(by Elias Deutscher) Can the owner of a well-known brand lawfully prevent its independent retailers from using its trademarks and brands to advertise their products on Google? This question has major implications, in particular, for small- and medium-size retailers who largely benefit from the reduced costs of online search advertising and the new sales channels offered by online distribution. For quite some time, issues surrounding the use of trademarks as keywords in paid online search advertising have stirred considerable controversy amongst trademark and IP lawyers. Recently, this question also came into the focus of competition law enforcers. In the Case AT. 40428 Guess, the European Commission assessed for the first time the legality under European Union (‘EU’) competition law of a vertical agreement whereby a trademark proprietor restricted the ability of its licensed distributors to use or bid for its brand names and trademarks as keywords in Google AdWords.
The facts of the case and the Commission’s analysis
Guess is an American company that produces and sells fashionable apparel and accessories under various trademarks, such as ‘GUESS?’, ‘GUESS Jeans’ or ‘GUESS by Marciano’. To distribute its products, Guess uses a dual distribution model. On the one hand, it sells its products through its own physical stores in various Member States of the EU and operates its own online shop http://www.guess.eu. On the other hand, Guess has set up a selective distribution network with independent wholesalers, as well as mono-brand and multi-brand retailers. To ensure the exclusive and prestigious image of its brands, Guess selects its authorised wholesalers and retailers based on several qualitative criteria pertaining, for instance, to the store design and quality standards for pre-sale services or advertising (paras. 6, 20-31).
Guess’ selective distribution agreements contained several vertical restraints that raised competition law concerns under Article101 TFEU, which prohibits anticompetitive agreements. The most novel amongst these restraints concerned the restriction on online search advertising whereby Guess prohibited its authorised retailers from using or bidding for its trademarks and brand names as keywords for paid search in Google AdWords. By imposing these restrictions, Guess sought to prevent its authorised retailers from cannibalising its online sales, by diverting consumers away from its exclusive online shop guess.eu to their own websites. The restriction on online search advertising secured Guess’ own online sales channel the most prominent position on the Google search platform. At the same time, Guess made sure that the online shops of its retailers could only be found further down amongst the generic search results on Google’s website. By hindering its dealers from using and bidding for its trademarks and brand names in Google AdWords, Guess also tried to prevent them from driving up its search advertising costs. Google AdWords selects advertisers displayed in Google’s sponsored search results on the basis of an auction process that determines the visibility and cost-per-click rate of each advertiser. As a result, Guess’ advertising cost increased, the more authorised retailers were bidding for the same keywords (paras. 40-52).
The Commission’s assessment of the online search restraints turned on two principal considerations. First, the Commission observed that neither the origin nor the advertising function of Guess’ trademarks could justify the restriction of online search advertising it imposed on its retailers. The Commission, therefore, asserted that the restraints could not be explained by any legitimate objective of Guess’ selective distribution agreement in accordance with the Metro doctrine. ( paras. 115-123). Second, the Commission took the view that the restrictions reduced the ‘findability’ of retailers’ online presence and seriously hampered their ability to sell products outside their allocated sales territory. On this account, the Commission concluded that online search advertising restrictions constituted a clear-cut by-object restriction of competition, in breach of Art. 101 (1) TFEU (paras. 114, 120-125). The Commission further held that the agreement could not benefit from the block exemption provided under the Vertical Agreement Block Exemption Regulation (VABER) ( paras. 152-162). Nor would online search advertising restrictions qualify for an individual exemption under Art. 101 (3) TFEU, as they were not indispensable to address free-riding, or to protect Guess’ brand image (para. 164).
An alternative story of vertical online search advertising restrictions
Although the Commission was right in holding that vertical online search advertising restraints raise competition issues under Art. 101 TFEU, condemning them as by-object restrictions may overshoot the mark. The Commission’s strict approach in Guess turns on the assumption that there are no legitimate, pro-competitive reasons for a brand owner to restrict online search advertising by its retailers. I challenge this view, arguing that it disregards the role of vertical restraints in allowing manufacturer-brand owners to coordinate the complex relationship between online search, online display and traditional offline advertising within their distribution systems. Vertical online advertising restraints may help brand owners ensure a sufficient level of investment in display and traditional advertising that play a crucial role for competition with other brands (so-called ‘inter-brand competition’).
A major shortcoming of the Commission’s analysis in Guess is that it focused exclusively on the impact of these restraints on online search advertising. This narrow focus disregards that manufacturer-brand owners, like Guess, must carefully coordinate with the members of their distribution networks how they attribute advertising expenditure across various advertising channels: in particular, online search, online display (e.g., banners, images or videos) and traditional offline advertising (e.g., TV, radio and print media).
This is a delicate exercise because each form of advertising performs a different role in the so-called ‘conversion funnel’ that traces the different stages of consumers’ journey from becoming aware of a brand up to the point where they make a final purchasing decision. Online search advertising typically influences consumers only in the advanced stages of their decision-making process. As it interacts with consumers closer to their actual purchasing decision and routes them directly to the website of the advertising seller, the effectiveness of search advertising in generating sales outperforms that of display and offline advertising. In short, each pound a manufacturer or its retailers invest in search advertising will translate into more sales than the investment of the same amount of money in display or traditional advertising.
