1-2.7: Confronting the FCC Net Neutrality Order with European Regulatory Principles

1. Introduction
   

1. The FCC’s 2010 “open Internet” (net neutrality) order went into effect on November 20, 2011[1]. Verizon Communications and Metro PCS have challenged the FCC Order in court, alleging that the FCC lacked statutory power to impose net neutrality rules in connection with Internet access services[2]. Their principal argument is that the FCC does not have authority under the Communications Act to impose these regulations – Congress has not empowered the agency with authority over the Internet. In addition, the FCC lacked any showing of an existing harm requiring regulation at this point.[3]. Opponents of the FCC Order argue that while the FCC has considerable powers to adopt regulations in connection with telecommunications services under Title II of the Act, its powers are considerably more limited in connection with information services governed by Title I of the Communications Act, which the FCC relied upon (as well as other non-Title II sections) to establish its authority[4]. A net neutrality advocacy group Free Press has also challenged the FCC Order on the ground that the distinction between fixed and wireless operators (the FCC provided more lenient regulation of wireless on the ground that it is bandwidth constrained) is not justified[5]. 
   
2. The purpose of this article is not to examine these US court challenges to the FCC Order, but rather to compare the FCC Order to the net neutrality provisions contained in the 2009 revisions to the European directives on electronic communications[6]. This comparison will bring to light key similarities and differences between the two sets of rules. After comparing the two regimes, the article will apply European regulatory standards to the FCC Order: would the FCC Order withstand scrutiny if it were challenged in Europe? Submitting the FCC Order to a European “stress test” will yield several benefits. 
   
3. First, it will help readers better understand how net neutrality fits into the bigger picture of the European electronic communications framework, and in particular into the competition law principles that underpin most regulatory action in Europe. Second, the exercise will reveal potential flaws in the FCC’s reasoning, flaws that might be fatal to the FCC Order if it were subject to court review in Europe. Lastly, the exercise will help illuminate what flexibility European Member States and their national regulatory authorities (“NRAs”) have in connection with net neutrality rules. 
4. The Netherlands already adopted[7] national rules on net neutrality that go beyond the baseline measures provided for in the European directives. It is not beyond reason to anticipate that one or more national regulatory authorities in Europe might adopt measures that resemble the FCC Order. Our analysis will therefore help to determine whether national measures resembling the FCC Order would be permitted under the existing European framework, or whether they would be deemed incompatible with the European directives on electronic communications. The basic conclusion is that the European analysis starts with broad principles, including technological neutrality, specific proof of market failure before regulating, and the absence of a private common carrier regime that in the U.S. views non-discrimination as its hallmark. The result is more than a semantic difference. Instead, the starting place for analysis in Europe might well lead the U.S. regime to fail a test of legality, even if the day-to-day outcomes might not be so very different.

2. A comparison of the U.S. and European rules on net neutrality.

5. We will first examine the events leading up to the adoption of the FCC Order (A), before examining the content of the order itself (B). We will then examine the background to the European net neutrality measures (C), the content of those European measures (D), and finally summarize the differences between the two sets of measures (E).

1. The Genesis of the FCC Order

6. The genesis of the FCC’s order goes back to a 2005 complaint alleging port blocking by a small regional telephone company[8] of voice over IP services provided by Vonage. The regional telco did not want to lose its customers to Vonage’s “over-the-top” VOIP service. After quickly sanctioning the telco for blocking service, the FCC then issued in August 2005 a non-enforceable policy statement[9] describing what it referred to as the “four Internet freedoms”:

“To encourage broadband deployment and preserve and promote the open and interconnected nature of the public Internet[10],” consumers are entitled to:

• access the lawful Internet content of their choice.
• run applications and use services of their choice, subject to the needs of law enforcement.
• connect their choice of legal devices that do not harm the network.
• competition among network providers, application and service providers, and content providers.

