Retail Alliances and Joint Purchasing agreements: Evaluating Benefits and Challenges in light of the Revised Horizontal Guidelines.

Nowadays, retail alliances (‘RA’) are at the core of the European retail market. RA help their members to operate in the highly competitive and rapidly evolving retail sector, characterized by increased consolidation and internationalisation.¹ RA are a specific joint purchasing agreement (‘JPA’), where members collectively negotiate with suppliers to increase their buying power. The increased buying power helps the members to potentially achieve “lower prices, more variety or better quality products for consumers.” Additionally, joint purchasing can help to avert shortages, and prevent interruptions, thereby mitigating disruptions in the supply chain.² Consequently, the past decade has seen a growing number of alliances formed among large retail chains in Europe.³

The benefits derived from these alliances can potentially be passed on to consumers. The European Commission (‘EC’) recently acknowledged the potential consumer benefits associated with RA, stopping its antitrust investigation against two European Retail Alliances, AgeCore and Coopernic. The EC found no evidence of anticompetitive effects, and it emphasised the positive influence of those alliances on reducing prices, which is “particularly important in the current context of high inflation.” ⁴

While JPA may give rise to efficiencies, competition concerns regarding JPA cannot be neglected. First, the discounts negotiated may simply not be passed on to consumers and therefore no efficiencies are created. Moreover, if the buyer power of the JPA surpasses that of the suppliers (e.g., in negotiations with smaller suppliers or suppliers of fresh agricultural products⁵), it can potentially have adverse effects on product variety and innovation.⁶ In extreme cases, a joint purchasing strategy can also result in a buyer cartel, which then becomes illegal and causes disadvantages for consumers. Recent cases⁷ show the anticompetitive side of buyer cartels, such as the €260 million fine imposed on a chemical buyer cartel.⁸ These cases show that it is crucial to clearly distinguish between behaviour that leads to consumer efficiencies and those that result in anticompetitive effects.

The EC recently addressed the distinction between the positive and negative effects of JPAs in its new horizontal guidelines, which were released on July 1, 2023. The horizontal guidelines specifically address the applicability of Article 101 of the Treaty on the Functioning of the European Union (‘TFEU’).⁹ The new guidelines offer greater legal clarity and practical guidance. This guidance allows businesses to better assess the legality of their JPA.

This article's primary objective is to explore the general conditions under which European RA, such as AgeCore and Coopernic, can be beneficial for consumers and when they may not. This examination specifically concentrates on European RA to provide deeper insights into the EC's findings in its investigation of AgeCore and Coopernic.

To explore the conditions of these efficiencies, this article first derives insights from the revised horizontal guidelines. It examines the general conditions that can determine whether JPAs benefit or harm consumers and competitors. Subsequently, it conducts a comparative analysis between these conditions and the specific characteristics of European RA to understand why these characteristics might result in efficiency gains.

During the revision process of the horizontal guidelines stakeholders confirmed that overall, the chapter of the 2011 guidelines on purchasing agreements was useful for assessing JPAs. However, in some segments stakeholders missed the necessary clarity and voiced concerns about their application.¹⁰ Therefore, the new guidelines provide a more distinct differentiation between buyer cartels and JPAs¹¹. They offer a more comprehensive guidance on assessing the potential anticompetitive effects of JPAs, with a particular focus on the upstream market.¹² Additionally, they give more insights into the assessment of RA.¹³ Overall, the revised guidelines closely follow the structure and key points of the 2011 horizontal guidelines on purchasing agreements.

Members of the JPA can engage in collective product purchases or focus solely on negotiating prices and terms with suppliers, after which they make individual purchases based on the negotiated terms.¹⁴ The pooling of purchasing activities can be achieved through various means, including “a jointly controlled company, through a cooperative, by a contractual arrangement or (…) a representative negotiates on behalf of several undertakings”.¹⁵

The EC holds two primary concerns regarding JPAs. First, JPAs have the potential to restrict competition in both the relevant purchasing market and selling markets.¹⁶ Second, these agreements can be misused to form a cartel, either during the negotiation process in the upstream purchasing and/or subsequently on the downstream selling market.¹⁷ Figure 1 presents a simplified visualisation of the market situation. The JPA and other buyers obtain their goods from the upstream seller market. The upstream market is represented by the upstream sellers A and B. Subsequently, the buyers, depicted as downstream sellers A, B and C, distribute their goods further downstream to consumers (or firms), represented by consumer A, B, C and D.

