Rebooting Antitrust and Regulation for Digital Communications

This paper analyzes the development of a “tight oligopoly on steroids” in the communications sector. It uses Business Data Service as an example because it represents a new choke point in the sector. The other services in the “tight oligopoly” include wireline and wireless, cable and set top boxes, and broadband. Using antitrust concepts in the Merger Guidelines the paper argues that high levels of concentration are reinforced by geographic separations, technological specialization, and product segmentation to magnify market power. The paper also uses Alfred Kahn’s discussion of the justification for regulation to explain the long and successful dual jurisdiction (antitrust and regulation) used to control the pervasive market power in the sector. The paper concludes by showing how the concepts and recommendations can be applied to the digital platforms (search and social media), in which an even worse “tight oligopoly on steroids” exists.

By Mark Cooper & Amina Abdu¹

 

This paper uses Business Data Services (“BDS”)² as an example of a broad problem that affects antitrust and regulatory oversight of the communications sector and its primary consumer products – wireless and wireline telephone service, cable connectivity devices (set top boxes) and services,³ and broadband.⁴ The unique challenge is a “Tight Oligopoly on Steroids” that came to dominate the sector because of the inherent economic conditions in communications markets and a combination of lax antitrust enforcement and weak regulation since the Telecommunications Act of 1996.

Section 1 describes the how the “tight oligopoly on steroids” came to dominate the communications sector. Section 2 shows that the Trump administration reversed the oversight that the Obama administration had begun to impose with “Flip-Flop” orders that administratively repealed the Communications Act of 1934 by largely deregulating the market power of the dominant firms. It reviews the evidence of abuse that the FCC ignored in making its deregulatory decision based on its new, and erroneous theory of “sufficient” competition.

Section 3 introduces antitrust and regulatory concerns about market power. Section 4 discusses justification for regulation based on Alfred Kahn’s analysis Section. Section 5 identifies key steps that must be taken to prevent the abuse of market power in the “traditional” communications sector. In Section 6 the conclusion briefly explains why the analysis also applies to big digital information platforms.⁵

 

I. THE TIGHT OLIGOPOLY ON STEROIDS: AVOIDING REGULATION AND COMPETITION

BDS are a new chokepoint in the communications network, the point at which the ocean of data coursing through the digital network becomes a stream directed to each individual consumer. These services have been growing at a rate of almost 15 percent per year for a decade and a half, driven by the fact that high capacity, high quality, always-on connections are vital to a wide range of businesses and economic activities. As shown in Figure 1, affected services include not only communications – mobile, broadband and video – but all forms of high capacity connection for hospitals, ubiquitous networks like ATMs, and the evolving “Internet of Things.”

Figure 1: Business Data Services, the Choke Point of Access to the Digital Network⁶

The pervasive problem in the sector is high concentration, which is magnified by several other characteristics that are well recognized in the antitrust literature (see Figure 2).The same four firms constitute a tight oligopoly across four communications product markets, meaning that the number of firms needed to engage in parallel and reciprocal conduct is very small.⁷ Their history prior to the Telecommunications Act of 1996 and their pattern of expansion since have resulted in geographic separation of home (fortress) territories, technological specialization, and product segmentation. These are the steroids that enable them to dampen rivalry and abuse market power.

The tight oligopoly on steroids was driven by two reinforcing patterns of behavior by the dominant incumbents. The goal of the communications companies after the 1996 Telecommunication Act was a push to define services via the categories that carried the fewest public interest obligations and were generally the least regulated. Neither law nor economic convergence required this direction of policy convergence. Even if policymakers concluded that it was too complicated to maintain the distinctions (known and reviled as silos), policy could read the law in the opposite direction – declaring that, where services involved mixed functionalities, applying the strongest regulatory category and broadest public interest obligations was (more) consistent with the purpose and intent of the Communications Act.

Figure 2: The Tight Oligopoly on Steroids⁸

In fact, the 1996 Act explicitly stated that regulation should be relaxed only where competition, or other factors, had rendered it no longer necessary in the public interest.⁹ It stated that the definition of services, which would trigger the public interest obligations, should not be dictated by the technologies used.¹⁰ The fundamental values of the Communications Act, coupled with real world experience could have guided policy in a different direction. But given the tenor of the times, there was an “irrational exuberance” for deregulation.¹¹

The second thrust was to avoid competition. The dominant firms did not invade each other’s service territories and were slow (at best) in offering products that might compete head-to-head with other dominant firms. The problem was compounded by a weak view of antitrust that was dominant at the time. A massive sustained merger wave was allowed to severely concentrate all the communications markets. Twenty years after the passage of the 1996 Act, much of the old Bell system has been put back together (in three pieces) and that structure has been extended to mobile through the merger waves that affected both landline and wireless. Two cable providers, Comcast and Charter, have come to dominate broadband Internet access service, while continuing to hold a dominant share in the MVPD market.