By contrast, online display and traditional offline advertising interact with consumers in the early stages of their purchasing decision-making process. While prompting only to a limited extent direct purchasing decisions, display and traditional advertising are particularly effective in raising awareness for a brand, attracting new audiences and differentiating the brand from competitors. Recent studies also show that display advertising and, albeit to a lesser extent, traditional advertising may have significant positive spill-over effects on search advertising because they create consumer awareness and, thereby, indirectly elicit clicks and purchasing decisions. Without display advertising, consumers might simply not enter the brand name as a search term in Google’s search bar in the first place.
This positive spill-over effect on sales is, however, difficult to attribute to display advertising and traditional advertising. Investments in search advertising can also be more easily appropriated than investment in display and traditional advertising because they generate high click-through and conversion rates, while producing fewer positive externalities that benefit other members of the distribution network. Retailers, therefore, tend to consider search advertising as the least risky alternative: the return on investment is higher and less uncertain than for other forms of advertising. This creates a situation where retailers are likely to over-invest in search advertising, which intensifies so-called ‘intra-brand competition’ between the sellers of the same branded product. At the same time, they tend to under-invest in display advertising and traditional advertising, which are key to compete with other brands (inter-brand competition).
To address this problem of over-investment in search and under-investment in display and traditional advertising, manufacturer-brand owners, such as Guess, may end up centralising the provision of display and traditional advertising under their exclusive control. Brand owners’ task of providing the appropriate level of display and traditional advertising is, however, further complicated if retailers through their over-investment in search advertising and competitive bidding on keywords drive up their search advertising cost. Excessive intra-brand competition by retailers through search advertising thus may significantly reduce the amount of money the brand owner can spend on the riskier forms of display and offline advertising. The size of this adverse effect of the competitive use of keywords by retailers on the brand owner’s advertising cost depends largely on the number of authorised retailers bidding for the same trademark as a keyword in Google AdWords. This adverse effect is further amplified by the fact that online search advertising, due to its high degree of effectiveness in generating sales, allows retailers to take a free-ride on the brand owner’s advertising expenses for display and traditional advertising. This free-rider effect may seriously undermine the ability of the brand owner to recoup part of its investment in online display and offline advertising through the sales via its own online shop. Both effects, the increased search advertising costs and the free-rider effect, may, in the end, chill the brand owner’s incentive to invest in non-search advertising that performs a crucial role in expanding consumer awareness and stimulating inter-brand competition.
The imposition of online search advertising restrictions on retailers constitutes one way for the brand owner to get to grips with this problem. These restrictions offer a handy tool for the brand owner to recover the (fixed) cost it incurs by investing in non-search advertising. At the same time, they enable the brand owner to internalise the free-riding externality generated by the excessive investment in search advertising by its retailers. By limiting the retailers’ use of its trademarks as keywords and leaving them just with the opportunity to bid on a long tail of less effective search terms with lower click-through-rates, these restrictions increase search costs and search friction for consumers. This, in turn, allows the manufacturer to maintain price dispersion across its distribution system, if consumers have different valuations for its products and different search costs. Consumers with a high valuation, say, for Guess products and high search cost will have little incentive to try several search queries on Google or visit more than one website. Owing to the online search restraints, these high-value/high-search cost consumers are likely to be channelled to Guess’ online store where they can be charged a high ‘premium’ price. Consumers with lower valuation and lower search cost, by contrast, will invest more time and efforts in searching for stores selling Guess products at lower prices. They may, for example, enter a search term in Google which does not contain Guess’ brand names or trademarks, routing them to an authorised retailer who charges a lower price.
By segregating and isolating high-value, search-cost sensitive consumers from low-value, search-cost insensitive consumers, online search advertising restrictions allow the manufacturer to orchestrate brand-wide price discrimination taking place at the level of its distribution network level rather than the level of individual retailers. This form of price discrimination enables a brand owner, such as Guess, who operates its own online store, to extract a maximum of surplus and thereby recoup parts of its investment in display and offline advertising.
Without a doubt, brand owners, like Guess, may have alternative solutions at their disposal to recover part of the costs they incur by investing in display and traditional advertising. Guess could, for instance, simply decide to charge a higher wholesale price. Brand-wide price discrimination through search advertising restrictions may, however, be preferable to such a uniform increase of wholesale prices of Guess products because it allows for surplus extraction without quantity restriction. Low-value consumers who have a lower willingness or capacity to pay for Guess products are not priced out of the market but are still served by cheaper retailers they can find after incurring some additional search cost. While reducing intra-brand competition, online search advertising restraints may thus stimulate inter-brand competition and preserve the short- and long-term incentives of brand owners to invest in brand recognition and the creation of new brands.
