The merger between T-Mobile and Sprint is now complete. After receiving clearance from the DOJ and the FCC, the last significant hurdle it had to clear was the court challenge by a group of states. It was a rare event in which states found themselves pitted against both the merging parties and the federal agencies that had cleared the merger with remedies. A number of issues related to the merger have already been dissected in the public discussion. In this paper, we highlight three significant issues that haven’t received as much attention. These include, the court’s exclusion of Mobile Virtual Network Operators (“MVNOs”) as participants in the relevant market, T-Mobile’s estimation of standalone marginal costs and merger efficiencies, and the somewhat odd remedy that required dismantling of a fully operating network and building of a brand new network to replace that dismantled network.

By Nitin Dua & Keith Waehrer1



The recently closed merger between T-Mobile and Sprint went through a long legal process in front of antitrust enforcers across the United States. At one point, the merger was simultaneously being investigated by the Federal Communications Commission (“FCC”), Department of Justice (“DOJ”), California Public Utilities Commission (“CPUC”), and numerous state Attorneys General (“AGs”), including AGs of New York, California, and Texas. Even though the FCC and the DOJ approved the merger, a group of 13 states and the District of Columbia, led by New York and California, sued in the Southern District of New York (“SDNY”) to block the merger.2 After a short two-week trial in December 2019, Judge Victor Marrero cleared the merger in February of 2020 opining that the merger is not “reasonably likely to substantially lessen competition…”3 The merger was widely reported on, but as two economists who worked on the case, here we describe three underappreciated aspects of the litigation.

Prior to the merger, T-Mobile and Sprint were the third and fourth-largest mobile wireless carriers in United States, with AT&T and Verizon as the two leading carriers. Together, the four firms accounted for over 95 percent of all wireless network access in the country.4 Sprint and T-Mobile had considered merging in the past, including in 2012 and 2014, but didn’t proceed due to likely push back from FCC and DOJ at the time.5 In 2019, the two companies tried again and this time found a sympathetic regulator in the FCC and a less combative enforcer in the DOJ.

In its published opinion, the FCC stated that, per a “static” analysis, an unconditional approval of the merger will create upward pricing pressure. However, the FCC also agreed with T-Mobile and Sprint that the merger would yield “quality and dynamic competitive benefits” that will offset the upward pricing pressure.6 The FCC approved the merger with several conditions, the most significant of which was the divestiture of Sprint’s pre-paid brand Boost Mobile.7 In contrast to the FCC’s published opinion, the DOJ’s public complaint was unequivocal regarding the likely anticompetitive effects from the merger between T-Mobile and Sprint. The complaint stated, among other things, that the reduction in the number of national facilities-based wireless carriers from four to three will reduce competition in the wireless market and harm wireless consumers. The complaint further stated that the merger could lead to collusion between the three remaining facilities-based carriers and that any efficiencies generated from the merger would be unlikely to offset the anticompetitive effects on consumers.8

In spite of its concerns, the DOJ cleared the merger with a remedy that went beyond the FCC’s approval terms. As part of this remedy DISH, a satellite television company with substantial holdings of unused wireless spectrum, was identified as the divestiture buyer of Sprint’s Boost Mobile brand. In addition to Sprint’s prepaid business, DISH would to receive a seven-year network access agreement with T-Mobile and certain infrastructural assets that would facilitate the building of its own wireless network. The DOJ’s remedy was designed to keep the number of national facilities-based carriers at four. However, for the first few years, DISH would be what is known as a mobile virtual network operator (“MVNO”) before it hopefully transitions into a full-fledged facilities-based carrier with its own wireless network. This transition period and MVNOs ability to compete with national facilities-based carriers were among the central issues in the case litigated in front of Judge Marrero.

As has been widely recognized, this was a rare situation because the merger litigation was led by a group of states without any support from the two federal agencies responsible for enforcing antitrust in United States (DOJ and FTC). In fact, the DOJ and the FCC had already essentially cleared the merger with remedies when the States filed their complaint in Federal Court to block the merger. The coverage of the trial and the reporting afterwards focused on a number of relevant issues, including Sprint’s ability to be an effective competitor given the strength of its network and financial position, DISH’s incentive to build out a network within a seven-year period required by the remedy, and T-Mobile’s incentive to remain a maverick after the merger. Antitrust practitioners also found interesting Judge Marrero’s decision to put relatively little weight on the testimony from economic experts.9 The judge stated in his opinion that the “incompatible visions of the competitive future” laid out by both sides’ experts “essentially cancel each other out.”10 These are important issues and each of them may deserve a detailed examination of its own. In this piece, however, as two individuals deeply involved in the litigation we focus on three aspects that didn’t get as much attention either during the trial or in post-trial examination of the arguments made in favor of and against the merger.



