TPRC 2024: The 52nd Research Conference on Communications, Information and Inter

Link to paper

Abstract:
Mergers in mobile markets are of keen interest to policy makers and scholars. Because carrier networks are subject to pronounced economies of scale and scope and given that communications regulators create substantial barriers to entry by limiting spectrum allocations for mobile services, wireless services generally exhibit relatively high levels of industrial concentration. Hence, antitrust authorities often struggle with the tradeoff between enhanced scale economies and enhanced market power. Between 2012 to 2016, for instance, four E.U. nations (Austria, Ireland, Germany, and Italy) consummated “4-to-3” mobile mergers while two such combinations were blocked (in Denmark and the U.K.). In the U.S., 4-to-3 transactions were blocked by regulators in 2011 and again in 2014, but a recent merger -- between the No. 3 (T-Mobile) and No. 4 (Sprint) carriers was approved in February 2020. This combination remains a subject of intense debate. We examine post-merger evidence of retail mobile subscription prices, network investment, service quality, market shares, and industry profits in the U.S. mobile communications industry. We conclude that the data are consistent with the thesis that the T-Mobile/Sprint merger produced consumer gains. This outcome is particularly interesting given that the government remedy imposed to mitigate potential anti-competitive merger effects, the creation of a new fourth network (DISH), has produced no plausible pro-competitive impact.

Link to paper

Abstract:
The Digital Markets Act (DMA) is a cornerstone of European digital platform regulation. Gatekeepers are required to enable interoperability of their number-independent interpersonal communications services (NI-ICS), which are designated as core platform services (CPS) like e.g. messaging ser-vice WhatsApp. To this end the DMA defines a three-stage implementation of interoperability, start-ing with simple bilateral text communication with further features to follow over a period of four years (DMA, Art. 7(2)).
Using a consumer survey in Germany with 2,826 respondents, this study examines the potential effect of interoperability on consumer preferences for using specific NI-ICS. As NI-ICS are subject to network effects, demand tends to gravitate towards services with a larger user base, which has re-sulted in a highly concentrated market. Horizontal interoperability could counteract such tendencies towards concentration. However, this desired effect is highly dependent on the willingness of con-sumers to use NI-ICS in an interoperable manner, and which NI-ICS they choose to use as a result of interoperability.
There are two main findings from our consumer survey:
First, we find that almost all NI-ICS users in Germany tend to use NI-ICS (97%), which are designat-ed as CPS. WhatsApp alone is currently used by around 93% of NI-ICS users and is typically ac-cessed on a daily basis. In contrast, only 50% of NI-ICS users use an alternative (non-gatekeeper) service. Overall, alternative services are used less frequently and tend to play a subordinate role in the interpersonal communication of German users.
Second, very few NI-ICS users are opposed to interoperability, 64% consider it useful, and at each stage of implementation more than half of respondents indicate that they would use interoperability - at least in some cases – to communicate with specific contacts, via certain functions or via certain services. However, the majority of NI-ICS users that intend to adopt interoperability expect to contin-ue to use gatekeeper services to the same extent as they do now. In addition, the proportion of users who intend to use gatekeeper services to a greater extent than before the introduction of interopera-bility slightly outweighs the proportion of users who intend to use these services less. With regard to the use of alternative NI-ICS, the proportion of users who say they will use alternative NI-ICS less than before the introduction of interoperability is about twice as high as the proportion of users who intend to use these services more.
Therefore, interoperability as mandated by the DMA could lead to a reduction in the use of alternative NI-ICS by those NI-ICS users who choose to enable interoperability. It may even increase the incen-tive to use gatekeeper services, thereby contributing to the strengthening of gatekeeper ecosys-tems. Our results therefore call into question the incentive for alternative NI-ICS providers to de-mand interoperability from gatekeepers in the first place. Overall, our analysis suggests that the effect of the horizontal interoperability obligation for NI-ICS in the DMA may be contrary to the in-tended effect, or at least will not mitigate the concentration tendencies in the market for NI-ICS.

