Assessing Monopoly Power or Dominance in Platform Markets: A Summary
This post is part of a series covering different studies on antitrust enforcement of digital services released by various academics, think tanks, and regulatory agencies.
Recently Koren W. Wong-Ervin, Director of Antitrust Policy & Litigation at Qualcomm Inc., released a paper entitled “Assessing Monopoly Power or Dominance in Platform Markets.” This paper explores issues such as whether to define one integrated market or two separate-product markets; the use of market shares as a proxy for monopoly power; potential barriers to entry such as network effects, data, user switching costs, and consumer biases; and the role of multihoming. The paper in turn responds to key questions in each of these topics drawing from existing economic literature.
I. Challenges of Defining Markets
The paper starts by discerning the challenges around market definition of platforms. In this respect, the author notes that the key question is to define if the market at issue is one integrated market or two separate-product markets. The author takes the view that if a competitive-effects analysis is undertaken that rejects the separate-effects assumption, that harm to one group of consumers in the platform context equates to overall harm to competition.
The question then becomes whether or not efficiencies on one side of the market are weighed against an identified loss of competition on the other side. In doing so, it must be determined whether the markets at issue are considered to be two sides of the same market, or distinct markets that are interrelated. Economic literature states the type of market definition should depend on the strength and direction of the network effects. These effects are described as externalities originating on one side of the platform affecting participants on other sides of the platform or the operation of the platform itself. This definition presents a critical difference between a single-sided market and a platform. As former FTC Commissioner Joshua Wright and former FTC Acting Deputy Assistant Director John Yun have noted, a separate-markets approach can lead to the discounting of cross-group effects, but the use of an integrated-market approach can lead to a singular focus on a “net” price, which might be difficult for a court to employ.
Regardless of the approach taken to defining a relevant market, the most important step is in conducting an integrated competitive-effects analysis. Finally, she argues that competitive constraints both from single- and two-sided platforms must be considered to determine whether they may be viable substitutes. This is predicated upon defining the relevant market.
II. Market Shares
The paper then discusses whether market shares are a proxy for monopoly power. The short answer to this question is: “No.”
The author argues that there is very little empirical basis to presume any systemic relationship between market structure, competition, and innovation. It is noted that the empirical literature that attempts to link market structure and product market competition to innovation is based on cross-section analyses that do not produce causal inference and yield inconclusive results. It is especially difficult to determine market share when analyzing multi-sided platforms. As David Evans and Richard Schmalensee note, “[o]ne could compare shares of each sides (sic) across platforms and then make a judgment about market power based on looking at the shares for both sides, but there is not reason to expect those shares to be equal.” Evans’ suggested alternative to market shares is to look at the risk adjusted rate of return on investment, or assessing the degree of market power by determining the extent to which incumbents are constrained in pricing and innovation behavior by the prospect of new entry.
III. Barriers to Entry
The paper additionally addresses specific claims of barriers to entry in platform markets. The author argues that the definition of barriers to entry is a controversial topic. The author notes some argue that almost anything that makes it “hard” to gain entry to a market could be considered a barrier to entry. One possible challenge is that a new entrant may require a large sum of capital, however, it has been found that many platforms that become successful often start small and expand over time. Consumer confidence also plays a role as it is likely that a consumer may not switch to a new provider unless they are confident others will also switch to that provider. On a larger scale, coordination between consumers and firms internalizes the benefits of network effects, eliminating network externalities and avoiding technology lock-in. However, there are other barriers to entry such as network effects, data as an input, user switching costs, and consumer biases.
A. Network Effects
Network effects are usually indirect, between different kinds of customers. These indirect network effects occur when the size and intensity of participation on one side affects welfare on another side of a market. Network effects can have both positive and negative effects, in some cases they can be responsible for the concentration of market power by one firm, and in others they can help hasten the decline of a dominant player. As Evans points out this effect happens as consumers either enter or leave a market, and in doing so can entice others to enter or leave with them. He continues, noting that with regard to indirect network effects, consumers would not join or use a platform that did not contain most if not all of their personal network. Evans argues that consumers can use multiple online communications platforms and thus competition can exist even amongst platforms that may serve the same purpose. Further, network effects are not simply about garnering more customers, but the right type of customer that will be provided the most value from the platform. Finally, network effects are often limited to the user’s interactions with the digital platform rather than across the entire user base.
B. Data as an Input
This barrier, simply put, centers around the platform’s ability to access, collect relevant data, and turn that data into a useful product that can lead to value creation. The more efficiently a platform can access, collect and process data the more successful they are likely to be over those who cannot process data as efficiently.
C. User Switching Costs
Switching costs can also be seen as barriers to entry. Platforms that initially compete aggressively attract customers and because of high switching costs are then able to charge monopoly rents after their base is established. However, increases in technology to collect user data and decreases in pricing to attract new customers can combat this. Additionally, consumer coordination can mitigate or eliminate technology lock-in as previously stated, by orchestrating movement of consumers from one platform to another to increase benefits from consumption.
D. Consumer Biases
Default settings can save time and avoid transaction costs that the consumer would often rather not deal with at the outset of obtaining a product. The question remains, how easy is it for users to make alternative choices? This is based on the ease of switching and interoperability of services. It follows that a barrier to entry should not be inferred in markets where switching is relatively easy.
Wong-Ervin concludes with the role of “multihoming.”The author argues multihoming can have positive competitive effects. When multihoming exists on one side of a market, research has shown that price competition on the other side of that market has increased as well. This has been shown not just to be the case in social media but also in search engine platforms — if the user is unable to find what they are looking for on one, they may try another. The inducement of more consumers to multihome can make platform-specific content available to more consumers, which is beneficial not just to consumers but content providers as well.