The DOJ’s remedy envisions DISH acquiring Sprint’s prepaid business but not the facilities needed to provide wireless service to its subscribers. Thus, until it builds its own network DISH will operate as an MVNO. DISH will not be unique in its status as an MVNO. MVNOs have been around for decades. MVNOs in the United States do not own the physical infrastructure required for a mobile wireless network.11 They instead lease access to a mobile network on a wholesale basis from the big four (now three) facilities-based wireless carriers and then resell mobile services to end customers, usually under prepaid brands.12 Most MVNOs help traditional mobile wireless carriers reach specific consumer segments, such as low-income consumers.

One of the disputes at the center of the trial was whether these MVNOs, including the largest MVNO Tracfone (with over 20 million subscribers)13 and large cable companies like Comcast and Charter that have their own MVNO offerings, are independent market participants that should be attributed shares for the purpose of determining whether the merger is presumptively anticompetitive based on HHI thresholds. Before the start of the trial Judge Marrero identified this as one of the main issues of the case.14 If the court agreed with Sprint and T-Mobile and considered MVNOs as independent competitors in a “retail mobile wireless telecommunication services market,” the market shares could have generated HHI levels below the threshold required to meet the presumption of substantial harm due to the merger under the Horizontal Merger Guidelines (HMG).15 Not meeting the presumption would have been seen as a serious blow to the states’ case against the merger because US courts have typically followed a burden-shifting framework.16 The very first step of this framework requires plaintiffs (states, in this case) to show that the merger is likely to have significant anticompetitive effects. Plaintiffs generally rely on the HMG at this step. The general standards laid out in the HMG include a presumption of significant anticompetitive harm from mergers that lead to further increases in concentration in an already highly concentrated market.17

Sprint and T-Mobile argued that MVNOs should be treated as market participants because, among other reasons, MVNOs account for a large number of subscribers and independently set their own prices.18 For calculating concentration statistics, this would require attribution of shares to each MVNO as opposed to attributing all MVNO subscribers to the wireless facilities-based carriers that provide network access. The approach of attributing MVNO subscribers to facilities-based carriers is consistent with what the FCC has done for years in its annual wireless competition reports, citing it as industry practice.19 However, T-Mobile and Sprint argued that existence of large MVNOs like Tracfone (operating as a reseller for years), and new large entrants like Google (with its Google fi MVNO product), Comcast, Altice, and Charter in the MVNO space increased the importance of MVNOs as competitors. Because large MVNOs, due to their size and presence in other related markets, could, in theory, bargain with the four major wireless facilities-based carriers to get network access at sufficiently low rates to make them competitively independent of the carriers that lease them their network capacity. Additionally, some of the MVNOs, like cable companies, could use their mobile products as loss leaders in an effort to gain more cable subscribers. However, the key issue from an antitrust standpoint was whether MVNOs could effectively act as competitive restraints on the facilities-based wireless carriers. The court had to settle whether MVNOs could be considered independent competitors to the four nationwide carriers. That is, could an MVNO be an effective competitor to the carrier it leased network capacity from?

One of the states’ economic experts, Professor Carl Shapiro, testified to the states’ case on this question. There are at least two main aspects of MVNOs that render them ineffective as competitive constraints on facilities-based carriers. First, retail sellers of mobile wireless services compete on network quality (e.g. network speed, network coverage) and invest billions of dollars in maintaining and upgrading their network to meet the data needs of their subscribers. MVNOs simply cannot compete on network quality because they do not own a network. This, in turn, limits their ability to steal subscribers from facilities-based carriers by increasing quality. Second, even if MVNOs were able to survive in the market as low margin competitors, given their low margins, they have limited ability to adjust prices in order to steal business from facilities-based carriers. Importantly, facilities-based carriers do not have much incentive to compete with MVNOs either. This is because of the large margin that facilities-based carriers earn on MVNO subscribers.