Given the specialized nature of network industries, it was reasonable to expect that these firms would be the “ideal” candidates to engage in head-to-head competition by geographic extension (overbuilding their neighbors) or product extension (adding a new product to an existing line), but they merged instead, removing the best candidates to promote competition. Given their central location, some markets, like BDS possess unique forms of vertical market power that pose a broad threat to competition and consumers. Other markets, like the one for video programming, are national and the problem of monopsony power is important.

 

II. IRRATIONAL EXUBERANCE FOR DEREGULATION CONTRARY TO THE EVIDENTIARY RECORD

Given the significant and repeated examples of coordination – sometime explicit, frequently parallel, and the reinforcing behaviors in multiple market, it is proper to call the current situation a “virtual cartel” or a “tight oligopoly on steroids.” That being the case, there should be no pretense that competition is sufficient to protect consumers. The amount of scrutiny they require is magnified by the important role they play and their central location as chokepoints and bottlenecks in the digital communications sector and the digital economy.

The Obama FCC, understanding this problem, banned illegal business practices and was addressing how to restore Fair Reasonable and Nondiscriminatory rates terms and conditions (FRAND in antitrust, just and reasonable rates for in FCC-speak). The Trump administration, however, headed in exactly the opposite direction. In a series of “Flip Flop” orders, the FCC sidestepped the data, rather than try to refute it.¹² It invented a new theory of “sufficient” competition.

The data show that the BDS market is not only one of the most concentrated markets in the entire digital communications sector (with four firm concentration ratio [CR4] values close to 100 percent and HHI indices in the range of 6000 to 7000), but also that it is rife with market power abuse in contracting practices. The firms that dominate the BDS market (like AT&T and Verizon), have a near monopoly derived from the long-standing franchise services offered and the ubiquitous deployment of the network during the legal monopoly period. New entrants could not overcome the huge advantage of a fully deployed network and the anticompetitive practices implemented by the incumbents. Perhaps the most telling evidence is that when members of the tight oligopoly must purchase BDS services outside of their home territories, they do not buy services from outside the oligopoly lest they create more viable competition.

The BDS proceeding that was the basis of the FCC’s 2016 order showed that in-building competition and nearby potential competition, (in-census block) were the key to competitive pressures. The FCC chose a much larger geographic market using 3,000 counties to incorrectly deregulate about 90 percent of customers, because the cost of extending facilities across the county to reach customers makes them uneconomic as a competitor.¹³

The FCC “Flip Flop” went on to claim that no actual competitors were necessary to discipline market power, since the threat of entry was enough to prevent abuse. It claimed that it did not have to regulate because market power abuse would not occur even where there were no actual competitors, hence it is a “0 competition” rule. The theory of sufficient competition was identical to the theory of contestability that was thoroughly refuted when it was erroneously applied to transportation networks in the 1980a, as shown in Table 1.¹⁴ Communications markets clearly exhibit a large number of characteristics that make theory inapplicable to the majority of its properly defined markets. The FCC’s own data showed that adding in building competitors lowered prices much more than in-census block competitors and adding the eight competitor still lowered prices by 10 percent.¹⁵

Table 1: Market Conditions that Render Contestability (Potential Competition) Ineffective in Disciplining the Abuse of Market Power¹⁶

 

III. ANTITRUST CONCERNS ABOUT MARKET POWER

These data are generally consistent with the recent change in the DOJ guidelines that “relaxed” the threshold for highly concentrated markets from six to four and the threshold for moderately concentrated markets from ten to six.

In highly concentrated markets (now four or fewer) the DOJ finds:

Highly Concentrated Markets: Mergers resulting in highly concentrated markets that involve an increase in the HHI of between 100 points and 200 points potentially raise significant competitive concerns and often warrant scrutiny. Mergers resulting in highly concentrated markets that involve an increase in the HHI of more than 200 points will be presumed to be likely to enhance market power. The presumption may be rebutted by persuasive evidence showing that the merger is unlikely to enhance market power.¹⁷ While highly concentrated markets trigger the greatest concern, moderately concentrated markets are also a concern.

