Key Points In Coty, the CJEU relaxed the law against vertical restraints on distribution of luxury goods over the internet. This is a welcome step towards a sensible law on these restraints. Simple economics shows that in the absence of vertical restraints, retailers are likely to underinvest in maintaining a prestigious image for a luxury product or indeed any product for which demand is sensitive to brand image. Reselling over the internet is an extreme case of this underinvestment. A manufacturer’s choice to protect its image via vertical restraints at the cost of forgoing lower retail prices and the broad reach of the internet is often in the interest of consumers. Consumers clearly value prestigious product image. The decision narrows the application of the paternalistic reasoning in Pierre Fabre that a prestigious image is not worthy of protection via vertical restraints. I. Introduction In its decision of 6 December 2017 in Coty, the Court of Justice of the EU (CJEU) relaxed the law against restraints placed by manufacturers on retailers’ decisions to distribute goods over the internet. In a previous case, Pierre Fabre (C-439/09), the Court had stated that, as regards cosmetics, ‘the aim of maintaining a prestigious image is not a legitimate aim for restricting competition.' In Coty, the court narrowed the possible interpretations of Pierre Fabre, stating that this ruling concerned only the total ban of internet sales of cosmetic products at issue in that case. The Court decided in Coty that the preservation of a luxury image in fact can justify restrictions on internet sales (via selective distribution agreements) providing certain conditions apply.1 In particular, a manufacturer can now restrict selling on third-party platforms such as Amazon or E-Bay if these conditions apply. Coty is an important case in the development of EU law on selective distribution agreements and indeed in the entire area of vertical restraints. The case is interesting in light of the gap between European and US competition law on vertical restraints. This gap has grown as the US law on vertical restraints has become much more responsive to the underlying economic logic of the contractual agreements while EU law, until Coty, had not.2Pierre Fabre and Coty are cases that would never have been brought in a jurisdiction, like the USA, in which a vertical restraint is evaluated according to its impact on market incentives and ultimately on consumers. The Court of Justice of the EU (CJEU) in adopting an effects-based analysis in Coty has moved EU competition law in the right direction from an economist’s perspective. The Court concluded with justification that there are legitimate reasons for a supplier of luxury goods to restrict the distribution of its products over the internet. The Coty decision presents us with an opportune time to review the law and economics of vertical restraints on internet distribution. I offer in this note a brief perspective on the economics of vertical restraints in general and an application of the economic perspective to the central issue in Pierre Fabre and Coty: the use of restraint to protect product image.3 II. Vertical restraints Contracts between a manufacturer and distributor are much more complicated in reality than in an undergraduate textbook. In the textbook model, a supplier sets a price and anybody who wants to buy at the established price can do so. The purchaser acquires all property rights associated with the product and is unrestrained in any resale of the product. In reality, manufacturers and retailers often agree to contracts that impose restraints on the price at which the retailer may sell the product, on the consumers to whom the retailer may sell, on the channels for reselling, and so on. In Pierre Fabre and Coty, the issue was the extent to which manufacturers can restrict resale of their products over the internet. Before we can assess the Coty decision, we need a theory of why a manufacturer would impose restraints in its contracts with retailers. Why not leave retailers free to set prices as low as they want, free to sell to any consumers, and free to resell through any channels? Allowing retailers freedom to compete results in the lowest retail price. Once a manufacturer has set its wholesale price, w, then its wholesale margin is determined. The profits from sale of any additional quantity are equal to the product of this quantity and the wholesale margin. Anything that increases quantity therefore increases profit (at a given wholesale price). Surely lower retail prices would seem to increase quantity. And allowing retailers to use the internet would seem to allow more sales because the internet is, as the European Vertical Restraints Guidelines state, ‘a powerful tool to reach more and different customers.'4 Vertical restraints against sales over the internet are, at first blush, a puzzle. One set of possible answers that we must always keep in mind in explaining vertical restraints is that restraints can facilitate collusion among suppliers or exclusion by a dominant supplier, dampen price competition among upstream suppliers, or serve as instruments for collusion or exclusion at the downstream retail level.5 But in most cases, including Pierre Fabre and Coty, evidence that restraints are used