The evidence presented at the trial showed that Tracfone, the largest MVNO in the market, had a one percent profit margin, after covering marginal costs, including the cost of network access. T-Mobile, one of the four nationwide carriers, in comparison, had about a 50 percent profit margin, after covering marginal costs. Tracfone’s low margins are explained by the high fees it pays for network access. Operating at low margin and high costs simply means that Tracfone and other MVNOs (with similar or likely even smaller margins) have very limited ability to compete for new customers on the basis of prices.

The evidence presented at the trial further showed that on every new customer Tracfone pays, in addition to subscriber acquisition costs, nearly $15 per month in network access costs. Tracfone’s MVNO products have an average monthly price of about $25. At a one percent margin, Tracfone earns approximately 30 cents on every new subscriber. In contrast, on an average MVNO subscriber, T-Mobile earns around $11 from what it charges MVNOs like Tracfone for network access. This ability of the facilities-based carriers to extract most of the value from a new MVNO subscriber clearly shows that MVNOs are best considered as partners instead of competitors to the four nationwide facilities-based carriers. Indeed, the sentiment is reflected in Verizon’s recent statement announcing its intention to acquire TracFone in a $6 billion deal.20 Verizon explained in its public announcement that the acquisition of a “longtime partner” (i.e. TracFone) expands Verizon’s portfolio “into the value segment.”21 MVNOs’ lack of independence does not mean that they do not compete with the facilities-based carriers at all. Rather, it means that facilities-based carriers own or control the important levers of competition and can easily limit MVNOs from acting as a competitive constraint. This was echoed in the judge’s opinion. He stated that, “the mere fact that a firm may be termed a competitor in the overall marketplace does not necessarily require that it be included in the relevant product market for antitrust purposes.”22

The analysis underlying the court’s decision to not consider MVNOs as independent competitors is generalizable and can be effectively used for studying competition in other industries where resellers and manufacturers sell products alongside each other in a retail market. To understand if a reseller is an independent competitor to the manufacturer, one may ask if the reseller has the wherewithal to independently compete on price and quality. If the inputs supplied by the manufacturers account for a significant percentage of the reseller’s production cost and there are few available substitutes to those inputs, it is difficult for the reseller to meaningfully act as a competitive constraint on the manufacturer.



Early in their advocacy for getting the merger through without divestitures, Sprint/T-Mobile presented an argument that the efficiencies from the merger outweighed the anticompetitive effects. The Horizontal Merger Guidelines place a high bar on efficiency evidence.23 While the FCC was sympathetic, the DOJ did not accept these arguments and required parties to enable DISH’s entry in the retail mobile wireless market.24 Efficiency arguments were a large part of the testimony of Sprint/T-Mobile’s economic expert, Professor Michael Katz. Most of the arguments in court about the efficiencies claimed by Sprint/T-Mobile related to the sensitivity of the marginal cost reduction estimates to assumptions relating to the availability of spectrum to the standalone firms and the network speed requirements of mobile wireless users. A little discussed aspect of the models that Sprint/T-Mobile used to estimate cost savings is that a significant fraction of the predicted efficiencies were the result of network cost increases that the standalone Sprint and T-Mobile would face in the period immediately after the planned closing date of the merger, as opposed to the merged firm enjoying substantially lower network costs than the pre-merger standalone firms.

That is, the model used by Sprint and T-Mobile predicted large marginal cost increases for the standalone firms post-merger rather than a large marginal cost decrease for the New T-Mobile. This was notable because it implied that without the merger there would be a significant increase in marginal costs that there had been no indication of in the pre-merger period. This increase in marginal costs however has a somewhat contradictory source as it requires a large increase in the demand for speed by consumers, which would seem unlikely to arise unless speed became less expensive and not more expensive.

The core of Sprint/T-Mobile’s intuitive explanation for its efficiency claims was that the combination would be multiplicative as opposed to additive in terms of wireless network capacity. T-Mobile’s President of Technology, Neville Ray, explained in public filings that “[t]he combination of the two companies does not simply double the network capabilities, but instead provides a multiplicative effect for the overall capacity of the New T-Mobile network.”25 The effect of the merger on network capacity is central to the issue of merger efficiencies relevant to the antitrust review. More potential capacity implies lower marginal costs.

T-Mobile’s public filings to the FCC explained how network capacity is calculated with the following simple formula:

Network capacity = (# cell sites) × (MHz per site) × (Spectral efficiency).26

In words, network capacity is equal to the number of cell sites (i.e., cell towers) times the amount of spectrum deployed at those sites times a factor relating to the efficiency of the communications standard being used. Sprint/T-Mobile did not claim that there would be an improvement in spectral efficiency as a result of the merger, but that the merger would have a multiplicative effect on network capacity because more spectrum would be deployed on more cell sites.