Moderately Concentrated Markets: Mergers resulting in moderately concentrated markets that involve an increase in the HHI of more than 100 points potentially raise significant competitive concerns and often warrant scrutiny.¹⁸

We suggest that a superior conceptualization of thresholds lies between the two, “four is few, six may be okay and ten is competitive, depending on the inherent characteristics of the sector.” For the communications products addressed here, the distinction is largely moot, since these services generally have fewer than four equal sized competitors. The important point here is that, while there is certainly more competition in the communications sector or the digital information sector (i.e. digital platforms like search (google) and social media (Facebook)

This was the thrust of the Order and Further Notice of Proposed Rulemaking (“FNPRM”) published by the FCC in May 2016.¹⁹ Ignoring this record, the FCC changed direction, reiterating the claim that potential competition and weak actual competition are sufficient to eventually deliver just, reasonable and non-discriminatory rates several years in the future.²⁰ The new theory of potential competition deregulates more markets and products than the failed old theory. The record shows not only that the old theory failed, but it also shows why the new theory will fail as well.

• The key elements of the market structure analysis are well-supported in economic theory and practice, applied to the data in the record.
• Measured by customers, the markets are highly concentrated, but more importantly, the overwhelming majority are monopolies with the remainder being duopolies. This conclusion is dependent on the product and market definition.
• The product is defined as always-on guaranteed quality of service. Different products that do not guarantee service, like best-effort cable, are not considered good substitutes, even by the cable companies who supply them.
• The geographic market is defined as a building, or a census block, because of the prohibitive cost of extending connection over large distances.
• The need for ubiquity is also a barrier to competitive entry.
• These problems are more intense in the lower capacity time division multiplexing (TDM) services.

 

IV. REASONS TO REGULATE TO PREVENT ABUSE AND PROMOTE PROGRESS IN COMMUNICATIONS

In the second edition of his classic work, Economics of Regulation,²¹ published less than a decade before the enactment of the Telecommunications Act of 1996, Alfred Kahn identified a series of characteristics that could justify regulation, although he was generally critical of the way regulatory oversight had been practiced.

A. Infrastructure and Externalities

Making the case for economic regulation, Kahn pointed to the fact that, because communications networks exhibit economies of scale, the market will support only a small number of large firms compared to other sectors of the economy.²² In addition, because of the essential inputs the communications networks provide, they influence the growth of other sectors and the economy.²³ They are infrastructure.

Kahn added two other characteristics as potential justifications for regulation: “natural monopoly” and markets where “for one or another of many possible reasons, competition does not work well.”²⁴ Although Kahn was skeptical of the monopoly rationale for regulation, he later argued that the nature and extent of competition is an empirical question: “It is a question also of what, in the circumstances of each regulated industry, is the proper definition, what are the prerequisites, of effective competition.”²⁵

B. Market Structure

The second rationale offered by Kahn is a market structure problem. Very large economies of scale mean that building multiple networks raises costs. The market will not support competition. In the extreme, we run into the problem of a natural monopoly. Firms that become too large behind high barriers to entry, with high transaction costs on the supply side or high switching costs or other behavioral flaws on the demand side, obtain market power. Monopolists (natural or otherwise) have market power and there is a strong incentive to abuse it. With the incentive and ability to exercise it, they engage in behaviors that harm competition (by creating additional obstacles to entry or extending their market power to complementary markets) and to harm consumers (by raising prices and restricting choices). Regulation controls market power. However, monopoly is not the only reason to implement public policy – e.g., it has never been a necessary condition to impose common carriage in the communications and transportation sectors.

C. Social Values

We turn next to Kahn’s third reason for regulation – “other.” Although it is less specific, it can be given several referents in the communications space. Competitive markets do not deliver universal service because there are significant parts of society where the rate of profit does not support extending infrastructure or making it affordable. Rural or high-cost areas and low-income populations may not be very attractive from an investment point of view, but they are important from public policy and social values points of view.

Freedom and diversity of opinion and voices are extremely important socio-political values that may not be accomplished by a competitive market. Society cannot leave them to the vagaries of the market. Speech is perhaps the most important example of these values,²⁶ diversity is too. Communication is well-recognized as a key to democracy and many consider it a human right.²⁷ The challenge is not simply to ensure that all have the opportunity to speak, but also to address gross imbalances in those opportunities.