for purposes of anticompetitive exclusion or collusion is not produced. Manufacturers imposing vertical restraints do so to increase profit without strategic regard to facilitating collusion or exclusion. Vertical restraints must encourage behaviour on the part of retailers that raises somehow quantity or adds value to the product – in a way that more than offsets the negative impact of higher retail price on quantity. This point is so important that it is worth stating more precisely, with a little algebra. Suppose that a manufacturer faces a unit cost, c, of a product and in its sales strategy chooses two dimensions: a wholesale price, w, and a set of restraints, r ϵ S. The set of restraints, S, includes anything in the contract apart from the wholesale price. It may refer to a constraint that the retailers not sell below a specified price floor, or that retailers not sell outside a given territory, or over the internet, or through third parties, and so on. Any set of possible contractual restrictions is included in the set of possible restraints, S. Given the wholesale price, w, and the set of restraints, r, retailers will respond with actions such as retail price, sales effort, sales strategies in general that determine the total quantity, Q, of sales of the manufacturer’s product. We can therefore express the dependence of quantity on wholesale price and the restraints as a function, Q(w, r).6 The total profits of the manufacturer, also a function of (w, r), are given by π(w,r)=(w–c)·Q(w,r) (1) From equation (1), the choice of restraint, r, affects profit only through its impact on volume, Q. It follows as a simple matter of logic that the vertical restraint, r, that maximises profits also maximises quantity, given the manufacturer’s choice of w. Maximising π(w, r) with respect to r in (1) is equivalent to maximising Q(w, r) with respect to r, since the right hand side of (1) is separable, with r appearing only in Q(w, r). Suppose for example that the cost c is €5, and that the manufacturer’s choice of optimal wholesale price w is €8. Then the manufacturer makes €3 per unit, however many units are sold. The more units sold, the greater the profit; therefore to maximise profits, the manufacturer will choose a set of restraints r that maximises output, Q, given the chosen w. As a caution, we should not conclude from this that any vertical restraint chosen by a manufacturer maximises output and therefore than any limitations on the choice of r reduces output. The volume-maximising principle is qualified with the phrase ‘at the manufacturer’s choice of w'. The variable w varies endogenously when the law changes the set of feasible restraints, S, so one cannot conclude that a laissez-faire policy maximises output.7 But the manufacturer has the option of leaving w unchanged and eliminating all restraints entirely. (We could represent choice of no restraints as r = 0.) The fact that the manufacturer chooses a contract with restraints rather than option of r = 0 means that the restraints must have a role in maximising quantity. It must be true that at the chosen w, allowing unrestrained retailer decisions and competition would somehow reduce quantity. The higher prices and more restricted set of consumers with the vertical restraint are more than offset, in terms of impact on quantity, by strategies on the part of retailers that the restraints elicit. The simplest example is resale price maintenance. Suppose that a manufacturer sets a wholesale price of €5 and that in the absence of any restraint the retail price would be €8. The retail margin is €3. If the manufacturer sets a retail price floor of €11, then it ensures a doubling of the retail margin to €6. This doubling of the retail margin will surely elicit greater sales effort on the part of retailers – sales effort defined broadly to include effective and enthusiastic sales staff, a comfortable and attractive shopping environment, a prominent display of products, and provision of information about the product, and local advertising. The increased retail margin will also increase the number of retail outlets choosing to carry the product. All of these effects will increase the sales of the product. And it follows from the mere observation of the restraint, if collusive or exclusionary theories can be set aside, that the quantity impact of the additional sales effort and outlets more than offsets the negative impact of the higher retail price (€11 instead of €8). This general logic applies to any vertical restraints, not just resale price maintenance. I apply the perspective to restrictions on internet resales below. First, however, it is important to discuss at a general level the policy implications of this positive theory of why we observe vertical restraints. Does the theory imply that the manufacturer chooses restraints if and only if the restraints are in consumers’ interest? That is, are we guaranteed that the market ‘works' without legal restrictions? The answer might seem to be yes, given that the manufacturer chooses restraints to maximise quantity conditional upon the wholesale price. But it is in fact easy to construct some examples for which consumers’ welfare (or total welfare, including shareholders’ profits) falls with vertical restraints and other examples for which welfare