Price Parity Clauses on Internet Platform Markets

Table of Contents

1. Introduction

2. Legal framework and case study
2.1 Legal perspective
2.2 The Amazon case

3. Platform Markets
3.1 Definition
3.2 Critical mass
3.3 Pricing
3.4 Multi-homing

4. Price Parity Clauses
4.1 General
4.2 Meeting-Competition
4.3 Most Favored Customer
4.4 Combinations
4.5 Exclusive contracts
4.6 Predatory pricing and limit pricing

5. Application on Internet Platform Markets
5.1 Meeting-Competition
5.2 Most Favored Customer
5.3 Exclusive contracts
5.4 Predatory pricing and limit pricing

6. Conclusions

7. References

1. Introduction

On 30th August 2013, the internet platform Amazon stopped its price parity clause across Europe (Office of Fair Trading, 2013). This affected millions of internet users, because more than one billion articles were sold worldwide via Amazon in 2013 (Amazon, 2014a). Platform markets have become increasingly important since the dissemination of the internet (Evans & Schmalensee, 2010, p. 1). We can no longer imagine everyday life without internet platforms such as Amazon, the auction platform eBay, the booking website HRS and also social media platforms such as Facebook. They are "'must have platforms' for many businesses. [...] Access to these platforms, either to advertise or sell products, is a key requirement for many businesses" (OECD, 2013, p. 49).

Shopping on internet platforms saves time and reduces searching costs, because a wide range of products is easy to find on only one platform as well as through search engines and comparison portals (Bundeskartellamt, 2013a, p. 16). Furthermore, the internet provides the ability to compare prices and products very easily. However, disadvantages exist in the form of shipping costs and the fact that consumers are not able to physically "try on" a product before buying it (Loginova, 2009, p. 321). These problems especially occur in the case of goods that have "attributes that can be directly determined through our senses, particularly touch, smell, or sound before purchase a product" (Loginova, 2009, p. 321). Therefore, many platform operators offer services to reduce those problems. For example, Amazon offers free shipping costs for some consumers (Amazon, 2014b) while Zalando, a German fashion platform, offers fast shipping and free redemption, so consumers can try on the clothes and send them back for free, if they do not fit (Zalando, 2014).

Increasingly more new platforms arise every year and the already consisting firms have to compete with and on internet platforms. This leads to many challenges for all participants, prompting consumers, sellers and platform operators to adapt their behavior of interacting. These challenges go together with risks and problems. The reach of the sellers by using the internet increases dramatically. Theoretically, internet users can reach every online seller in the world from their home computers. However, this is usually not the case in reality, due to the large number of similar businesses (Bundeskartellamt, 2013a, p. 16). Firms normally need to advertise their platforms in their target areas. This is possible via the internet (i.e. banners or advertisements in search engines) or offline (i.e. advertisements in television or newspapers). Furthermore, shipping time and costs are significantly higher when ordering from other countries (Bundeskartellamt, 2013a, p. 17). Additionally, some products may be prohibited in other countries. Worldwide operating companies usually have offices in different areas to serve the local target groups. Accordingly, national borders still play an important role for the internet economy. In 2012, 98% of German online buyers ordered or bought their products from sellers in Germany, while 14% (also) bought inside the European Union and only 8% (also) made purchases outside the European Union (Bundeskartellamt, 2013a, p. 16). This clearly illustrates that borders still play an important role and thus high potential remains for growth and thereby increasing competition.

However, the reduced transaction costs in the internet economy are very strong compared to offline markets. Consumers can find thousands of products on Amazon's website by using their computer or smartphone from any place where they have an internet connection. In turn, sellers are less limited in their exhibition and storage space. Especially the rise of social media platforms such as Facebook has allowed both sellers and buyers to intensify communication. This gives firms the possibility to better position their products to their target group, while consumers receive a better service. As a result of the wide reach, it can be profitable for firms to offer niche products and build up specific know-how for them (Bundeskartellamt, 2013a, p. 17). As I will show in the next chapters, there is a growing trend towards this direction. Many well common platform operators such as Amazon (books) and Facebook (Harvard students) were initially launched as niche platforms, before subsequently expanding their supply.