The challenge of preventing the abuse of market power in communications and network industries was particularly severe, From the beginning of the deployment of the telephone in the earliest days of the second industrial revolution the problem of market power abuse had been addressed by both antitrust and regulation. By the early 1900s, the two pillars of oversight – the Sherman Act (by a consent decree in 1913) and the Interstate Commerce Act (by the Mann Elkins Act of 1910) had been extended to the telephone network. The animus behind the ICA and the target of regulation was the abuse of market power by another network industry, railroads. Access to important networks was addressed by antitrust in early Supreme Court cases (St. Louis Bridge, 1912, the AT&T Consent Decree, 1913), The strands of federal oversight over telecommunications were pulled together in separate agencies by the Communications Act of 1934 of Thus history of dual jurisdiction is long and rich, as shown in Table 2.

 

V. CORRECTING THE MISTAKES OF THE TRUMP ADMINISTRATION

After a period of dormancy under a Republican administration and uncertainty in the wake of the Telecommunications Act of 1996, the Obama administration had restored this dual jurisdiction. The Federal Communications Commission had re-established its authority to ensure just and reasonable rates (BDS, order issued and further notice put out), nondiscriminatory access and interconnection (network neutrality, wireless roaming, both upheld by Appeals Courts), as well as the promotion of universal service (Order upheld by 10th Circuit). These were the primary goals of the ’34 Act. The Department of Justice had reasserted control over mergers, rejecting some (Comcast/Time Warner) and imposing conditions on others (Comcast/NBC). In these merger cases, the FCC exercised its independent merger oversight by adopting parallel decisions that it would enforce.

The FCC’s actions rest on the basic principle that market power is so pervasive in this network industry that it cannot be met by antitrust or regulation alone. The abuse of market power on a daily basis strains antitrust laws, which hesitate to engage in behavioral regulation. The big competitive issues (mergers, divestiture) are generally beyond the reach of the regulatory agency.

After the Trump administration’s effort to essentially repeal the ’34 Act by administrative inaction and explicit effort to and abandon its authority to other agencies, the challenge will be to restore effective oversight and pull the strands of policy into a coherent overall approach. This effort will require a mix of regulatory and legislative actions including:

1. reversing the theory of “sufficient” competition and the deregulation to which it gave rise,
2. asserting full Title II authority over nondiscrimination and universal service,
3. bringing the full weight of Title II authority to bear on achieving universal service to broadband,
4. restoring full dual jurisdiction by repealing the Trinko decision that neuters antitrust even when there is only a whiff of regulation present.

Table 2: The long History of Dual Oversight in the Communications Sector²⁸

 

VI. CONCLUSION: THE IMPORTANCE OF TIGHT OLIGOPOLIES ON STEROIDS, BIG DATA PLATFORMS

There is one final reason to introduce the concept of a tight oligopoly on steroids.²⁹ Not only is it useful in describing the current market structure and harms in the traditional communication sector (regardless of the technology used, i.e. broadband), it is also useful in describing the current situation in Big Data Platforms, which requires an independent analysis. As shown in Table 3, the four conditions for steroids to amp up market power and to frustrate competitive entry are found in the Big Data Platforms. The manifestation of the traits and the policies necessary to prevent abuse and promote competition are somewhat different, but the magnification of market power and the need to have policy to prevent abuse and promote competition remains the same.

Table 3: Tight Oligopoly on Steroids: Broadband Networks and Data Platforms³⁰

The approach to Big Data platforms builds on the broader view of antitrust and regulation outlined in this paper, but the principles must be adapted to these new industries.

• While the principles on which they rest are the same, the practice must be adapted to the new techno-economic relationships in the economy.
• Antitrust and regulation need to be rebooted, after a long period in activity.
• Antitrust and regulation also need to be recalibrated to fit the new economy.
• Antitrust practice needs to be redefined to be better equipped for the challenges of the new economy.
• Regulation, which was designed and is well-suited for communications networks (Big Broadband Networks), needs to be redefined for the challenges of digital information (Big Data Platforms).
• A new regulatory agency is necessary because the existing sector specific expert agencies do not have portfolios and approaches (authority and power) to meet the task of overseeing the digital information sector.
• A new agency designed for the digital sector is important not only to provide oversight, but also to ensure flexibility and adaptability and promote competition.