According to this formula, the network capacity of the New T-Mobile would be:

New T-Mobile capacity

= [(# cell sites)S + (# cell sites)T ] × [(MHz per site)S + (MHz per site)T ]

× (Spectral efficiency),

Subscripts S and T in this formula stand for Sprint and T-Mobile. Simple algebra using the two formulas above shows that the increase in capacity due to the combination of the two networks can be written as:

∆Capacity = [(# cell sites)S × (MHz per site)T + (# cell sites)T × (MHz per site)S ]

×(Spectral efficiency).

Depending on the magnitude of the inputs, this formulation has the potential to imply a very large increase in capacity relative to the sum of the capacities of standalone Sprint and T-Mobile. For example, for a hypothetical merger of two symmetric networks, the formula implies that the merged firm’s network capacity would be double of what could be achieved by just summing up the capacity of standalone firms.

While the simple formula provides a good approximation of the technological explanation for potential efficiencies, it does not account for a number of real-world complications that are important for a detailed quantification of efficiency claims in the context of a merger review. For example, to achieve the efficiencies implied by the formula above, New T-Mobile would need to make significant investments in deploying new radios on the cell towers it intends to keep in service. This leads one to ask if the merged firm would have the incentive to make such an investment. Likely recognizing this, Sprint/T-Mobile did not rely on this formula for quantification of efficiency claims. Instead, T-Mobile created a complicated model that had its origins in an ordinary course network engineering tool that helped T-Mobile predict cell site congestion.

Putting aside plaintiff arguments that the efficiencies associated with deploying Sprint’s spectrum on T-Mobile’s cell sites is not merger specific and thus should not be considered as an offset to the anticompetitive effects of the merger, there was yet another issue with Sprint/T-Mobile’s efficiency claims. The New T-Mobile planned to keep only 30 percent of Sprint’s former cell sites.27 This fact undercut the efficiency logic of the simple formula, as it has the effect of reducing the potential capacity gains. In the illustrative example of symmetric networks described above, if only 30 percent of the cell sites from one of the merging companies are kept, capacity increases are greatly reduced. Instead of a hypothetical 100 percent increase in capacity, the simple formula would imply only a 30 percent increase in capacity. While still potentially large, not quite the multiplicative effect that was claimed publicly.

A quantitative estimate of the effects of the claimed efficiencies needs to be compared with an estimate of the anticompetitive effects arising from the merger. While the simple formula implies significant potential efficiencies, measures of unilateral effects presented in the court implied that the actual efficiencies would need to be extraordinarily large to reverse the predicted price increase from the merger. For example, Professor Carl Shapiro testified that in order to reverse the price effect on T-Mobile’s post-paid customers, the New T-Mobile’s marginal costs would need to decrease by 90 percent.28 The supersized cost reductions needed to reverse the post-merger unilateral incentive to raise price were due to the relatively high diversions between Sprint and T-Mobile and the relatively high margins that facilities-based wireless carriers enjoy. Even with the multiplicative effect described in the Sprint/T-Mobile filings, it is a steep uphill climb to establish efficiencies large enough to overcome the unilateral price effects predicted from the merger.

To quantify marginal cost savings, Sprint/T-Mobile relied on a network congestion model that predicted capacity expansion needed in response to predicted data demand increases for each of the three networks (standalone Sprint, standalone T-Mobile, and the New T-Mobile). The marginal cost was derived from the cost of adding the capacity needed for the increased data use associated with an increase in the number of subscribers. As described in the testimony of Profession Fiona Scott Morton, one of the plaintiff’s economic experts, most of the difference between the predicted cost of the New T-Mobile and predicted costs of the standalone Sprint and T-Mobile were the result of a large marginal cost increase for the standalone firms that the congestion model conveniently predicted would occur soon after the closing of the merger. According to a demonstrative accompanying Professor Scott Morton’s testimony, the congestion model predicted a threefold increase in standalone Sprint’s costs from under $1 per subscriber per month in 2020 to almost $3 per subscriber per month in 2022. The model also predicted a sudden twofold increase in marginal costs for the standalone T-Mobile, from a little over $3 in 2020 to over $6 in 2022.29 Such large cost increases seem unlikely to have occurred in the past as wireless plans prices have been mostly stable or decreasing.30