In conclusion, it can be said that the internet leads to a significant reduction of transaction costs and an intensification of competition, which benefits consumers via lower prices, higher quality due to more innovation, more product variety, more information on different prices and quality, as well as improved service. However, there is increasing pressure on competition for the sellers (Bundeskartellamt, 2013a, p. 17). This is true for both producers and retailers. Therefore, offline sellers often need to change their strategy. While some use the internet as a complement to the traditional offline business, others have started to focus on online sales only. In addition, many completely new internet-based products and services have been developed and compete with traditional products (i.e. e-books versus traditional paper books). This increase of competition has a welfare-enhancing effect (OECD, 2013, p. 111).

Different forms of vertical restrictions are well known and used within the internet economy due to the increasing competitive pressure. Some firms such as the French producer of pharmaceutical and cosmetic products Pierre Fabre have used restrictions to avoid competition across the internet, which has been prohibited by the European Court of Justice (Bundeskartellamt, 2013a, p. 23). Since restrictions on internet platforms are still relatively new, the analysis from a legal perspective is quite difficult. I would like to illustrate this situation based on the case of Amazon, who introduced a price parity clause, which has been under investigation by antitrust authorities.

Before I start with the case, I like to demonstrate the economic growth of the internet economy by introducing some important facts. This will highlight the relevance of this topic and illustrate the importance of the case for future development. The number of internet users in Germany and worldwide has been growing rapidly in recent years, and will be ongoing in the next years. In May 2012, 57 million people in Germany used the internet at least once in three months, with 74% of these people, namely 42.3 million (in other words, half of all German inhabitants), making purchases via the internet during the past 12 months (Bundeskartellamt, 2013a, p. 14). This illustrates that using the internet for purchases is already very common in society, with further growth expected. The internet provides many different ways to promote products through photos, videos, texts, etc. (Bundeskartellamt, 2013a, p. 14). To ensure the provision and usage of the wide range of promotion, a fast internet connection is often necessary. From 30% in 2006 and over 50% in 2008, in 2012 already 75% of households in Germany had access to a fast internet connection. It is planned that all German households will gain access in the future. Therefore, there is still a high potential for even more purchases via the internet. Most consumers use their home computer for internet purchases. The share of purchases via smartphone was only 10%, although this number is rapidly increasing (Bundeskartellamt, 2013a, p. 15). Therefore, people will be connected to the internet even more often and hence will be able to make more purchases in the future. Of course, the increasing usage of the internet and the related growing number of internet sales also affect revenues. The total expenditure for physical products and digital services via the internet by end consumers was 37.3 billion Euro in 2012, reflecting an increase of 25% compared to the year before. Goods that have been distributed via the internet were sold for a total of 27.6 billion Euro in 2012, which has been an increase of even 27.2% compared to the year before. The expected increase for 2013 was still 21.3%, which is 33.5 billion Euro in total. While there are different independent calculations about the exact amount of revenue and growth, all of them agree on a rapid increase and an ongoing growth within the next years. Relatively cheap prices for end consumers and low transaction costs are considered as the main reasons for the rapid increase (Bundeskartellamt, 2013a, p. 15).

The share of the total revenue of internet sales shows that the potential of this branch for even more revenue remains very high. The 37.3 billion Euro represent less than 10% of the total revenue of retailers in Germany. The share becomes a little higher (14%) when excluding non-durable goods such as food and drugstore articles, which traditionally need to be checked in the store prior to purchase (Bundeskartellamt, 2013a, p. 17). The share of online sellers in the German retail market strongly varies between different branches. For instance, the share of electronic goods such as TV, computers or phones was 17% in 2012, while the share of fashion was 16.8% (Bundeskartellamt, 2013a, p. 18). The purchase of electronic goods mainly depends on measurable facts such as power, capacity or speed, with such details relatively easy to compare on the internet. The high share of fashion can be explained by the trend of rising platforms such as Zalando, which offer quick deliveries and free redemption. On the other hand, the share of home improvement and gardening was relatively low with only 2.6% in 2012 (Bundeskartellamt, 2013a, p. 18). Home improvement articles usually need to be experienced by consumers with their own hands before purchasing. Gardening articles are usually relatively sizeable and thus entail high shipping costs. Furthermore, consumers need to see them for real to consider whether they fit into their garden. This shows that whether something "fits" into the internet depends on the kind of good. Depending on the different branches, shares up to 50% are expected (Bundeskartellamt, 2013a, p. 18).

