digital advertising. In
December 2020, the FTC issued orders to 9 social media and video streaming companies under Section
6B of FTC Act, which allows them to collect information from companies
616
. The orders require the
companies to provide data on how they collect, use, and present personal information, their
advertising and user engagement practices, and how their practices affect children and teens. The FTC
is also using this Section 6B tool in an ongoing study examining the privacy practices of broadband
providers
617
.
609
See: Federal Trade Commission, Children's Online Privacy Protection Rule ("COPPA"), Children's Online Privacy Protection Act of 1998, 15
U.S.C. 6501–6505.
610
See: Smith, A., 2020, Using Artificial Intelligence and Algorithms, blog.
611
See: Federal Trade Commission, 29 April 2021, Bringing Dark Patterns to Light: An FTC Workshop, 29 April 2021.
612
Ibid.
613
See: Federal Trade Commission, 2020, FTC Issues Orders to Nine Social Media and Video Streaming Services Seeking Data About How They
Collect, Use, and Present Information, Press Release.
614
See: Federal Trade Commission, Truth in Advertising, Media Resources.
615
See: Bbb National Programs, National Advertising Division, Information Page.
616
See: Federal Trade Commission, 2020, FTC Issues Orders to Nine Social Media and Video Streaming Services Seeking Data About How They
Collect, Use, and Present Information, Press Release.
617
See: Federal Trade Commission, 2019, FTC Seeks to Examine the Privacy Practices of Broadband Providers, Press Release.
Online advertising: the impact of targeted advertising on advertisers, market access and consumer choice
151 PE 662.913
3. Analysis and Evaluation
The significant number of cases the FTC deals with, their variety, as well as the compensation it has
issued to customers indicate the FTC does meet its objectives in terms of consumer protection and
privacy. Last year, the FTC provided USD 483 million
618
to consumers in the US and other countries
619
.
It also provided a number of reports to the Congress on digital advertising issues including one about
social media bots and deceptive advertising
620
.
Two cases which reflect the FTC’s activity in these areas are the Flo Health
621
and the Everalbum
622
case.
The first which involved allegations that the developer of a period and fertility-tracking app shared
users’ health information with third parties such as Facebook, Google, and others after promising to
keep the information private. The settlement requires Flo Health, Inc. to, among other things, obtain
an independent review of its privacy practices and get app users’ consent before sharing their health
information. In the latter, the FTC sought action against Everalbum which involved allegations about
an online company’s use of facial recognition technology and its retention of the photos and videos of
users who deactivated their accounts. The proposed order includes a novel algorithmic remedy,
requiring the company to delete models and algorithms it developed by using the photos and videos
uploaded by its users.
618
This corresponds roughly to EUR 393 million [exchange rate 31.12.2020].
619
See: Federal Trade Commission, 2020, Data on Refunds to Consumers, Cases and Proceedings.
620
Federal Trade Commission, 2020, FTC Sends Report to Congress on Social Media Bots and Deceptive Advertising, Press Releases.
621
Federal Trade Commission, 2021, Developer of Popular Women’s Fertility-Tracking App Settles FTC Allegations that It Misled Consumers
About the Disclosure of their Health Data, Press Release.
622
Federal Trade Commission, 2021, California Company Settles FTC Allegations It Deceived Consumers about use of Facial Recognition in
Photo Storage App, Press Release.
PE 662.913
IP/A/IMCO/2020-30
Print ISBN 978-92-846-8130-3 | doi: 10.2861/288740| QA-02-21-683-EN-C
PDF ISBN 978-92-846-8131-0 | doi: 10.2861/80 | QA-02-21-683-EN-N
In this research paper, we provide a comprehensive overview of online advertising markets and we
analyse the challenges and opportunities concerning digital advertising. We review the degree to
which existing and proposed legislation at EU level addresses the identified problems, and identify
potential solutions, with reference to experience from EU Member States and third countries. We
conclude with a synthesis and specific policy recommendations, drawing on stakeholder interviews.
This document was provided by the Policy Department for Economic, Scientific and Quality of Life
Policies for the committee on the Internal Market and Consumer Protection (IMCO).