Figure 1. Simplified visualisation of the market situation

Source: CEG

Compared to the 2011 guidelines, the revised guidelines offer a more precise definition of the distinguishing characteristics between JPAs and buyer cartels and offer more clarity on which actions are lawful:

• Representation and Transparency: The JPA must clearly state that it negotiates “on behalf of its members”, who are “bound by the agreed terms and conditions”.¹⁸ While the identity of the members doesn’t necessarily have to be disclosed, it should not be actively concealed either.
• Defined Cooperation: The “form, scope and functioning of their cooperation” must be defined clearly in a written agreement.¹⁹
• Agreement Clarity: The agreement can help to confirm compliance with Article 101 provisions. However, it’s crucial to understand that a well-defined agreement does not make a JPA immune to EC investigations.²⁰
• Imposition of Restrictions: The JPA can impose restrictions on its members, provided these restrictions are “objectively necessary and proportionate” to the cooperation’s functioning.²¹ For example, it is allowed to impose restrictions on non-participation in other JPAs, as such behaviour could potentially limit the JPA’s buying power.
• Commercially Sensitive Information: The exchange of commercially sensitive information is not necessarily forbidden, as long as the JPA itself “does not fall within the Article 101(1) prohibition because it has neutral or positive effects on competition”.²² In such cases, the information exchange should be objectively necessary for implementing the JPA and proportionate to its objectives. To avoid falling under Article 101(1), sensitive data can be collected in ways that mitigate the risk of a spillover (e.g., a separate entity).²³
• Negotiation Threats: The JPA is allowed to use its buying power to make negotiation threats, such as leaving the negotiations or ceasing purchases, to secure better conditions (the same is allowed for the counterparties).²⁴
  

JPA turn into buyer cartels, when they “have as their object the restriction of competition in the internal market contrary to Article 101(1).” ²⁵ In the negotiation process, a buyer cartel aims to eliminate or reduce strategic uncertainty by coordinating individual actions, such as fixing market variables like prices, before entering into negotiations with suppliers. Due to its inherent harmful nature for firms and consumers, evaluating its negative effects is not required.²⁶

The revised guidelines maintain a soft safe harbour provision. The provision states that competition concerns diminish when members lack market power, which is indicated by market shares.²⁷ In this context, the EC defines a 15% threshold for the combined market share of the JPA members, either on the selling or purchasing market[s].

However, it is important to note that a market share exceeding 15% “does not in itself indicate that the joint purchasing agreement is likely to give rise to restrictive effects on competition”.²⁸ Therefore, if substantial competitive concerns arise regarding the behaviour of a JPA, three assessments should be conducted. This assessment is now described in greater detail in comparison to the 2011 guidelines.²⁹

• First, the buying power of the JPA has to be assessed in the context of the countervailing selling power.³⁰ A significant degree of buyer’s power allows the JPA to dictate terms in negotiations with a selling side that lacks the countervailing seller’s power. In such cases, the seller’s investments incentives may decrease, potentially leading to a reduction in the quantity and quality of products. Additionally, the JPA may use its buying power to limit access for competing purchasers, especially in markets with few suppliers and barriers to entry.³¹
• Second, the composition of JPA members in the downstream selling market should be evaluated. If the JPA members are competitors downstream and collectively purchase a substantial share of the products they compete on, their incentives to compete on prices could be “considerably reduced”.³² This effect is more pronounced when these members have substantial market power in the selling market, allowing them to retain the reduced purchaser prices. Moreover, when the JPA holds a significant market share within the downstream selling market, the coordination established during purchasing could extend to coordinated actions when selling these products.³³ This coordination is more likely to occur if the structure (high market concentration, transparency, or a high degree of communality) of the selling market facilitates collusion.³⁴
• Finally, after evaluating the potential restrictive effects, these should be weighed against potentially generated efficiencies. Article 101(3) TFEU, allows restrictive JPAs to be lawful if they contribute to improving production, distribution, or economic progress.^35,36 These positive effects are not limited to the JPA alone but can also extend to the sellers involved. The JPA can incentivize suppliers to “innovate and introduce new or improved products” and help “expanding distribution of their products”.³⁷ The efficiency defence does not apply when assessing a cartel or when restrictions imposed by the JPA result in the elimination of competition.³⁸ Once efficiencies are established, the question arises whether these benefits have been passed on to the consumer. The efficiencies must be passed on to consumers to an extent that outweighs any restrictive effects.³⁹

European RA can be defined by the following key characteristics. First, their members operate in different national markets.⁴⁰ However, it should be noted that retailers, pursue simultaneously multiple alliance strategies and are often active in national RA as well.⁴¹ Second, these alliances often center their efforts on collaborating with premium manufacturing brands to establish agreements, such as rebates.⁴² Alternatively, they may focus on the procurement of private label products by formulating tenders. Third, the decision-making for purchases is decentralized, allowing members to individually decide how much to buy.