Based on a survey in November 2012 concerning the online and offline purchasing behavior of consumers in Germany, 7% of all transactions are made via the internet, creating a resulting revenue of 16%. The purchases were mainly made in specific branches. For instance, the share of electronic goods is 27% of the amount that was spent on online purchases in total, while 22% was spent on fashion. Sometimes, online and offline purchases are connected with each other. In the survey, 20% of online purchases were prepared offline, while 17% of offline purchases were based on online research. Due to the significantly higher revenue generated in offline retail, the purchases that were provided online and performed offline generated a revenue eleven times higher than those that were researched offline and executed online (Bundeskartellamt, 2013a, p. 18). Based on this survey, the range of consumer types buying online is very wide. However, there is a tendency towards young people using the internet more than old people do. This can be considered as another trend towards an increasing use of the internet for purchases in the future. Furthermore, it has been shown that the most important question for the decision whether to buy online or offline was whether a consumer likes to take the product directly home or have it delivered (Bundeskartellamt, 2013a, p. 18).

To summarize, one can say that the internet economy is already a very important field of study in today's economics and will be even more important in future. Firms will seek to benefit from the internet by using different restraints. The question whether these restraints are legal is not easy to answer. Given that research on the internet economy is still relatively new, there are not many cases from the past to compare. However, some important cases have led to pioneering decisions made by antitrust authorities. One of the most common cases of vertical restraints in the form of a price parity clause is the case of Amazon. I would like to analyze this case as well as the decision of the antitrust authorities based on literature about platform markets and different kinds of price parities. Furthermore, I would like to analyze the reason for the decision, before concluding whether the decision and/or the reason in this case was correct based on the literature I used.

In the second chapter, I will introduce the current legal perspective under which the German antitrust authority, the Bundeskartellamt, decided to prohibit Amazon's price parity clause. I will show that the law on which the Bundeskartellamt referred to was the Treaty on the Functioning of the European Union (TFEU), which covers the entire market of the European Union. The Bundeskartellamt worked together with the Office of Fair Trading in the United Kingdom. As result, Amazon stopped its price parity clause, not only in Germany but also all over the European Union. Subsequently, I will illustrate the Amazon case in detail. Amazon used the price parity clause for about three years. After the antitrust authorities started to investigate the clause, Amazon stopped the price parity clause by itself, and thus the Bundeskartellamt also stopped the investigation after a while (Bundeskartellamt, 2013b).

In the third chapter, I will introduce the theory of platform markets. Platform markets usually connect two sides with each other, often buyers and sellers, although they can have various different forms. I will show how platform operators charge prices and what difficulties occur when launching a new platform by achieving the so-called critical mass. Furthermore, I will illustrate the acting of multi-homing, which means using multiple platforms at the same time and how this behavior affects pricing and competition between platforms as well as between sellers.

In the fourth chapter, I will introduce different kinds of price parities. Meeting-Competition as well as Most Favored Customer clauses are very common in today's life and used by a wide range of firms in different markets. Such clauses are often used as a marketing strategy to give the impression of particularly cheap prices (Arnold, Baake & Schwalbe, 2012, p. 12). The assumptions in research are sometimes different. The basic theory assumes an automatically paid rebate, while extended research adds so-called hassle costs, which reduce the price increasing effect of those clauses. Exclusive contracts are often necessary to launch a new platform. However, such contracts usually have anticompetitive effects. Predatory pricing and limit pricing are difficult to evaluate from a legal perspective, since they actually lead to lower prices. Based on the theory, the effects are anticompetitive because they cause market entry barriers and market exit.