rises. A market unrestrained by competition law does not guarantee the ideal market outcome even in the absence of collusion. If one imposed in the law a burden of proof on suppliers to justify their restraints on competition among retailers, then many uses of restraints would end up being prohibited because suppliers could not, through economic reasoning, justify their contracts with retailers as maximising consumers’ interest. But asking whether the market achieves the ideal outcome in the absence of intervention by the state is posing entirely the wrong question. We intervene in markets when there is (at a minimum) a high likelihood of improving the market outcome – not because such intervention might improve the market outcome. It is always true that an unrestricted firm might be wrong in choosing between (for example) greater sales effort or promotion with restraints versus lower prices without restraints. In any intervention, by competition law or regulation, the burden of proof should lie on the side of the intervention. We do not intervene in a manufacturer’s choice of prices versus promotion when this choice is made directly; nor should we intervene when this choice is implemented through vertical restraints.8 The policy suggested by this approach is one in which either the burden falls on the plaintiff in a vertical restraints case or a structured rule of reason approach in which the burden shifts between the plaintiff and defendant depending on whether preliminary evidence indicates the possibility that the restraint is motivated by anticompetitive collusion or exclusion.9 An aspect of the economic theory outlined to this point is important but may not be obvious. Public policy towards vertical restraints inherently involves a double negative. A rule against a vertical restraint is a prohibition of a prohibition: the prohibition of a manufacturer’s right to prohibit certain actions of retailers. To express this in terms of internet-selling restrictions, the question is not whether consumers should have the right to purchase over the internet; the question is whether manufacturers should have the right to prohibit the sales by retailers over the internet. To assess a law or policy on vertical restraint we therefore need a hierarchical model a theory of how final consumers respond to various decisions on the part of retailers; a theory of how retailers will respond to various vertical restraints, given the theory (1) of how consumers respond to retailer decisions; a theory of how the manufacturer, free of legal restraints, would choose its optimal contract offer to retailers, given the answers to (1) and (2); and finally a theory of how a policy maker should choose design an optimal set of laws given the answers to (1)–(3). This framework for establishing an economic theory of optimal policy in this area is completely natural to an industrial organisation or competition policy economist, given the game-theoretic foundation of these areas of economics. But it is, I think, subtle. The European Commission misses the point entirely in its Vertical Restraints Guidelines. According to the Guidelines, the Commission regards as a hardcore restriction10 any obligations that dissuade dealers from using the internet to reach a wider range of customers by imposing criteria for online sales that are not approximately equivalent to the criteria imposed on sales from brick and mortar shops.11 The Guidelines offer the following rationale for this rule: The internet is a powerful tool to reach more and different customers than will be reached when only more traditional sales methods are used and this is why certain restrictions on the use of the internet are dealt with as (re) sales restrictions. In principle, every distributor must be allowed to use the internet to sell products.12 If the power of the internet to reach more and different customers than can be reached through more traditional sales methods were the only factor involved in the use of the internet then we would never observe restrictions on internet sales. The manufacturer certainly has an interest in reaching more and different customers. The Guideline’s justification fails to provide a theory of why a manufacturer would restrict sales over the internet. The mere observation of the restraint tells us that something else is going on. The theory of vertical restraints that I outlined earlier in this note is a ‘reduced-form' or summary theory. It explains that the observation of a vertical restraint, if conditions for anticompetitive collusion or exclusion are ruled out, in and of itself tells us that the restraint has a role in increasing consumer value or quantity demanded.13 But in any particular case, we must have a more detailed or ‘structural' theory of (1) why the simplest contract without the restraint fails to elicit retailer incentives compatible with the manufacturer’s quantity-maximisation problem (at the chosen wholesale price); and (2) how the restraint solves this. I discuss in the next section such a theory, and its policy implications, for the central issue in Pierre Fabre and Coty. III. Restrictions on internet sales: Pierre Fabre and Coty The brief background I have offered on the economics of vertical restraints can be applied immediately to the issue at the