7. No incidents were reported until 2008, The FCC next acted on a complaint alleging that Comcast was wrongfully blocking BitTorrent applications and misled customers as to why a session had been interrupted. The FCC found the behavior unlawful, and particularly criticized Comcast for its lack of transparency in the matter[11]. The FCC’s Comcast decision explained the notion of “reasonable network management” and the proportionality test that the FCC would apply to determine whether a given instance of network management is reasonable or not. The FCC also explained the principle of transparency that would apply to any action by a network operator to block or shape Internet traffic. 
8. Comcast challenged the FCC’s decision in court, and the U.S. Court of Appeals for the District of Columbia annulled the FCC’s decision on April 6, 2010[12]. The court did not reach the merits of the FCC’s decision, but found that the FCC lacked statutory authority to issue the kind of order it had issued against Comcast. In spite of the court reversal of its Comcast decision, the FCC pursued a rulemaking procedure to develop comprehensive rules on net neutrality. The FCC adopted its final rules on December 21, 2010 after an extensive public consultation. The rules came into effect on November 20, 2011 after their publication in the Federal Register on September 23, 2011[13]. 
9. In parallel to the creation of the new rules, the FCC imposed net neutrality obligations on a temporary basis on AT&T as one of its obligations in connection with its merger with SBC[14]. When licensing the valuable spectrum in the 700Mhz band, the FCC also set aside a separate block, called “Block C,” that would be subject to wireless net neutrality rules[15]. The other blocks of spectrum were not burdened by this obligation. Verizon Wireless purchased the Block C spectrum at auction, and is currently deploying a network to provide 3G services while respecting the wireless net neutrality rules contained in the Block C spectrum license. Finally, the FCC imposed some net neutrality obligations on Comcast in connection with Comcast merger with NBC Universal[16].

2. The content of the FCC Order.

10. The FCC Order imposes three different obligations on U.S. network operators. All of the obligations are applicable to fixed operators and only some of them are applicable to mobile operators. The first obligation is to be transparent vis-à-vis customers: operators must disclose to customers in a clear and comprehensive manner the network management practices they apply.

A person engaged in the provision of broadband Internet access service shall publicly disclose accurate information regarding the network management practices, performance, and commercial terms of its broadband Internet access services sufficient for consumers to make informed choices regarding use of such services and for content, application, service, and device providers to develop, market, and maintain Internet offerings.[17]

This transparency rule applies to both fixed and mobile operators in the U.S. The FCC has recently published guidelines on how these transparency rules should be applied[18]. 

11. The second obligation is called the no-blocking rule. The no-blocking rule comes in two different flavors: The no-blocking rule applicable to fixed operators prohibits the blocking of any lawful website, application, service or device, subject to reasonable network management. The no-blocking rule applicable to mobile operators prohibits only the blocking of any lawful website or the blocking of any service or application that competes with a service or application provided by the mobile operator. Other types of blocking, i.e., downloads of large files, may be allowed by mobile operators, As for fixed operators, the no-blocking rule applicable to mobile operators is subject to reasonable network management. In other words, blocking that occurs because of “reasonable network management” will not be deemed a violation of the FCC Order. The no-blocking rule is formulated as follows:

A person engaged in the provision of fixed broadband Internet access service, insofar as such person is so engaged, shall not block lawful content, applications, services, or non-harmful devices, subject to reasonable network management.

A person engaged in the provision of mobile broadband Internet access service, insofar as such person is so engaged, shall not block consumers from accessing lawful websites, subject to reasonable network management; nor shall such person block applications that compete with the provider’s voice or video telephony services, subject to reasonable network management.[19]

12. The third obligation contained in the FCC’s Order is the rule prohibiting unreasonable discrimination. This rule applies only to fixed operators. In other words, mobile operators are free to partake in “unreasonable discrimination,” whatever that term means. The rule prohibiting unreasonable discrimination is also subject to reasonable network management: discrimination dictated by reasonable network management will not violate the FCC’s rules. We will discuss the FCC’s no unreasonable discrimination rule in more detail below[20]. It is by far the most controversial aspect of the FCC’s order insofar as it is designed to prohibit paid prioritization arrangements between an Internet access provider and upstream content, application or service providers. The rule is worded as follows:

A person engaged in the provision of fixed broadband Internet access service, insofar as such person is so engaged, shall not unreasonably discriminate in transmitting lawful Internet traffic over a consumer’s broadband Internet access service. Reasonable network management shall not constitute unreasonable discrimination.[21]