In the fifth chapter, I will apply the theory used in chapters 3 and 4, showing how differently the effects work on platform markets. Meeting-Competition and Most Favored Customer clauses do not have a strong effect when applied by platforms due to different pricing strategies on various platforms based on the framework. However, when sellers on platforms introduce such clauses, the effect is very strong due to the wide reach of sellers via internet, and thus the range of affected competitors is large. Exclusive contracts turn out to be necessary tools for platform operators to achieve critical mass in order to enter the market. However, when critical mass is achieved, the effects become anticompetitive because the contracts complicate market entry for other platforms. Predatory pricing is very difficult to detect when it is performed by platforms, because the optimal price can even be below marginal costs in some cases. When sellers use predatory pricing strategies on platforms, the anticompetitive effect is also very strong, owing to the high number of affected competitors.

In the last chapter, I will analyze these effects and compare them with the Amazon case. I will show that Amazon's price parity clause does not reflect any of the introduced restrictions, but has similar effects and hence should be handled similarly. Based on that, I will show that the decision of the antitrust authorities to prohibit Amazon's price parity clause was correct, although the reason that the clause represents a horizontal restriction was wrong. Finally, I will conclude this case and suggest what should be undertaken in future cases.

2. Legal framework and case study

2.1 Legal perspective

To answer the question of whether price parity clauses should be prohibited and, if so, which conditions are important, it is necessary to analyze the legal framework. Article 101 TFEU covers the "prohibition of anticompetitive agreements and concerted practices" (Article 101 TFEU). The article contains vertical as well as horizontal restraints (Bundeskartellamt, 2013a, p. 18). The first paragraph describes in which cases vertical or horizontal restraints are anticompetitive and hence should be prohibited:

"The following shall be prohibited as incompatible with the internal market: all agreements between undertakings, decisions by associations of undertakings and concerted practices which may affect trade between Member States and which have as their object or effect the prevention, restriction or distortion of competition within the internal market, and in particular those which:

(a) directly or indirectly fix purchase or selling prices or any other trading conditions;
(b) limit or control production, markets, technical development, or investment;
(c) share markets or sources of supply;
(d) apply dissimilar conditions to equivalent transactions with other trading parties, thereby placing them at a competitive disadvantage;
(e) make the conclusion of contracts subject to acceptance by the other parties of supplementary obligations which, by their nature or according to commercial usage, have no connection with the subject of such contracts" (Article 101(1) TFEU).

This paragraph primarily applies to collusions and cartels. Based on economic theory, both entail higher prices than the competitive one and are therefore legally prohibited in most industrial countries (Bester, 2004, p. 133.). In result, cartels reduce the incentive for innovation and more efficiency and complicate the market entry for new competitors that might be more efficient.

It is not always easy to monitor whether vertical restraints contain collusion. Exclusive contracts as well as price parity clauses on platform markets can be considered as vertical restraints because they hinder potential competitors from achieving a sufficient number of participants to run a platform (Bundeskartellamt, 2013a, p. 8). The third paragraph of the article relativizes the first paragraph by naming conditions for exemptions, stating:

"The provisions of paragraph 1 may, however, be declared inapplicable in the case of:

- any agreement or category of agreements between undertakings,
- any decision or category of decisions by associations of undertakings,
- any concerted practice or category of concerted practices,

which contributes to improving the production or distribution of goods or to promoting technical or economic progress, while allowing consumers a fair share of the resulting benefit, and which does not:

(a) impose on the undertakings concerned restrictions which are not indispensable to the attainment of these objectives;
(b) afford such undertakings the possibility of eliminating competition in respect of a substantial part of the products in question" (Article 101(3) TFEU).

This paragraph shows that horizontal or vertical restrictions can be legal in special cases, even when they entail fixed prices, limited production, etc., when restrictions lead to positive effects such as improved distribution or benefits for the consumers. Therefore, it is necessary to audit individual cases and check whether positive effects sufficiently outweigh negative effects. Theoretical efficiencies are not sufficient to gain exemption under Article 101(3) TFEU. It is not enough if they are only merely conceivable; rather, they need to be presented in concrete terms and their probability must be considered (Bundeskartellamt, 2013a, p. 11).