heart of Pierre Fabre and Coty: whether the manufacturer of luxury or image goods should be allowed to prohibit or restrict distributor sales over the internet. The economics of the two cases are very similar. The general theory offered by both Pierre Fabre and Coty is persuasive: the restrictions help the manufacturer maintain its brand name value or and image.14 The Court in Pierre Fabre did not necessarily reject the brand-building logic, but instead rejected the idea that brand image was worthy of protection. The Court in Coty narrowed the scope of the Pierre Fabre decision, allowing restrictions against resale on third-party platforms, explaining that the Pierre Fabre decision applied narrowly to the total ban on resale over the internet to the products at issue in Pierre Fabre. The facts of Pierre Fabre illustrate, contrary to the CJEU decision in the case, that restrictions on internet sales can increase the economic effectiveness of the vertical distribution system as a whole. A simple wholesale contract fails to elicit efficient retailer incentive because of a free riding or ‘positive externality' story. Downstream retailers of Pierre Fabre products are concerned not about the overall sales of Pierre Fabre products, but about their own sales. Pierre Fabre, in contrast, is concerned about the profitability of its distribution system as a whole. Each retailer’s investment in brand image, which may involve simply the decision to display the product rather than sell it over the internet, benefits not only that retailer but other retailers as well. As in any decision involving positive externalities, the individual retailer’s investment is too low for the profits of the system as a whole. In competition among any cosmetic product lines, it is clear that brand image is a significant selling point. Pierre Fabre would therefore place considerable importance on promoting its brand image to consumers. Retailers are concerned much less about contributing to Pierre Fabre’s brand through their own investment, since they capture such a small share of the benefits of this investment. In the context of brand-building services, for either Pierre Fabre’s cosmetic products or Coty’s luxury goods, shoppers will respond to the overall investment by retailers in enhancing product image even if they never visit the store. Because of the acute positive externality from retailer services associated with brand building, the divergence between downstream and upstream incentives to provide such services is especially strong.15 In both Pierre Fabre and Coty, the supplier decisions to restrict internet sales by its retailers are a clear attempt to protect the brand image of the products. Brand image is valued by consumers. Consumers pay thousands of dollars for a luxury purse, for example, not because of the quality of the zippers or the leather in the product but because of the prestige and image of the product. If consumers attach such value to image, then policy or law should respect manufacturers’ attempts to arrange their distribution systems so as to protect image. Any intervention in the market to promote lower prices over product image represents a paternalistic decision that consumers’ preferences are not a valid representation of their interests. But wealthy consumers, purchasing luxury products, do not need the intervention of courts to direct markets to provide lower prices at the expense of reduced investment in brand image. Pierre Fabre was a paternalistic decision by the Court of Justice of the EU, and the narrowing of the decision in Coty is a welcome step in the right direction. I conclude with a remark about how broadly the economic reasoning that I have outlined can apply. Pierre Fabre and Coty were about luxury goods. But the theory applies equally well to any good for which brand image is important. Running shoes or basketball shoes, for example, may command prices well over €100 not because of the quality of the laces, stitching or construction but because consumers value brand names such as Air Jordan, Nike or New Balance.16 The normative principle of consumer sovereignty demands that public policy respect consumers’ choices – and that public policy respect manufacturers’ efforts to protect brand image so highly valued by consumers. Footnotes 1 The conditions are set out on page 4 of the decision: ‘… Article 101(1) TFEU must be interpreted as meaning that a selective distribution system for luxury goods designed, primarily, to preserve the luxury image of those goods complies with that provision to the extent that resellers are chosen on the basis of objective criteria of a qualitative nature that are laid down uniformly for all potential resellers and applied in a nondiscriminatory fashion and that the criteria laid down do not go beyond what is necessary.' 2 The prominent US cases in this trend are Continental Television v. GTE Sylvania, 433 U.S. 36 (1977); State Oil Co. v. Khan, 522 U.S. 3 (1997); and Leegin Creative Leather Products, Inc. v. PSKS, Inc., 551 U.S. 877 (2007). 3 This note draws on my article with Edward Iacobucci, ‘European Law on Selective Distribution and Internet Sales: An Economic Perspective,' Antitrust Law Journal 81 (2016). 4 Commission Notice – Guidelines on Vertical Restraints SEC (2010) 411 fianl (May 10, 2010) (hereafter, Vertical Restraints Guidelines) at para 23. 