In tight oligopolies with scale economies and barriers to entry, we need strong pro-competition regulation. Antitrust enforcers’ failure to block mergers exacerbated this situation greatly for telecom, but there were also an array of anti-competitive business practices that antitrust authorities could have challenged, but did not. The breakup of AT&T took place in a context where (was made necessary by) regulators had failed to prevent the abuse of vertical market power. For digital platforms, this historical example shows we both need a regulator and clearer antitrust guidelines, both of which must do the hard work that wasn’t done for BDS. It is exactly this coordination and backstopping with regulators doing the daily work, backed up by the threat of an antitrust action, that the Obama administration began to achieve. It needs to go back in that direction, but must be more systematic in developing oversight for digital data platforms that tend even more strongly to smaller numbers and larger market power.

---

1 Director of Research, and Antitrust Advocacy Associate, Consumer Federation of America.

2 Traditionally and more narrowly, these were called special access, or middle mile services. See, Mark Cooper, & Amina Abdu, Business data services: Another Failure Of Free Market Fundamentalism To Promote Competition Or Prevent Abuse Of Market Power, Consumer Federation of America, September 2020.

3 Mark Cooper, Overcharged and Underserved, Consumer Federation of America, December 2016.

4 Mark Cooper & Amina Abdu, Pragmatic, Progressive Capitalism at Its Best: Network Neutrality How an Entrepreneurial State Used Public Policy to Foster Experimental Entrepreneurialism and Create The Internet, Consumer Federation Of America, August, 2020.

5 Mark Cooper and Amina Abdu, Big Data Platforms A New Chokepoint in the Digital Communications Sector: Meeting New Challenges with Successful Progressive Principles, Consumer Federation of America, September 2020.

6 See note 2, pp. 2-8, 48-52.

7 C. Scott Hemphill & Tim Wu, “Parallel Exclusion,” Yale Law Journal, 122, 2013.

8 See note 2, p. 48.

9 Section 402.

10 See the definition of telecommunications in the ’96 Act.

11 The term was popularized in an Alan Greenspan in a 1996 speech, “The Challenge of Central Banking in a Democratic Society.” He attested to its deeper meaning in congressional testimony (October 22, 2008) after the financial meltdown a decade later. “Those of us who have looked to the self-interest of lending institutions to protect shareholder’s equity — myself especially — are in a state of shocked disbelief.”

12 This was true of both the BDS order (as explained in the citation in note 2 above) and the Network Neutrality Order (as explained in the citation in note 4 above).

13 Competing products must offer products that are close substitutes on quality and price.

14 In addition to the reference in note 2, a recent paper provides an extensive critique of the “ersatz” version of capitalism espoused in the theory of “sufficient” competition and the nearly identical theory of “perfect” contestability before it: Mark Cooper, From Brandeis to Stiglitz: Into & Beyond The 2020 Election: The Brandeis Protocol and the Stiglitz Model Create a Framework for Pragmatic, Progressive Capitalism, Consumer Federation of America, September 2020.

15 See source in note 2, chapter 7 for this data and review of the empirical literature that makes this point.

16 See note 2.

17 DOJ/FTC, 2010, p. 19.

18 Id. p. 19.

19 Federal Communications Commission, 2016 Final rule and FNPRM). Business Data Services in an Internet Protocol Environment et al., 31 FCC Rcd. 4723 (rel. May 2, 2016).

20 Federal Communications Commission, 2017 (Flip-Flop Order), Business Data Services in an Internet Protocol Environment et al., Report and Order, WC Docket Nos. 16-143 et al. (rel. Apr. 28).

21 Kahn, Alfred, 1988, The Economics of Regulation: Principles and Institutions, (Cambridge: MIT Press).

22 Kahn, 1988, p. 11.

23 Id.

24 Id.

25 Id. at 114.

26 Associated Press v. United States, 1945.

27 Cooper, 2013, 2014.

28 See note 4, p. 57.

29 In additional to the sector specific discussions above, a much earlier discussion of antitrust in high technology industries can be found in Mark Cooper, “Antitrust as Consumer Protection in the New Economy: Lessons from the Microsoft Case,” Hastings Law Journal, 52:4, 2001. Broader discussion of the role of antitrust with application to the communication sector can be found in Gene Kimmelman and Mark Cooper, “Antitrust and Economic Regulation: Essential and Complementary Tools to Maximize Consumer Welfare and Freedom of Expression in the Digital Age,” Harvard Law & Policy Review, Volume 9-2.

30 See note 5, p. 57.