The so-called Vertical Restraints Block Exemption Regulation (VRBER) is a regulation designed by the European Commission. Its task is to define a group of agreements that normally meet the conditions defined in Article 101(3) TFEU (Bundeskartellamt, 2013a, p. 11). Therefore, the objective of the VRBER is to filter out agreements that are likely to have benefits for the competition and the consumer and hence justify a general exemption (Bundeskartellamt, 2013a, p. 11). In general, the level of market power plays an important role for the VRBER when defining conditions for exemptions. The European Commission defines market power by market share, because it is relatively easy to measure. As a condition for a group exemption to apply, the market share of supplier and consumer must each be 30% or less. However, there is no presumption that agreements with a higher market share violate Article 101(1) TFEU or do not satisfy the conditions to fulfill Article 101(3) TFEU, although neither is there the presumption that agreements falling under Article 101(1) TFEU will meet the conditions of Article 101(3) TFEU (European Commission, 2010a, p. 10). Furthermore, the Commission defined so-called hardcore restrictions that, independent from the market power, raise the assumption that they violate Article 101(1) TFEU. These hardcore restrictions include restrictions of the possibility of the buyers to independently set their prices by themselves or restrictions of the place or the customer groups that are served (Bundeskartellamt, 2013a, p. 11). The European Comission stated the definition as follows:

"Including such a hardcore restriction in an agreement gives rise to the presumption that the agreement falls within Article 101(1). It also gives rise to the presumption that the agreement is unlikely to fulfill the conditions of Article 101(3), for which reason the block exemption does not apply" (European Commission, 2010a, p. 17).

This means that in case of such restrictions, the contract is no longer in the part of application of the Vertical Restraints Block Exemption Regulation and hence should be prohibited. However, it is not a per se prohibition, because in there exist possible constellations in some specific cases that apply to the exemption conditions based under Article 101(3) TFEU.

For a better understanding, I would like to show whether a restraint violates Article 101(1) TFEU and/or applies to Article 101(3) TFEU based on case studies in Germany in recent years. In 2005, the Bundeskartellamt imposed a fine upon the manufacturer of hearing aids Phonak GmbH for the amount of 4.2 million Euro (OECD, 2013, p. 115). Hearing aids are usually sold by retailers to end consumers. In the case of Phonak, a retailer published prices of several producers on the internet. Phonak's prices turned out to be significantly below the market standard minimum price. After that, other retailers of hearing aids claimed about the price cutter. As a result, Phonak threatened them with a delivery block to force them to raise their resale prices (Bundeskartellamt, 2010, p. 103). While the application of a recommended retail price is usually permitted, the threat of advantages or disadvantages for retailers to put this price through has anticompetitive effects. A delivery block is obviously a disadvantage and therefore Phonak was penalized (Bundeskartellamt, 2010, p. 104). This case applies to Article 101(1) TFEU, which states that restrictions that "directly or indirectly fix purchase or selling prices or any other trading conditions", should be prohibited (Article 101(1) TFEU).

In a similar case, the Bundeskartellamt imposed a fine for the amount of 11.5 million Euro to the CIBA Vision Vertriebs GmbH in 2009. CIBA is a producer of contact lenses (Bundeskartellamt, 2010, p. 98). The company used a monitoring and intervention system to control retail prices. When retailers set their prices below the recommended retail price, CIBA contacted them and forced them to change the price. Similar to the Phonak case, such behavior must be prohibited based on Article 101(1) TFEU (Bundeskartellamt, 2010, p. 99). These cases are interesting in terms of the comparability of the case of Amazon. When, inter alia, analyzing the case from the legal perspective in the fifth chapter, I will include these examples. In the next subchapter, I would like to present the case of Amazon in detail.