The marginal cost increases for the standalone firms, as predicted by the model, occur during the 5G implementation period (i.e. 2020 and beyond). While the facilities-based carriers will need to make significant investments to upgrade their networks as part of the transition to 5G, these costs are different from marginal costs that are key to the efficiency claims. Among the technological benefits of 5G are increased spectral efficiency and an increase in the wave lengths of usable spectrum, both of which have the effect of expanding network capacity as defined in the simple formula that Sprint/T-Mobile claimed as support for their efficiency argument. Holding other factors constant, these technological advancements should lower the marginal costs of the standalone firms.

The key assumption underlying Sprint/T-Mobile’s efficiency quantification was that mobile wireless users would require significantly higher data speeds with the introduction of 5G. Professor Scott Morton testified that these assumed data speeds were much greater than those required by applications currently used on mobile phones, such as streaming high-definition video.31 Significantly higher speeds predicted by Sprint/T-Mobile would only be used by mobile consumers if there was widespread adoption of new uses such as applications involving augmented or virtual reality in environments where Wifi access is not available. Putting aside the likelihood that Wifi would remain the preferred source of high-speed data applications, one would only expect such an adoption on mobile wireless networks if the cost of achieving the required faster speeds decreased, so it seems inconsistent to simultaneously predict an increase in marginal costs per subscriber (without the merger) while at the same time assume a large increase in the demand for network speed that would likely only occur if the cost of that speed were decreasing.32



According to DOJ’s complaint and other filings, but-for the remedy, the merger was anticompetitive and thus would have been challenged.33 The remedy required T-Mobile and Sprint to divest selected assets to DISH and bound DISH to commitments, enforced with financial penalties, to follow through with the construction of a new wireless network.34 Thus, the remedy envisions recreation of a fourth independent facilities-based wireless competitor. The competitive impact statement published by the DOJ indicates that creation of a new fourth competitor to replace Sprint is critical.35 DISH was seen as uniquely situated for this role given its large unused spectrum holdings that could be used in the creation of this new network.36

The remedy negotiated by the DOJ requires the divestiture of all of Sprint’s prepaid business and subscribers to DISH.37 In addition, DISH has been given the option to acquire at least 400 retail locations that the New T-Mobile decided not to retain. However, the asset package divested to DISH is missing the network infrastructure required to provide service to DISH’s newly acquired subscribers, which is the key asset needed to operate as an independent facilities-based competitor. To solve this issue, DOJ’s remedy creates incentives for DISH to build a new network of its own. To help facilitate the building of a new network, the divestiture package includes options for DISH to acquire 20,000 decommissioned Sprint cell sites and certain spectrum assets.38 Given that building a network from scratch takes years and DISH will need to provide service to its acquired subscribers beginning immediately, for a period of up to seven years, the New T-Mobile will provide the required network service through a MVNO agreement with DISH.39

Critical commentary related to the remedy has focused on it behavioral aspects, DOJ’s reliance on the somewhat uncertain prospect that DISH will follow through with its commitment to build its own network, and the reduced coverage of the planned DISH network relative to the coverage of the pre-merger Sprint network.40 While DOJ refers to the remedy as structural and DISH as an independent fourth competitor, the remedy involves DISH competing as an MVNO for the first few years operating on networks owned and operated by the New T-Mobile. DOJ and the FCC have generally not considered MVNOs as independent competitors,41 and as discussed above the trial court here agreed.42 As such the remedy does not appear to meet DOJ’s own stated requirement that a structural divestiture include all of the assets needed to compete effectively and independently.43

However, here we seek to point out a different aspect of the remedy that has gotten less attention. Recognizing the presence of a fourth facilities-based carrier as critical to replace the competition lost as a result of the merger, DOJ’s remedy allowed for the complete dismantling of a working network and at the same time incentivized DISH to build a new network from scratch.

The New T-Mobile’s plan was to upgrade the old T-Mobile network in order to have the capacity to move Sprint subscribers onto that network and eventually discontinue the use of the old Sprint network. That means the physical infrastructure of the Sprint network would eventually be unused and taken down. The acquired assets that the New T-Mobile intends to use includes, perhaps most importantly, Sprint’s spectrum. After the transition to the new network the New T-Mobile will have also kept 30 percent of Sprint’s towers and of course the acquired subscribers that it is able to retain.44 However, the pre-merger Sprint assets that the New T-Mobile planned to use are not unique in the sense that additional spectrum and towers are available elsewhere, including from DISH itself.