13.  Under the FCC Order, the notion of reasonable network management is similar to the rules developed by the FCC in the Comcast decision. To qualify as reasonable network management, a measure must target a legitimate objective such as fighting spam or computer viruses or undue congestion of the network. The measure must then be narrowly targeted to achieve the identified objective and not create unnecessary spill-over effects. In essence, the FCC will apply a proportionality test to determine if network management measures are reasonable. The FCC described the proportionality test as follows:

A network management practice is reasonable if it is appropriate and tailored to achieving a legitimate network management purpose, taking into account the particular network architecture and technology of the broadband Internet access service.[22]

14. Finally, the FCC’s rule indicates that the net neutrality measures will not apply to so-called specialized services, i.e. services that do not consist of broadband Internet access. Examples of specialized services include managed IPTV, such as AT&T’s U-Verse service and managed online health services.

3. Background to the European Net Neutrality Measures

15. The 2009 amendments to the European directives on electronic communications provide national regulatory authorities (“NRAs”) with several tools to deal with net neutrality. European directives do not have direct effect. They do not immediately apply to companies and individuals in each Member State. Instead, each Member State must first adopt a national law that implements the directive. This gives each Member State flexibility to interpret the provisions of the directive and in some cases to adopt laws that go farther than what the directives provide. Member States were supposed to adopt national laws to implement the revised directives no later than May 25, 2011. However, a number of Member States have still not completely implemented the provisions of the revised directives[23]. 
16. Once a Member State enacts legislation to implement the directives, the Member State’s NRA has the responsibility for developing detailed rules on issues such as net neutrality. Those rules would then be subject to judicial review before national courts. When evaluating the legality of the measure taken by the NRA, the national court will review the measure in light of the directive’s provisions and the general regulatory principles underpinning the European framework for electronic communications. This review process is becoming quite routine for NRA decisions, which is why it is fairly easy to imagine the hypothetical case of a European national court reviewing a net neutrality measure adopted by an NRA in Europe, and which is similar to the FCC Order.

4. Content of the European Net Neutrality Rules

17. The 2009 revisions to the directives impose transparency obligations similar to those imposed by the FCC Order. All electronic communications operators in Europe must disclose in a clear and comprehensive manner the limitations imposed on customers in connection with accessing certain content, services or applications. The existence of this transparency provision suggests that Internet access providers are permitted to block or limit access to certain content, services and applications, provided those limitations are clearly disclosed to customers in advance. The Directive uses the following language:

This Directive neither mandates nor prohibits conditions, imposed by providers of publicly available electronic communications and services, limiting end-users’ access to, and/or use of, services and applications, where allowed under national law and in conformity with Community law, but lays down an obligation to provide information regarding such conditions.[24]

18. This contrasts with the FCC’s rule that prohibits, for fixed operators at least, any blocking measures unless they are necessitated by reasonable network management. 
19. The second branch of the transparency rule in Europe consists in the obligation to clearly disclose to customers the nature of the traffic management measures applied by the Internet access provider. 

Member States shall ensure that national regulatory authorities are able to oblige undertakings providing public electronic communications networks and/or publicly available electronic communications services to inter alia:

….

(d) provide information on any procedures put in place by the provider to measure and shape traffic so as to avoid filling or overfilling a network link, and on how those procedures could impact on service quality.[25]

The body of European regulators of electronic communications, the BEREC, has already issued guidelines on how operators should apply the European transparency rule[26].

