Georgios Petropoulos, Geoffrey Parker and Marshall Van Alstyne review what the Meta antitrust case reveals about its merger and acquisition strategy and what lessons it holds for regulators as they seek to protect competition in digital platform markets.
One key objective in the ongoing FTC v. Meta court case is to decide whether Meta’s acquisitions of Instagram in 2012 and WhatsApp in 2014 reduced competition and harmed consumers.
The Federal Trade Commission argues these acquisitions prevented the emergence of competitors to Meta (previously Facebook) in the social networking market, which produced several societal harms. These include increased user ad loads, reduced investment in service quality and content, and more aggressive invasion of user privacy. The FTC compares these quality reductions to the counterfactual case where the Instagram and WhatsApp acquisitions never took place, implying that competition would have compelled Facebook to improve quality on all three fronts to keep users from leaving.
To understand the validity of these concerns, we need to look first at Meta’s merger and acquisitions strategies and its objectives. What this analysis reveals is a three-prong M&A strategy, some of which is potentially harmful, and some of which is not. Regulators must focus on devising new strategies for preventing harmful M&A.
Our research has documented Big Tech acquisition strategies, including Meta’s, over two decades to categorize the logic. Meta started its M&A activity in 2007 with a focus on creating a user-friendly social network experience. That motivated the acquisition of companies providing such functionalities as shaping online conversation (e.g., Zenbe, Octazen), enabling photo sharing (e.g., Karma, Atlas Solutions), creating an environment to share travel stories (e.g., Chai Labs), and providing updates for live events or online instant messaging (e.g., Snaptu, Oculus VR). Other acquisitions focused on monetization through targeted advertising (e.g., Pursuit, Storylane). In the past several years, Meta has actively acquired companies that specialize in computer vision (e.g., Scape Technologies), virtual and augmented reality (e.g., Sanzaru Games), artificial intelligence, and machine learning (e.g., Servicefriend, Atlas ML, Fayteg AG).
One important dimension distinguishing Meta from other Big Tech platforms is that it has consistently acted to acquire new talent through its acquisitions. An M&A strategy that mostly involves hiring new talent but not transferring new technology is commonly referred to as “acquihiring.” Big Tech can acquire talent, with or without acquiring competitors, in order to thwart competition.
Working from primary data, we developed a taxonomy of overlapping categories of motives behind Meta’s acquisition strategies. These motives include:
- Complementary functionality can help Meta provide more efficient services related to its core business, such as virtual reality (examples: Spool, Storylane, Pebbles).
- New functionality adds products or services in the vertical value chain that make the platform more attractive to users, such as the provision of mobile data access (examples: Pryte, Infiniled, Wit.ai).
- Substitute functionality incorporates competing services for a firm’s core intermediation function at the platform layer, reducing overall market competition (examples: FriendFeed, Instagram, WhatsApp).
This third category of M&A strategies generates the most concern with respect to market competition and consumer welfare.
The acquisitions of Instagram and WhatsApp involve dynamic innovation challenges related to the FTC’s three main concerns. Did an incumbent firm acquire targets to discontinue the target’s innovation projects and to preempt future competition? In theory, consumer welfare suffers because consumers miss the benefits from increased competition as well as the alternative consumption choices from new products and services within the same market that would have developed if the acquisition had not taken place.
At the heart of the FTC’s argument, Instagram’s growing 30 million users at the time of its acquisition signaled that it could emerge as a viable Facebook competitor, especially in mobile ecosystems. In the counterfactual world, more competitive pressure would have led Meta to keep its investment in service quality high to improve its value to users. Instead, the FTC argues that the Instagram acquisition helped Meta protect its market position and ensure prolonged dominance in social networks. In the same spirit, the FTC argues that WhatsApp’s growing 465 million mobile-messaging users could likewise have developed into a competing mobile social network. Acquiring core user bases is a common platform growth strategy. Hence, Meta sought not only to protect Facebook Messenger, its main messaging app, from a new competitor, but also to shield its core social network platform from a similar service entrant.
The fact that debate continues more than a decade after these 2012 and 2014 acquisitions reveals how challenging their assessment is. Clearly, merger review at that time required more careful consideration of possible anticompetitive factors. Factors include dynamic innovation concerns (ex-ante effects), how the merger investigation changed Facebook’s behavior (maneuvering effects), and estimating the merger’s actual impact on industry structure and user experience, compared to the counterfactual where it did not occur (ex post effects).
Assessment is particularly challenging in digital markets because of tradeoffs. Network effects do create value for users by granting access to larger networks, but platforms can cause privacy or competition harms and jealously guard that information. They explain their cagey behavior based on claims of protecting users’ privacy or guarding trade secrets. In a bout of circular logic, they assert regulators cannot show the occurrence of harms necessary to block mergers yet they hide any data that would demonstrate harm. Yet, when regulators do gain access, as the Meta case shows, harms have indeed occurred.