5 Some of the more prominent articles exploring the anticompetitive roles of vertical restraints are Telser, Lester, ‘Why Would Manufacturers Want Fair Trade?,' Journal of Law and Economics 1960; Bruno Jullien, and Patrick Rey, ‘Resale Price Maintenance and Collusion', The RAND Journal of Economics, vol. 38, n. 4, Winter 2007, pp. 983–1001; Raising Retailers’ Profits: On Vertical Practices and the Exclusion of Rivals.' American Economic Review · vol. 104, no. 2, February 2014. (pp. 672–686); Daniel P. O’Brien and Greg Shaffer, ‘Vertical Control with Bilateral Contracts,' The RAND Journal of Economics, Vol. 23, No. 3 (Autumn, 1992), pp. 299–308; Dobson, P.W and M. Waterson, 1997, Countervailing power and consumer prices, Economic Journal 107, 418–430. 6 Q(w, r) is in economic terminology a ‘reduced-form' relationship in that it summarises the structure whereby a given choice of w and r elicits retailer choices, including the retail price, that determine the quantity, Q. 7 For example, under a facilitating-practices theory of vertical restraints, a collusive level for the wholesale price, w, is supported by the selected vertical restraint. 8 The Standing Committee on Industry, Science and Technology (SCIST) of the Canadian Parliament stated, ‘The classical example of [vertical] price maintenance is where a supplier requires someone to whom it sells … to maintain prices at a particular level as a way of encouraging that retailer or wholesaler to engage in competition on something other than price. A higher retail margin thus encouraged the retailer to engage in providing a high level of service to clients or to ensure that the brand image associated with the product is maintained and not sullied in any way. (¶) From the consumer’s perspective, vertical price maintenance results in more services, which we would regard as good, but higher prices, which we would view as bad… Prohibiting resale price maintenance under the per se rule is effectively regulating the manufacturer’s decisions on how best to maximise the sale of his products. By way of an analogy, we do not prohibit by law high levels of advertising even when such advertising raises prices; for the same reason we should not prohibit vertical price maintenance under a per se rule.' (SCIST) Report: A Plan to Modernize Canada’s Competition Regime, (2002), chapter 5. 9 Christine A. Varney suggests a burden-shifting approach to resale price maintenance, in A Post-Leegin Approach to Resale Price Maintenance Using a Structure Rule of Reason, Antitrust, Fall 2009 at 22. Varney was at the time this article was published Assistant Attorney General in charge of the Antitrust Division of the U.S. Department of Justice. 10 A hardcore restriction is a restraint that presumptively meets the conditions of Article 101(1) of the Treaty of the Functioning of the European Union. 11 Vertical Restraints Guidelines, at para 56. I leave open for legal scholars the question of whether this specific guideline is less likely to be enforced in light of Coty. 12 Id. ¶ 52. The exceptions to the restraints on Internet distribution as hardcore restrictions are provided by Article 101(3). 13 Increasing quantity and increasing value amount to the same thing. The restrict shifts out the demand curve, which can be described as an increase in marginal value at any quantity, or an increase in quantity at any price. 14 Pierre Fabre also argued that customers should be able to see the products, but the Court’s response was that visual examination was unlikely to yield any useful information prior to purchase. 15 Coty’s contracts demonstrate the concern for retailer investment to enhance or protect product image. The Coty contract, as described on p. 2 of the decision, includes the following: ‘The décor and furnishing of the sales location, the selection of goods, advertising and the sales presentation must highlight and promote the luxury character of Coty Prestige’s brands. Taken into account when evaluating this criterion are, in particular, the façade, interior décor, floor coverings, type of walls, ceilings and furniture, sales space and lighting, as well as an overall clean and orderly appearance’. 16 The value that consumers place on brand image has positive implications upstream in product supply chains. ‘Nike is one of the business world’s shining examples of how to clean up an image: In the 1990s, the company was plagued by reports that it used sweatshops and child labour. Pressure grew until 1998, when Nike cofounder Phil Knight publicly committed to changing the company’s practices, and Nike spent the next decade doing just that.' [Quartz online magazine, Aug 1, 2017: https://qz.com/1042298/nike-is-facing-a-new-wave-of-anti-sweatshop-protests/] Nike’s efforts to improve its labour practices may be due in part to corporate social responsibility, but are surely driven by the sensitivity of demand to its brand image. © The Author(s) 2018. Published by Oxford University Press. All rights reserved. For Permissions, please email: firstname.lastname@example.org This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)
Journal of European Competition Law & Practice – Oxford University Press
Published: Mar 1, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera