Themes and Takeaways from the FTC Hearings on the Intersection of Big Data, Privacy, and Competition By Elena Kamenir (Orrick, Herrington & Sutcliffe)1
On November 6-8, 2018, the U.S. FTC held hearings on privacy, big data, and competition as a part of its Hearings on Competition and Consumer Protection in the 21st Century series.2 The hearings covered a wide array of issues, including whether privacy concerns affect competition analysis, how to evaluate conduct and mergers involving big data, which remedies are appropriate to address antitrust harm involving certain datasets, and the policy recommendations and implications surrounding these issues. This article first summarizes common issues discussed on the panels. Although there was much consensus among panelists, differences emerged regarding the intersection of competition, privacy, and big data. This article concludes by outlining three issues on which antitrust and consumer protection practitioners will continue to focus on and debate.
Common Issues Discussed During the Hearings
- What Makes Data Valuable?
One of the threshold questions when analyzing the impact of big data is whether the data is valuable. Panelists generally agreed that the value of data is highly dependent on how the data will be used and analyzed. Florian Zettelmeyer identified three categories exemplifying how data analytics can be useful. Data analytics can: (1) enable business initiatives; (2) enable idea generation; and (3) help decision-makers evaluate whether a strategy is working (e.g., advertising analysis). Mark MacCarthy echoed the sentiment that the value of data depends on its intended use, noting both that sometimes a lot of data is necessary to achieve a result and that the value of some data diminishes quickly, while other data could remain valuable for years.
On a separate panel, Alex Okuliar testified that not all data is created equal and that the value of data depends on several factors, including its use and whether the data is unique. For example, because machine learning is a process of dynamic experimentation, varied data that offers a multiplicity of signals tends to be more useful. In a panel involving former enforcers, former FTC Chair Maureen K. Ohlhausen recognized that, in the merger context, it is theoretically possible that certain combinations in which consumer data is a key asset could lead to market power but that this outcome seems less likely unless a showing is made that the data could not be replicated or has no substitutes.
- What Competition Issues Can Big Data and Privacy Concerns Present?
Panel discussions also covered whether consumers consider privacy practices when purchasing goods or services and whether firms compete on privacy. Several panelists agreed that there is consumer demand for privacy. Alessandro Acquisti noted that there is evidence that consumers demand privacy protections when the price to protect privacy is small. As privacy protections become costlier, consumers will demand privacy when there is a clear benefit. However, especially online, it is difficult to predict what the cost of privacy will be because privacy tradeoffs are intertemporal. On another panel, former FTC Commissioner Julie Brill stated that, from her perspective, companies are competing on privacy but in ways that may not be as intuitive or obvious.
Another common topic was whether large volumes of data create scale economies or can act as a barrier to entry. Florian Zettelmeyer stated that large amounts of data can be extremely important in training algorithms. More data can enable an algorithm to make extraordinary predictions and this ability could affect competitive strategy and business models. Ginger Zhe Jin noted that there are examples of entrants disrupting incumbents without a data advantage, although it is possible that the incumbent did not make efficient uses of its data at the time. Following up on this point, Alessandro Acquisti opined that for some firms without a first-mover advantage, one important characteristic differentiating these firms from the incumbents is that they benefitted from the self-reinforcing dynamics of network effects. On a separate panel, Allen Grunes explained that barriers to entry can be higher where data is a critical input or where there is an ability to scale up. Michael Baye added that, if data is viewed as a barrier to entry, one potential reason is because there are economies of scale and scope associated with collecting data.
Panelists also debated whether big data enables price discrimination. As Ginger Zhe Jin explained, price discrimination from an economic perspective does not necessarily reduce consumer welfare or total welfare. Liad Wagman stated that we are starting to see individualized pricing and that these efforts are likely to intensify. Alessandro Acquisti added that we may not be observing first-degree price discrimination but may instead be observing forms of self-selection, which is a form of second-degree price discrimination. On another panel, Jonathan Baker noted that, although selective discounting can be good for consumers, it can also harm consumers where it operates as a mechanism to exclude rivals. During a panel focusing on online advertising, Allie Bohm stated that data-driven advertising can have procompetitive benefits but may also allow companies to gain insights, via algorithms, that could decrease their incentive to cut prices to remain competitive.
- What Kind of Experience Do Agencies Have in Analyzing Data-Related Competition Issues?
Testimony throughout the hearings made clear that U.S. enforcement agencies have ample experience analyzing data issues arising in mergers. Haidee Schwartz identified several FTC examples. The FTC evaluated data as a product in CoreLogic/DataQuick, data as a key input in Verisk/EagleView, whether access to data presented a barrier to entry in CCC/Mitchell, and whether the data was unique in Microsoft/LinkedIn.3 Similarly, the U.S. DOJ considered data issues in its investigation of the search agreement between Microsoft/Yahoo!4 as well as the Bazaarvoice/PowerReviews and Thomson/Reuters mergers.5
- Which Remedies Can Address Data-Related Competition Issues?