2.2 The Amazon case

Amazon.com, Inc. was founded in 1994 as a platform for books and is currently the largest online retailer in the world. It has 113,000 employees and its annual revenue was 74.45 billion US-Dollar in 2013 (Amazon, 2014c). Amazon is a typical business-to-consumer online platform, whereby its consumers are end users. Amazon acts as a retailer selling goods by itself, as well as a platform for external sellers. The share of articles sold by third-party sellers has strongly increased over recent years, from 28% in 2005 to around 40% in 2012 (Mulpuru & Walker, 2012, p. 6). In 2012, Amazon's share of internet sales in Germany was between 30% and 40% (Bundeskartellamt, 2013c, p. 1). Amazon's total revenue in Germany in 2012 was 6.6 billion Euro (Bundeskartellamt, 2013a, p. 14). About one billion articles were sold worldwide on Amazon in 2013, while the number of active sellers increased by 65%. Today, more than two million sellers offer products on the platform. The increasing trend of fashion sold via the internet (i.e. Zalando in Germany) also affected Amazon. The total amount of clothes sold via Amazon doubled in 2013 compared to 2012 (Amazon, 2014a). This clearly demonstrates the company's position as a market leader in Germany.

Before sellers are allowed to place their products on the platform, they need to sign a contract with Amazon. This contract contains general terms and conditions, as well as several rules with which sellers have to comply. Here, Amazon provides a product presentation, user accounts, payment method, mode of shipment and the conditions for return. The sellers post their products on the platform, so they become presented to the consumers in a specific ranking. These products are presented together with the products that Amazon offers by itself as a retailer. Based on the final price, sellers need to pay a provision to Amazon after each purchase (Bundeskartellamt, 2013c, p. 1). Part of the contract was the so-called price parity clause, which was introduced in March 2010 in Germany (Greif, 2013) and in the United Kingdom (Office of Fair Trading, 2013). This clause prohibited sellers from setting a lower price on other platforms, including their own website, than they set on Amazon. The clause stated:

"By our General Pricing rule, you must always ensure that the item price and total price of an item you list on Amazon are at or below the item price and total price at which you offer and/or sell the item via any other online sales channel" (Amazon, 2014d).

A very important point is that the clause only applies to online sales channels and thus excludes offline markets. The price parity clause is defined in much detail, so there are not many ways to avoid the clause, i.e. by having different launch dates, charging different shipping costs or making "low price" offers (Le Roux, 2012). Amazon regularly monitored whether its sellers complied with the price parity clause. In the case of non-compliance, Amazon sent reminders (Bundeskartellamt, 2013c p. 2). In these reminders, Amazon listed the products that are monitored to be cheaper on other platforms than Amazon and their total prices, including discounts, rebates, promotions and shipping costs (Ogborne, 2012). In the case of non-compliance, Amazon threatened to exclude sellers from the platform.

The Bundeskartellamt started an investigation based on Article 101(1) TFEU about the price parity clause. In their argumentation, the platform represents a trade cooperation with the purpose and effect of a restraint of competition. The price parity clause thereby represents a hardcore restriction that neither "contributes to improving the production or distribution of goods or to promoting technical or economic progress" nor allows "consumers a fair share of the resulting benefit" based on Article 101(3) TFEU. Since Amazon sells its own articles as a retailer, the Bundeskartellamt does not see a vertical agreement between Amazon and its sellers using the platform. From its perspective, a platform itself is harmless. Even though the harmonization of competition parameters is necessary, it ensures efficiencies and hence is exempted by Article 101(3) TFEU. Parameters that entail inadequate disadvantages related to the efficiencies cannot be exempted. This is the case forthe price parity clause inthe Bundeskartellamt's opinion (Bundeskartellamt, 2013c, p. 2). An application for exemption based on Article 101(3) TFEU would be theoretically possible, if the platform improved the distribution of goods and thereby allows the consumers a fair share of the benefit. Given that Amazon is the market leader and so well known and used by most online consumers, it strongly reduces searching costs. However, this happens due to Amazon's high number of consumers and would hardly change in the case of no price parity clause. Additionally, the clause still cannot be considered as an improvement for the production or a fair shade for the consumer and hence represents a hardcore restriction (Bundeskartellamt, 2013c, p. 3).