From a purely economic welfare perspective, it seems rather odd, and inefficient, to allow a working network with a planned path to 5G to be torn down at the same time as requiring resources to be expended to build a new network that replaces the torn down network. Nowhere in the analysis of the remedy was there an accounting of the welfare costs associated with replacing the Sprint network.

1 Nitin Dua is a Principal at Bates White, LLC; Keith Waehrer is a Principal at Economists Incorporated. The views expressed in this paper are solely those of the authors and do not necessarily reflect the opinions of Bates White, Economists Incorporated, or their clients.

2 See

3 Judge Marrero Opinion, p. 167–168;

4 Fig. A-3, Communications Marketplace Report (FCC-18-181A1; Dec 26, 2018).

5 Judge Marrero Opinion, p. 28;

6 Memorandum Opinion and Order, Declaratory Ruling, and Order of Proposed Modification, p. 5, para. 9–11.

7 Boost would have operated as an MVNO with superior wholesale network access terms than a typical MVNO.

8 See DOJ Complaint, “United States et al v. Deutsche Telekom AG; T-Mobile US, Inc.; SoftBank Group Corp.; and Sprint Corp.”

9 See

10 Judge Marrero Opinion, p. 4.

11 Mobile wireless network includes radio access network, core network, and backhaul connections that connect the cells in the radio access network to the core. MVNOs do not operate radio access networks, which are critical for the provision of mobile wireless telecommunication services. Most MVNOs do not own any facilities—i.e., radio access network, core network, or backhaul.

12 See FCC Annual Report: Federal Communications Commission, “Consolidated Communications Marketplace Report – 2018” (Report FCC 18-181, Dec. 26, 2018), at ¶¶7, 16.

13 Memorandum Opinion and Order, Declaratory Ruling, and Order of Proposed Modification, p. 128, footnote 191.

14 Court transcript, December 6.

15 U.S. DOJ and Federal Trade Commission, Horizontal Merger Guidelines, August 19, 2010. “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.” According to the guidelines, markets with HHI above 2,500 are considered highly concentrated.

16 Antitrust, Vol. 33, No. 2, Spring 2019. “The Four-Step Rule of Reason.”

17 Once the plaintiff has met their burden, the burden shifts to the defendants (Sprint and T-Mobile, here) who have to show that there are aspects of the merger that would offset the anticompetitive effects (such as, efficiencies from the merger). And, finally, the burden moves back to the plaintiffs who have to successfully rebut the defendant’s case and show that there are reasonable alternatives to a merger that would help the defendant achieve its procompetitive objectives.

18 FCC opinion, p. 32, para. 75. Judge Marrero Opinion, p. 37.

19 FCC footnote 90, 20th competition report, “Following widespread industry practices, the Commission generally attributes the subscribers of MVNOs to their host facilities-based service providers, including when it calculates market concentration metrics.” DISH, a new MVNO as a result of the merger, also stated in 2018 (in a petition to deny the Sprint/T-Mobile merger) that “MVNOs are likely not effective competitors to facilities based carriers in light of these operators’ dependence on their landlord carriers’ consent…” Petition to Deny of DISH Network Corporation at p. 6, In re Applications of T-Mobile US, Inc. and Sprint Corporation, No. 18-197 (F.C.C. Aug. 27, 2018), ¶103.

20 See

21 See

22 Judge Marrero Opinion, p. 37.

23 HMG, §10.

24 FCC Memorandum Opinion, p. 69, para. 157: “While we are confident that the transaction will lead to significant marginal cost savings, developing a precise estimate of those savings is a difficult and inherently uncertain task.” DOJ Complaint p. 8, para. 24: “Any efficiencies generated by this merger are unlikely to be sufficient to offset the likely anticompetitive effects on American consumers in the retail mobile wireless service market, particularly in the short term, unless additional relief is granted.”

25 DECLARATION OF NEVILLE R. RAY, p. 13. (Page 177 of the public interest statement and appendices).

26 Ibid.

27 DECLARATION OF NEVILLE R. RAY, June 2018 states that New T-Mobile would keep 11,000 of Sprint sites. Dish’s petition to deny of dish network corporation, “…T-Mobile’s AWS-3 spectrum will be deployed on all retained Sprint sites (11,000) (out of Sprint’s existing 46,000 sites).”