20. The second measure in the European package is not an obligation per se, but a declaration of principle. The 2009 amendments to the directives include a new principle that NRAs must strive to achieve: in their decisions, national regulatory authorities must “promote the ability of end-users to access and distribute information or run applications and services of their choice.”[27] By framing this statement as a principle instead as a hard and fast obligation, the European framework gives NRAs flexibility to adapt the principle to different circumstances. Moreover this net neutrality principle is in competition with other equally important principles in the framework, such as ensuring the promotion of competition, efficient investment, and innovation[28]. This allows NRAs to apply a balancing test and make trade-offs when certain principles are in competition with each other, as is often the case.
21. The next measure contained in the revised European framework is the power for NRAs to fix minimum quality of service requirements for Internet access services. NRAs may only do this if they can show that the competitive market is not providing users with sufficient choice and quality levels. NRAs must communicate their proposal and its justification to the European Commission, which may provide comments or object. The reason for this provision is to allow NRAs to intervene in the event Internet access providers all begin promoting managed services – which, like “specialized services” are not subject to net neutrality rules — and gradually degrade the quality of basic Internet access. 
22. European policymakers fear that managed services such as IPTV will be more remunerative for access providers than basic Internet access, and that consequently access providers will gradually try to push their customers toward these managed services and that as a consequence the quality of broadband Internet access will decline. 

23. The European approach starts, however, from the principle that robust competition in the broadband access market will be sufficient to provide consumers with sufficient choice and quality in services, and that regulatory intervention to impose quality of service should occur only if there is a demonstrable market failure. This is well summarized in recital 34 to the Consumer Rights Directive:

A competitive market should ensure that end-users enjoy the quality of service they require, but in particular cases it may be necessary to ensure that public communications networks attain minimum quality levels so as to prevent degradation of services, the blocking of access and the slowing of traffic over networks.[29]

24. The last net neutrality measure contained in the new European package provides that NRAs should be in a position to arbitrate disputes between network operators and providers of content, applications and services. The wording of the revised directive on this subject is relatively unclear. The directive refers to disputes between network operators and enterprises that benefit from interconnection[30]. Because of this imprecise wording, national measures adopted to implement the directive may take divergent approaches. However, the idea behind the provision is to ensure that NRAs are able to intervene in dispute resolution proceedings in cases where a content or service provider upstream may accuse an Internet access provider of unreasonably blocking or discriminating against his content or service When deciding such cases, the NRA would reach a decision that would achieve to the extent possible all the various objectives listed in article 8 of the Framework Directive, including the promotion of competition and the objective of allowing users to have access to the content and services of their choice.

5. Summary of Differences

25. This quick comparison between the FCC Order and the new European net neutrality provisions reveals several key differences between the two regimes.

• The FCC Order imposes an affirmative no-blocking rule, whereas the European framework simply allows NRAs to intervene in dispute resolution proceedings should a service or content provider believes that it is the victim of unreasonable blocking. When deciding such a case, an NRA in Europe will take into account a number of factors and objectives, including reasonable network management and the need to ensure that end-users have access to the content and services of their choice. However, in practice a no-blocking rule may come into play only upon the complaint of a party which believes the rule has been violated.

• The FCC Order contains a non-discrimination provision, whereas the European framework does not. Under the European framework, a non-discrimination obligation can be imposed by NRAs only after a market analysis and an identification of an operator as holding significant market power. We will review this in more detail below.

• The FCC Order makes a distinction between fixed and mobile access providers, whereas the European framework makes no such distinction. We will examine this aspect of the FCC Order in more detail below, in the context of technological neutrality.

3. Review of EU Regulatory principles

26. We will now look at four of the regulatory principles that form part of road map for European regulators and courts when evaluating regulatory options: (A) the proportionality test, (B) the difference between asymmetric and symmetric regulation, (C) the principle of technology neutrality, and (D) the principles of objectivity and transparency.