This is, however, not to say that restrictions on online search advertising are unambiguously innocuous. All to the contrary. As they increase information and search costs, such restraints may considerably harm consumer welfare. Restrictions on online search advertising introduce significant search frictions and reduce the ability of consumers to carry out informed intra-brand price comparisons between different online shops selling Guess products. Since Guess’ online search advertising restrictions also constrained how authorised multi-brand dealers advertised its products on Google, the ability of consumers to carry out meaningful price and quality comparisons between brands is also likely to suffer. The adverse impact of online search advertising bans on consumers’ ability to engage in intra- and inter-brand quality and price comparisons is particularly worrisome in situations where product differentiation already makes such side-by-side comparison very difficult. Online search advertising restraints also deprive retailers of new opportunities to reach consumers that the emergence of online distribution models has opened up to them.
Online search advertising restrictions thus involve a complex trade-off: while they may seriously undermine intra-brand competition, they have the potential to stimulate inter-brand competition. Their overall competitive impact on competition may, therefore, be less clear-cut than their characterisation as by-object restrictions in Guess suggests. The very fact that they generate potentially ambiguous welfare effects counsels against the categorical prohibition of online search advertising restrictions under Art. 101 TFEU adopted by the Commission in Guess.
In light of the ongoing reform of the Vertical Agreements Block Exemption Regulation and the Guidelines on Vertical Agreements by the European Commission, it is important to note that such a restrictive policy against online search advertising restrictions may entail two unintended consequences. First, while benefitting consumers in the short-run by increasing consumer choice and lowering prices, consumers may be worse-off in the long-run if the costs this policy imposes on trademark owners undermine their dynamic incentives to invest in new, innovative and exciting brands. The second unintended outcome of this policy is that it ultimately benefits large online platforms, such as Google. By preventing retailers from bidding on Google AdWords, Guess also tried to keep its advertising costs and, hence, the slice Google takes of its distribution revenues to a minimum. The use of Art. 101 TFEU to protect retailers’ ability to harness the new opportunities of online search advertising and thus to intensify intra-brand competition does not only help small and medium-size online dealers. Above all, more competition on AdWords means more advertising revenue for Google. The Commission’s decision thus shows that antitrust intervention aimed at helping the ‘little guy’ (here small and medium-size retailers) may also have the unintended consequence of supporting the ‘big guy’ (here Google). Greater attention should be paid to both of these unintended consequences in the ongoing policy debate about the reform of the competition law approach towards vertical online restraints.
 Initially published on the UEA ISP blog
 Case COMP/AT.40428 Guess. C(2018) 8455 final.
 Case 26/76 Metro v Commission ECLI:EU:C:1977:167 para. 20.
 Commission Regulation (EU) No 330/2010 of 20 April 2010 on the application of Article 101(3) of the Treaty on the Functioning of the European Union to categories of vertical agreements and concerted practices OJ L 102, 23.4.2010, p. 1-7.
 Competition and Markets Authority, ‘Online platforms and digital advertising: Market study interim report’ (2019) 153–157 <https://assets.publishing.service.gov.uk/media/5dfa0580ed915d0933009761/Interim_report.pdf> accessed 30 March 2020.
 E. Bayer and others, ‘The impact of online display advertising and paid search advertising relative to offline advertising on firm performance and firm value’  International Journal of Research in Marketing, 2–3; I. M. Dinner, H. J. van Heerde and S. A. Neslin, ‘Driving Online and Offline Sales: The Cross-Channel Effects of Traditional, Online Display, and Paid Search Advertising’ (2013) 51(5) Journal of Marketing Research 527 542; Competition and Markets Authority, ‘Online platforms and digital advertising: Market study interim report’ (n 4) 49, 157.
 Dinner, van Heerde and Neslin (n 6), 530, 539-540; P. Kireyev, K. Pauwels and S. Gupta, ‘Do display ads influence search?: Attribution and dynamics in online advertising’ (2016) 33(3) International Journal of Research in Marketing 475 475-476, 487-489.
 In the case of Guess, this number was particularly high. In the European Economic Area (EEA), Guess’ products are distributed by about 113 authorised mono-brand stores operated by 30 companies and by up to 3000-5500 independent multi-brand dealers Case COMP/AT.40428 Guess (n 2) para. 21.
 The following argument draws upon the analysis of price-related information restraints by J. Asker and H. Bar-Isaac, Vertical Information Restraints: Pro- and Anti-Competitive Impacts of Minimum Advertised Price Restrictions (National Bureau of Economic Research 2016) 2, 11-17; J. Asker and H. Bar-Isaac, ‘Advertising and Related Restraints’  Competition Policy International Antitrust Chronicle 1.
 Asker and Bar-Isaac (n 9), 4.
 See for instance R. L. Steiner, ‘The Nature of Vertical Restraints’ (1985) 30 Antitrust Bulletin 143 146, 183, 190.R. L. Steiner, ‘Intraband Competition—Stepchild of Antitrust’ (1991) 36(1) The Antitrust Bulletin 155 193–195.