To provide but one example of data misuse, a United Kingdom parliamentary inquiry into Meta’s privacy protocols revealed that Meta used Israeli analytics company Onavo “to conduct global surveys of the usage of mobile apps by customers, apparently without their knowledge. They used this data to assess not just how many people had downloaded apps, but how often they used them. This knowledge helped them to decide which companies to acquire, and which to treat as a threat.”
Big Tech platforms are much more likely to have such insights than the authorities responsible for overseeing them. This information asymmetry limits competition authorities’ abilities to assign motives and measure harms when assessing M&A activity.
Moving forward, we suggest three lessons for competition authorities on closing information gaps, on data rights, and on shifting the burden of proof.
First, authorities should use digital survey techniques to run massive online choice experiments, examining the preferences of hundreds of thousands of consumers across different services. Academics have developed robust methodologies to run such experiments, which offer snapshots of social network substitutability at given points in time. These tests yield at least two insights. First, they offer data defining digital market boundaries. Second, they provide convincing counterfactual analysis to assess and explain the potential competition effects when authorities do block mergers. Implementation would then require authorities to build the proper in-house expertise to follow this route.
Second, we should re-weight the allocation of merger review and litigation costs based on this approach. At one extreme, some scholars have argued that laws should reverse the burden of proof for investigations into and litigation of large platform mergers. Currently, regulators must prove that a merger would harm competition, else it may happen by default. Reversal would imply that Big Tech platforms should objectively justify efficiency gains before regulators or the courts would allow an acquisition. This shift could bring investigation costs down as authorities would need to only check the validity of efficiency justifications provided by the platform.
Complete reversal goes too far, however, with potential harm for entrepreneurial activity. Small firms often launch with the goal of becoming acquired by Big Tech platforms, while executives and investors view buyouts and launching yet more startups as a major success. Our analysis, in fact, has indicated that a majority of platform mergers involves such small innovative firms where the risk of foreclosing potential competitors is low.
Given this, we argue elsewhere that the universal reversal of the burden of proof, which presumes all platform mergers are anticompetitive, is excessive given likely damage to entrepreneurship. Instead, we would advocate for a reversal of the burden of proof for those limited horizontal mergers between potential substitutes, or direct competitors, where they appear most problematic and anticompetitive. Leave the burden with regulators for vertical app-on-top new features that serve as complements but reverse the burden of proof for horizontal infrastructure-on-bottom that serves as substitute. The size of the user base and online traffic statistics also help determine relevant thresholds.
Third, a legislative rather than judicial remedy is to grant users “in situ” data rights, allowing them to import algorithms of their own choice into the platforms where their data reside. This contrasts with data portability or “ex situ,” which removes data from the platform causing it to obsolesce, lose context, and become inactionable off-platform. In situ data rights improve transparency. When New York University researchers received users’ permission to track Facebook ads, misinformation, and what role Facebook played in the January 6 insurrection, Facebook terminated the researchers’ access. Researchers also showed that the archive of political ads Facebook made available for public analysis omitted over 100,000 ads. Users who wanted to know how Facebook was (mis)treating them were denied access to that knowledge. In situ data rights not only close the information gap that lets platforms hide their behavior from users and regulators but also foster competition. Startups would not need Facebook’s permission to launch new features; instead, they just need to share innovation value with users to get users’ permission.
The ongoing FTC v. Meta case, which follows the Department of Justice’s success in the recent Google Ad Tech ruling, highlights the revival of Section 2 of the Sherman Antitrust Act. As United States antitrust enforcers become more active, we call for equipping users with in-situ data rights, equipping regulators with access to information and tools for acting upon that information, and shifting the burden of proof in those limited, most potentially problematic cases where features are not vertical complements on top of a platform but horizontal substitutes for that platform.
Authors’ Disclosures:
Georgios Petropoulos was affiliated with the economics think tank Bruegel from November 2015 to March 2022 as an employee and from March 2022 to November 2024 as a non-resident (non-paid) fellow. He has not had any affiliation with Bruegel since November 2024. Meta is a corporate member of Bruegel.
Geoffrey Parker testified in Federal Tax Court for the Internal Revenue Services in its lawsuit against Facebook in Docket No. 21959-16. The testimony took place in March 2022.
Marshall Van Alstyne provided expert advice on Facebook Marketplace relative to Craigslist in August 2022. He declined compensation to avoid a conflict of interest.
Articles represent the opinions of their writers, not necessarily those of the University of Chicago, the Booth School of Business, or its faculty.