1. Proportionality test

27. If US net neutrality rules were subject to scrutiny under European regulatory principles, the first question to analyze would be whether the US rules respect the principle of proportionality. The principle of proportionality requires that the regulator have identified a precise harm or market failure that needs to be cured. Under European regulatory practice applicable to electronic communications, the regulatory harm is generally identified through a market analysis that the regulator is required to conduct before imposing any asymmetric remedies. The relevant harm or market failure should be one that actually exists and can be observed in the market. In some cases, a regulatory authority can identify a market failure that is likely to occur in the future based on evidence collected during the market analysis. However, when the market failure has not already occurred, the burden of proof on the regulator is relatively high to show that the market failure is likely.
28. Once the regulatory authority has identified a particular market failure that needs to be addressed, the regulatory authority must choose a proportionate remedy, i.e. a remedy that is the least burdensome possible and still permits the problem to be addressed. To choose the least burdensome remedy, the regulator must generally consider several scenarios and consider the costs and benefits of each. This requires that the regulator have some idea of the cost that any given remedy would generate for the affected firms and for the market, and that the regulator make an estimate of the effectiveness of the relevant scenarios in terms of achieving the desired objective. A remedy with high potential costs should be used only as a last resort, where other less burdensome remedies will not be effective. 
29. Some of the regulatory authorities in Europe go so far as to conduct impact assessments evaluating the effect of several scenarios on consumers and on the relevant firms. Implicit in this test is the requirement that the regulator consider the scenario of doing nothing – of not imposing any regulatory remedies — and examining what the outcome for the market is likely to be under that scenario, bearing in mind the possibility of using other tools such as competition law. 
30. Where two or more regulatory options yield roughly the same results in terms of cost and benefits, the regulatory authority will look to other factors that might justify choosing one scenario over another. For example, where one scenario is more consistent with a recommendation by the European Commission, or with practices of other European regulatory authorities, a national regulator would give a preference to that scenario in order to promote European harmonization. Similarly, if one of the regulatory solutions would be more conducive to innovation, the regulatory authority would give preference to that regulatory solution over another that is less favourable to innovation. 

2. Asymmetric versus symmetric regulation

31. Another question that would have to be asked if the US rules were to be analyzed under European regulatory principles is whether the rules are designed to address anticompetitive conduct committed by a firm holding a dominant position on the telecommunications market. If the objective of the measure is to address anticompetitive conduct by a dominant operator, then the relevant cure for that anticompetitive conduct would be considered under European regulatory principles as an asymmetric remedy, i.e. a remedy that applies only to firms holding significant market power. 
32. Under European regulatory principles, any measure designed to counter market power must follow the rigorous market analysis procedure set forth in the directives. This requires that the regulatory authorities define a relevant market and show that a given firm holds significant market power on that market. The regulator must then show that reliance on competition law alone would not suffice to permit competition to emerge on the relevant market. A regulatory authority must satisfy a number of tests before imposing an asymmetric regulatory remedy. If a regulator attempts to shortcut this procedure by disguising what should be an asymmetric remedy as a so-called symmetric remedy, the regulator’s action can be challenged under European law.

3. Technology neutrality

33. Another requirement of the European framework is that regulations be to the full extent possible technologically neutral. Technological neutrality was included in the European framework in 2002, in order to respect the philosophy that all electronic communications networks should be treated alike from a regulatory standpoint. Before 2002, European countries had separate rules for cable networks, for mobile networks and for fixed wireline networks. In 2002 the difference between these networks was abolished from the regulatory framework. 
34. Today, all networks in Europe are subject to the same rules. Where networks make use of a radio spectrum, separate rules apply to the assignment and use of spectrum. Those separate spectrum rules are in most cases limited to rules to prevent harmful interference and to ensure efficient use of the scarce resource. But the rules applicable to the operation of the underlying network, as well as the rules applicable to an operator’s relations with its customers and with other carriers, are the same regardless of whether the network is fixed line or mobile. This is why the European rules on net neutrality make no distinction between fixed and mobile networks. 
35. In terms of asymmetric regulation, there are generally less regulations on mobile than on fixed networks. However, this is purely a result of the market analysis process and the conclusion that for mobile networks the competitive conditions are not the same as for fixed networks.
36. The European Commission has made efforts to eliminate any remaining differences between fixed and mobile networks in order to encourage converged fixed and mobile networks and service offerings. The latest example of this is the European Commission’s recommendation on setting regulatory tariffs for the termination of voice calls on fixed and mobile networks. The Commission has recommended that national regulators use a single methodology (LRIC) for the calculation of tariffs on both fixed and mobile networks. The reason for this was to eliminate artificial differences between calls made on fixed and mobile networks and thereby encourage the development of service offerings that combine unlimited fixed and mobile calls. The underlying policy objective behind this is to encourage innovation and make sure that consumers receive the full benefits of converged IP-based networks. 
37. Technology neutrality is akin to the idea of setting standards that do not prescribe a given technology but instead specify the performance standard to be attained.[31] Regulation should not prejudge technological choices, by picking technological winners and losers. But termination rates are not the same type of issue as capacity constraints – so while uniformity there makes sense, the differential treatment of restricted bandwidth in the US system also makes sense.
38. Technology neutrality is also a form of non-discrimination. One of the separate requirements under the European regulatory framework is that regulations treat in the same way similarly situated operators. Technology neutrality goes even farther, stating that regulation should not discriminate based on the kind of technology used.