The hearings also touched on what remedies, if any, are appropriate for addressing competitive harm in data markets. Questions on this issue covered a variety of topics, from how remedies can be tailored to solve data issues to how remedies can affect innovation. For example, Courtney Dyer stated both that the aim should be on getting the remedial party to use the data in a way that enhances competition and that imposing long-term entanglements between the parties is not necessarily always beneficial. Andrew Gavil posited that it is important to consider whether the data is separable. The panelists further considered whether data is an essential facility under the current legal framework. D. Daniel Sokol explained that it is not clear that data could be viewed as an essential facility because of its characteristics, such as the ability for users to multihome and the ability to assemble and disassemble data, among others.
Areas of Focus and Debate Moving Forward
The hearings represent a positive step in analyzing issues concerning big data and privacy. Moving forward, the debate among panelists suggests greater focus on the following topics as businesses continue to evaluate their privacy practices and incorporate data-driven tools and solutions:
- The Types of Big Data that Should Be Scrutinized. There is some consensus that big data can benefit consumers and, in certain circumstances, can be used to harm consumers by creating barriers to entry. Nevertheless, debate remains about which characteristics enable anticompetitive conduct and whether these concerns are overstated. Analysis on this topic is likely to intensify as newer and larger uses of big data emerge.
- The Appropriate Framework of Analysis. The agencies will need to evaluate whether the antitrust or consumer protection laws (or both) provide the appropriate framework for analyzing certain types of conduct at the intersection of big data and privacy. There was some consensus among panelists that businesses do compete on privacy, but questions remain as to whether the antitrust laws are the appropriate mechanism for evaluating data collection-practices that pose threats to privacy. For antitrust issues concerning big data, the appropriate framework will in part depend on the conduct at issue. There will likely be greater focus on whether certain doctrines—such as the essential facilities doctrine or the two-pronged test for predatory pricing—are applicable to the collection and use of big data.
- Special Considerations for Data Collection and Advertising Practices by Two-Sided Platforms. As two-sided platforms rely more heavily on data-driven tools, additional commentary on whether big data can create barriers to entry or enable firms to engage in anticompetitive behavior will likely come into stronger focus and debate. For example, more economic and legal analysis on whether data collection can enable scale and network effects that cause high barriers to entry is likely to emerge. Similarly, debate about the competitive effects of targeted advertising strategies that allow for price discrimination or product discrimination is likely to continue.
No matter the outcome of these debates, it is important that any new enforcement approaches to data and privacy practices be supported by empirical evidence and experience so as to strike an appropriate balance between addressing anticompetitive behavior without harming innovation.
1 Elena Kamenir is an associate in the Washington, D.C. office of Orrick, Herrington & Sutcliffe LLP. The views expressed in this article are solely her own and are not meant to reflect the views of Orrick or any current or former clients.
2 Fed. Trade Comm’n, FTC Hearing #6: Competition and Consumer Protection in the 21st Century (Nov. 6-8, 2018), https://www.ftc.gov/news-events/events-calendar/ftc-hearing-6-competition-consumer-protection-21st-century.
3 Haidee L. Schwartz, FTC Experience with Data Markets, Hearings on Competition and Consumer Protection in 21st Century (Nov. 6, 2018) at 129-141, https://www.ftc.gov/system/files/documents/public_events/1418633/cpc-hearings-au_11-6-18.pdf.
4 Press Release, Dep’t of Justice, Statement of the Department of Justice Antitrust Division on Its Decision to Close Its Investigation of the Internet Search and Paid Search Advertising Agreement Between Microsoft Corporation and Yahoo! Inc. (Feb. 18, 2010), https://www.justice.gov/opa/pr/statement-department-justice-antitrust-division-its-decision-close-its-investigation-internet. The DOJ also evaluated an advertising agreement between Yahoo! and Google. See Press Release, Dep’t of Justice, Yahoo! Inc. and Google Inc. Abandon Their Advertising Agreement (Nov. 5, 2008), https://www.justice.gov/archive/opa/pr/2008/November/08-at-981.html.
5 Complaint, United States v. Bazaarvoice, Inc., No. 13-cv-00133 (N.D. Cal. Jan. 20, 2013), https://www.justice.gov/atr/case-document/file/488911/download; Complaint, United States v. Thomson Corp., No. 1:08-cv-00262 (D.D.C. Feb. 19, 2008), http://www.justice.gov/atr/cases/f230200/230281.htm.