28 Trial Transcript, p. 708.

29 Trial demonstratives of Prof. Fiona Scott Morton, p. 20.

30 Trial transcript, p. 684.

31 4K video is not yet widely used on mobile handsets and even it requires significantly lower speeds than Sprint/T-Mobile’s efficiency quantification assumed would be required in the near future. See Trial transcript, p. 2210.

32 A reasonable questions to ask here is whether such an increase in demand might be driven by AT&T and Verizon’s low costs for high speeds. However, given the spectrum holdings of the 4 national facilities-based carriers pre-merger, it would seem that Sprint was better positioned than any to expand capacity at the lowest cost. Sprint by far had the most spectrum on a per subscriber basis than any of the other carriers. See

33 See DOJ complaint, United States et al v. Deutsche Telekom AG; T-Mobile US, Inc.; SoftBank Group Corp.; and Sprint Corp.

34; DOJ Stipulation and Order, United States et al v. Deutsche Telekom AG; T-Mobile US, Inc.; SoftBank Group Corp.; and Sprint Corp.

35 DOJ Competitive Impact Statement, United States et al v. Deutsche Telekom AG; T-Mobile US, Inc.; SoftBank Group Corp.; and Sprint Corp. “The elimination of a fourth national facilities-based mobile wireless carrier would remove competition from Sprint and restructure the retail mobile wireless service market.” “Increasing DISH’s incentives to complete the buildout of a fourth nationwide wireless network also serves to decrease the likelihood of coordinated effects that arise out of the merger.” Without the relief provided in the proposed Final Judgment, neither entry nor expansion is likely to occur in a timely manner or on a scale sufficient to replace the competitive influence now exerted on the market by Sprint.”

36 “Under the agreement’s terms, Mr. Delrahim said, Dish is in a unique position to succeed.”

37 In addition to the terms intended to make DISH a new forth competitor, DOJ’s remedy include a number of “behavioral” components. It includes such commitments in the form of merging parties’ commitment to abide by the terms of their existing MVNO agreements and to extend such agreements on their existing terms, subject to certain limitations. The remedy also requires New T-Mobile and DISH to provide support for eSIM technology and requires them to adopt certain policies and procedures regarding the “unlocking” of handsets. See Proposed Final Judgment at §§VII.A, VII.B– VII.F, United States et al. v. Deutsche Telekom AG, et al, No. 1:19-cv-02232 (D.D.C. July 26, 2019), available at

38 Department of Justice, “Justice Department Settles with T-Mobile and Sprint in Their Proposed Merger by Requiring a Package of Divestitures to Dish,” news release, July 26, 2019, The Proposed Final Judgment (PFJ) requires the merging parties to divest Sprint’s prepaid businesses, including the Boost Mobile, Sprint-branded prepaid and Virgin Mobile businesses to DISH. The PFJ also requires New T-Mobile to divest certain 800 MHz spectrum licenses to DISH. DISH can elect not to acquire the 800 MHz spectrum if it has met certain network deployment milestones or in exchange for making a payment to the United States. See Proposed Final Judgment at V.B.2., United States et al. v. Deutsche Telekom AG, et al, No. 1:19-cv-02232 (D.D.C. July 26, 2019), available at

39 U.S. Department of Justice, U.S. and Plaintiff States v. Deutsche Telekom AG, et al., Proposed Final Judgment,, §§IV.A., VI.

40 For example, see


42 Judge Marrero Opinion, p. 38.

43 DOJ Merger Remedies Manual, Page 6, Section III.A, “Any divestiture must include the assets necessary to ensure the efficient current and future production and distribution of the relevant product or service and thereby preserve the competition that would have been lost as a result of the merger. A structural remedy requires a clear identification of the assets a competitor needs to compete effectively in a timely fashion and over the long term.” DOJ Merger Remedies Manual, Page 21, Section III.F, “Ongoing entanglements between the merged firm and the purchaser may put the purchaser in the position of having to rely on its rival in order to compete, and therefore call into question the purchaser’s position as a truly independent competitor.”

44 DECLARATION OF NEVILLE R. RAY, June 2018 states that New T-Mobile would keep 11,000 of Sprint sites. Dish’s petition to deny of dish network corporation, “…T-Mobile’s AWS-3 spectrum will be deployed on all retained Sprint sites (11,000) (out of Sprint’s existing 46,000 sites).”