Euro Commerce, an organisation representing European retailers and wholesalers, interprets the revised guidelines as an affirmation of the benefits provided by RA.⁴³ In the intensely competitive, consolidating, and international European market, RA play a vital role for their members.⁴⁴ They help their members “to remain competitive, to enhance supply chain efficiency, and to counter supplier power.” ⁴⁵ By joining a RA, members strengthen their bargaining position. Moreover, through resource pooling, RA can achieve economies of scale, such as reducing transaction or negotiation costs and coordinating promotional activities.⁴⁶ At last, RA can assist their members in avoiding vertical restraints imposed by suppliers, concerning the resale of their product.⁴⁷

European alliances often focus often on engaging with “premium manufacturing brands”. ⁴⁸ This characteristic is especially important, given that Coopernic and AgeCore have been established with a primary focus on branded products.⁴⁹ These premium manufacturing brands play a crucial role in the retail sector by exerting countervailing selling power. Their products are considered a must-have, as consumers expect them to be available when making a purchase. In return, RA, through their buying power, help to mitigate potential harm to competition upstream and ensure innovation and product quality downstream.

The countervailing selling power gains added significance, in the context of individual countries serving as distinct markets. Brands typically adhere to strict national purchasing rules, which make it difficult or impossible for supermarkets to purchase products abroad.⁵⁰ Hereby, products tend to be more expensive in smaller countries in comparison to larger countries, as these constitute smaller markets. In the context of supermarkets, smaller markets tend to lead to less competition and higher prices. This effect can be seen by comparing the prices of brands between countries. For example, Coca-Cola is more expensive in the Netherlands than in Germany. Geographic market separation is further reinforced by the labelling of products. Manufactures are not allowed to sell, for example, a product with a German label in the Netherlands. A JPA spanning multiple countries can therefore potentially enhance the buying power of retailers in smaller countries.

Another concern mentioned in the guidelines is the potential that a RA with immense buying power may be able to foreclose other purchasers.⁵¹ A powerful RA could be incentivized to limit the access to efficient suppliers. The presence of several powerful RA in Europe reduces the likelihood that a single RA can foreclose competing purchasers.⁵²

A key characteristic of European alliances that could foster efficiencies is that individual members often represent specific countries. This leads to a situation in which they “do not have overlapping markets and are not direct competitors on the consumer market.” ⁵³ This lack of overlap among alliance members mitigates the concerns mentioned in the revised guidelines. It helps to prevent a significant reduction in incentives “to compete on price on the downstream selling market(s)”.⁵⁴ As this characteristic does mainly apply to European RA, it safeguards to some extent against adverse effects. This member composition of European RA provides an additional layer of risk mitigation against coordination on the downstream selling market. When the combined market share of members on each national downstream selling market is lower, the risk of coordination between upstream purchasing and downstream selling decreases correspondingly.⁵⁵

Furthermore, the geographical dispersion of alliance members, enhances, to some degree, the probability of passing on lower purchase prices, to consumers. The likelihood of benefits from improved negotiations being passed on to consumers depends on the level of competition in the retail market. In general terms, the greater the level of competition, the more likely consumers will benefit from these negotiated cost savings. Given that the RA members lack significant collective market power in their national downstream retail market, the chances of these lower prices being passed on are heightened.⁵⁶ This effect could be further increased by the intense competition characterizing the European retail market.⁵⁷ As the EC noted in its investigation against AgeCore and Coopernic, the “extent of such price reductions depended notably on the degree of competition existing in the relevant downstream retail markets.” ⁵⁸ However, it should be noted that RA must prove that benefits are passed on to consumers. This assessment should rely on their pricing strategy and an (econometric) analysis of their purchase and sale prices. The extent of pass-on is a matter that must be substantiated on a case-by-case basis and cannot be assumed solely based on the prevailing market structure.

Further, it is important to note that in some countries, the retail markets are undergoing consolidation, and certain supermarket chains already individually surpass the previously mentioned 15% threshold, signifying substantial market power and elevating the risk of restrictive effects.⁵⁹ For instance, Epic Partners contains “Edeka, Germany’s number 1 retailer (26% market share); Magnit, Russia’s number 2 retailer (20% market share); ICA, Sweden’s number 1 retailer (48% market share); Jeronimo Martins, Poland’s number 1 and Portugal’s number 2 retailer; and Migros, Switzerland’s number 1 retailer.” ⁶⁰ Consequently, the European Brands Association (‘AIM’) is cautioning against RA becoming too dominant. More generally, concentration ratios are high and have been increasing over the past decades in European retailing.⁶¹ If buyers were to overpower sellers, this could have adverse effects on consumers. Recent studies have brought attention to the potential negative impact of increased buying power on product variety and innovation.⁶²

The future development of this delicate balance of powers between sellers and buyers is difficult to predict, because “European RAs have become increasingly more dynamic in their formation and reorganisation”.⁶³ As a result, the assessment of buying power may change in the coming years, necessitating ongoing observations.

The role of private label products in the market is also a significant consideration. Suppliers often highlight the growing proportion of private labels in total turnover, successfully competing against premium brands.⁶⁴ Suppliers see this as especially detrimental in combination with the increased consolidation and cooperation of retailers. This trend becomes even more relevant in times of inflation when consumers tend to turn to more affordable white-label products.