In particular, markets for products that are sold on Amazon's website suffer from this clause. For the Bundeskartellamt, Amazon's price parity clause is a horizontal price-fixing agreement for which the pricing of only one contractual party is sufficient in this case, as long as it is a direct competitor. Therefore, the clause represents a hardcore restriction in all product categories. Based on a survey of 2,500 sellers, prices increased after the introduction of the price parity clause. Furthermore, the clause complicates market entry and inhibits the expansion of competitors that are already in the market (Bundeskartellamt, 2013c, p. 3).

Following the investigation, Amazon stopped its price parity clause by itself, and thus the Bundeskartellamt also stopped the investigation. It was unclear in the first place when exactly Amazon stopped its price parity clause in Germany, since they did not officially announce it. However, on 27th August 2013, the Bundeskartellamt officially confirmed the end of Amazon's price parity clause in a press release, after Amazon told them they had stopped it (Bundeskartellamt, 2013d). Subsequently, the Bundeskartellamt took its time to ensure that the price parity clause has been completely terminated. After three months, the Bundeskartellamt officially announced in a press release that the investigation against Amazon had been terminated on 26th November 2013 (Bundeskartellamt, 2013b).

During the investigation, the Bundeskartellamt closely cooperated with the British Office of Fair Trading, which is the main antitrust authority in the United Kingdom, because Amazon used the same price parity throughout Europe and hence was under investigation for the same reasons as in Germany, given that Article 101 TFEU applies for the entire European Union (Bundeskartellamt, 2013c, p. 4). In the United Kingdom, Amazon announced the end of the price parity clause in the European Union via an official statement:

"Effective beginning August 30th, the provision in your seller agreement requiring price parity between your product listings [...] and your other sales channels will no longer be enforced. If we revise your agreement in the future we will remove this provision" (Edwards, 2013).

However, Amazon still uses the price parity clause in the United Statese. I would like to analyze the effects of price parity clauses on internet platform markets based on the theory of platform markets in combination with literature about different kinds of price parity clauses such as Meeting-Competition clauses, Most Favored Customer clauses, exclusive contracts, predatory pricing and limit pricing. Based on this analysis, I will conclude whether the decision of the Bundeskartellamt was correct or not. Furthermore, I will analyze their justification, showing that the decision was correct but the reason was incorrect.

3. Platform Markets

3.1 Definition

To analyze the effect of price parity clauses on internet platform markets, it is important to know about the peculiarities of these markets. The literature on platform markets is relatively new (Evans, 2013, p. 2). Reasons for starting research are that platforms feature several issues for pro- and anticompetitive analysis, while platform markets are often under investigation for the use of vertical restrictions, such as the case of Amazon. A platform brings customers of different sides together in order than they can deal with each other. It provides mechanisms that facilitate interactions such as matching, search and exchange (Evans, 2013, p. 3). Hence, platforms reduce transaction costs that potential customers of both sides would otherwise invest to interact with each other (Evans & Schmalensee, 2008, p. 673). There are also multi-sided platforms, although most platforms are two sided, as in the case of Amazon. Therefore, I will focus on the most common kind of platforms, connecting customers on the one side with customers on the other side. In his article from 2013, David S. Evans defines platforms as follows:

"Platforms create value by serving as intermediaries between two or more types of customers where one type of customer can realize value by interacting with another type of customer. The demand by one type of customer depends on the participation on the platform of one or more of the other types of customers" (Evans, 2013, p. 2).

A platform can be any place that is run to bring together two sides and can be designed in many different ways, such as shopping malls, payment systems, newspapers, dating websites, job agencies, auction houses, etc. The demand of one type of customer on one side of the platform depends on the participation of one or more customers of the other side due to positive indirect network externalities, which lead to positive feedback effects between the sides (Evans, 2013, p. 2). This means that the higher the number of participants on one side, the higher the probability for participants of the other side to find a match (Evans & Schmalensee, 2008, p. 670). From this perspective, the participation on one side increases when the participation on the other side increases, and vice versa. Hence, high participation on the platform makes the platform attractive and creates positive feedback effects, which signal attractiveness to other potential participants.