Link to Paper

Abstract:
A widening set of applications may soon be offered by broadband providers and edge providers, including home security, smart home, video conferencing, gaming, IoT device connectivity, autonomous vehicle connectivity, industrial control, and smart factory connectivity.

Multiple policy and regulatory issues surround these specialized services. Can they be offered over the Internet or do they require special treatment? When offered by broadband providers, are these offerings evading Open Internet rules regarding application-specific discrimination or prioritization? Can third-party application providers compete on a level playing field with ISPs offering such specialized services? What is the regulatory classification of the service and/or the underlying connectivity? Are existing regulatory schemes sufficient? If not, what rules should apply?

We consider two policy objectives. First, the development of specialized services should be encouraged, or at least not impeded, by the regulatory landscape. Second, to the extent there is a tradeoff between specialized services and broadband Internet access service, the regulatory landscape should strike a balance that is in the public interest.
Under the 2015 Open Internet Order, specialized services do not reach large parts of the Internet, are not a generic platform, and/or use isolated capacity. We propose an alternative approach, based on the Computer II framework, to classify the connectivity underlying specialized services as a telecommunications service, and to require this connectivity to be offered on a nondiscriminatory basis.

Under our proposed Computer II framework, a broadband provider could offer prioritized applications such as video conferencing and gaming, providing that it offers the same prioritization to end users on a nondiscriminatory basis. A broadband provider could also offer applications bundled with connectivity, such as home security, smart home, telemetry, automobile services, and industrial applications, providing that it offers the underlying connectivity to edge providers on a nondiscriminatory basis.

For prioritized applications, end users would be offered broadband plans that include the ability to prioritize applications of their choice. An end user could prioritize an application offered by a broadband provider and/or one offered by a third-party application provider. For bundled applications, end users could choose to obtain an application such as home security that runs over their broadband service or to obtain a bundle that includes both the application and the required underlying connectivity.

Link to paper

Abstract:
Robert Bork’s The Antitrust Paradox (1978) has been a justification for a lack of antitrust enforcement for over four decades. This paper addresses how the Bork criteria can be used today, if applied correctly, to supplement the approach to antitrust. His test asks if consumers are harmed by the pricing practices of the firm in the market in which they purchase the good or service. Even if these firms are monopolies or oligopolies in their fields with huge economic rents, if they pass this test, no action is taken against them. “Bigness is not bad.” This narrow view, inter alia, ignores two- and multisided markets (MSM) where the appearance of “no harm” is addressed to only one side of the market. The correct view is to examine all the markets impacting potential harm to consumers. It illustrates the harm which is “free” to the users, but advertisers pay dearly for the ability to micro-focus on potential consumers of their products. Facebook and Google are used as examples. This advertising cost is added to the sales price of the product, resulting in consumers being harmed by the embedded advertising costs in the products or services purchased. We argue here, using Bork’s own criterion – except to expand it to the other side of the market – that much needed antitrust action has been ignored by this narrow criterion. This analysis indicates that antitrust action is long overdue after considering two-sided markets. In addition, we argue that his “consumers’ welfare” criterion is misleading and liable to deceive, thus the deception. The Biden administration has taken a more aggressive approach to antitrust and consumer protection than previous administrations. Three antitrust suits have been filed against Alphabet (Google) and two against Meta (Facebook) in the United States. The European Community, which has been more aggressive in its pursuit of Big Tech – Alphabet (Google), Amazon, Apple, Meta (Facebook), Microsoft, etc., is considering additional rules and regulation of the sector. This paper addresses how the Bork criteria can be used today, if applied correctly, to supplement the neo-Brandeisian approach. We address the Bork Paradox in its own terms by examining the second side of the market which harms consumers indirectly by increasing the price of the products and services they purchase. Using the corrected Bork metric – both sides of the market and no producer’s surplus – the estimated loss of consumers’ welfare was $101.9 and $36.8 billion respectively from Alphabet (Google) and Meta (Facebook), respectively in 2023. The cumulative loss from 2013 to 2023 is over $888 billion

Link to paper

Abstract:
The paper covers a 40 year history of media ownership in the United States