A large platform would also have the advantage of economies of scale for the platform operators because service costs per user would decline. Especially when fixed costs are high and costs per user are low, economies of scale occur. In some cases, scale economies only exist on one side of the platform. For example, newspapers connect the consumers on the one side with advertisers on the other side. The costs for creating the newspaper (paying journalists, writing stories) might be relatively high, albeit cheap to print. Hence, by printing a high number of copies, the newspaper can reach more readers, although the space for advertisements does not change (Evans & Schmalensee, 2008, p. 679).

This would mean that the more participants there are on both sides of the platform, the better the platform would work. From this perspective, the best situation would be if all customers were on one single platform and the number of users was as high as possible. However, new problems would occur in such a situation, such as an overcrowded platform. On offline platforms, this is easy to imagine, such as an overcrowded shopping mall with busy shops and long waiting queues. Furthermore, the web servers do not have infinite space on internet platform markets. Another problem would be the costs when too many customers use the platform. In this case, the operator must invest money and/or labor to monitor whether the customers comply with the rules. Additionally, more service for several issues must be provided. As explained above, platforms seek to reduce transaction costs, including searching costs, which are lower than in case without the platform. However, the more participants there are on the platform, the higher the searching costs. Sometimes, one side benefits from searching costs on the other side (Evans & Schmalensee, 2008, p. 677). Imagine a shopping mall, in which consumers usually want all stores close to each other to save shopping time on the one hand. However, one the other hand, the stores want as much foot traffic as possible around their stores, and thus the most popular stores in the mall are placed on the opposite ends of the mall.

Some platforms intend to have a small number of customers on their platform. This is especially the case when they try to be different from other platforms via product differentiation (Evans & Schmalensee, 2008, p. 680). For instance, dating websites match men and women. On the one hand, the best situation for women would be if there were as many men as possible on the other side to have maximum choice and vice versa. On the other hand, this would entail high searching costs because both sides try to find an appropriate partner. Therefore, some dating platforms focus on a lower number of selected people. For example, the German dating platform Elitepartner.de focuses on people with high education for relatively high prices to reduce searching costs (Bruhns, 2012).

However, this does not mean that the number of participants on a platform should be low. Because platforms are usually profit maximizers, there is an optimum number for every platform depending on their strategy. However, before they start maximizing profit, a minimum number of participants on both sides of the platform must be achieved to run the platform. This minimum number is called the critical mass, "which involves a sufficient number of members of both sides to create enough value to attract more members of each side" (Evans, 2013, p. 4).

3.2 Critical mass

The market entry of a new platform is relatively difficult to perform. Imagine a new platform that wants to connect customers on both sides. At the beginning, the platform is almost empty. As a high number of customers on both sides of the platform is necessary to attract customers to participate on the other side, due to positive feedback effects, there is no reason for any customer to join because there is no customer on the other side to interact with. To start a new platform successfully, an operator needs to "get both sides on board" and "solve the chicken-and-egg-problem" (Evans & Schmalensee, 2010, p.4). When this is achieved, there will be positive feedback effects on both sides of the platform (Evans, 2013, p. 5). Solving this problem is the first task to undertake when operating a platform. Only after this, the platform operator can start to maximize its long-term profit (Evans & Schmalensee, 2010, p. 5). If this does not happen, whereby the critical mass is not achieved, negative feedback effects will occur and customers will leave the platform (Evans, 2013, p. 5).

Suppose a new platform that likes to compete with existing ones and assume that customers can only use one platform at the same time. There would be no reason for customers to leave their platform and change to a new one. Of course, consumers differ from each other. Therefore, platforms should focus on those potential customers who are likely to join new platforms and like to try new things, which are the so-called "early adopters", namely customers who expect that the platform will reach the critical mass (this can be seen as an investment) and customers with an especially high value of joining (Evans, 2013, p. 5). This high value could be living near a new shopping mall. Even when the new mall would be a copy of the former one, the value would be higher because the transaction costs are lower when it is closer to the consumers. In internet platform markets, this is rarely the case. Sometimes, new platform operators have already achieved the critical mass before launching the platform; for example, the Discover Card in the credit card market.

[...]