4. Objectivity and Transparency

39. Two other requirements of the European framework are that regulations be objective and transparent. The test of objectivity means that the reasons for the regulations must be based on objectively verifiable facts and methodologies. The test of objectivity is related to the test of transparency, which is designed to ensure that the reasons and factual evidence on which any regulatory action is based are publicly available. Objectivity and transparency are designed to ensure that regulations are not adopted on the basis of behind-closed-doors meetings and secret trade-offs between the regulator and the regulated firms. Objectivity and transparency contribute to the legitimacy of regulatory measures[32]. 
40. Objectivity and transparency are also necessary in order for courts to be able to effectively review the action of the regulator and annul any action by the regulator that does not respect the regulator’s statutory mandate. This dovetails with the principle that regulatory decisions must be subject to full judicial review. Objectivity and transparency have led regulatory authorities in Europe to use public consultations prior to adopting any regulatory measures. Typically the regulatory authority will publish a document describing in detail the proposed measures and the reason why the regulatory authority deems such measures necessary. 
41. Objectivity and transparency also promote the objective of predictability of regulatory action. Predictability was added as a specific regulatory objective in 2009[33]. The consistency and predictability of regulatory action facilitate investment and risk-taking by firms. Predictability is an important factor contributing to economic welfare.

4. Confronting the FCC Order with European regulatory principles

42. Two aspects of the FCC’s net neutrality order would be problematic under the principles of the European framework: (A) the FCC’s differing treatment of fixed and wireless networks, and (B) the "no unreasonable discrimination" rule imposed on fixed operators.

1. The FCC Order’s differentiation of fixed and mobile broadband access providers would violate the European principle of technology neutrality.

43. The FCC Order subjects fixed and mobile broadband access providers to two different sets of rules, as summarized by the table below:

44. To illustrate the difference more vividly, the FCC’s rules allow mobile operators to discriminate – or at least they are not banded from such behavior. This behavior can take the form of entering into agreements for paid prioritisation with upstream content or application providers, whereas fixed operators are prohibited from doing so in most cases^[34]. The rules allow mobile operators to block certain lawful applications, whereas fixed operators are prohibited from doing so.
45. The creation of two sets of rules — one for fixed operators and one for mobile operators — contradicts the European approach of technology neutrality, and the whole purpose of the 2002 European framework, which was to abolish regulatory differences between different kinds of electronic communications networks. 
46. As noted above[35], European net neutrality rules take into account the constraints of wireless networks. For example, in Europe "reasonable network management" would tolerate more aggressive traffic shaping over mobile networks than over fixed networks because of the limited capacity of the shared radio access network. Mobile operators are also subject to separate obligations related to the use of radio spectrum. Those obligations may relate to the need to ensure there is no harmful interference, or may relate to coverage obligations imposed when the mobile operator bid for its spectrum.
47. The FCC has taken a different approach, creating separate rules governing the economic conduct for fixed and mobile networks, unrelated to technical issues linked to spectrum, for example. How does the FCC justify the difference?
48. As regards the lighter version of the no-blocking rules, the FCC explains that mobile operators need the ability to effectively manage their mobile broadband networks, and that the lighter no-blocking rules for mobile operators strikes the right balance between the need to protect consumers from inappropriate blocking and mobile operators’ need to manage their networks^[36]. The main reason for the lighter rule therefore appears to be the increased needs for mobile operators to conduct network management. However the FCC also states that the standard of "reasonable network management" will take into account the specific needs of mobile operators^[37]. 
49. A national court in Europe would immediately note that the FCC’s rule appears discriminatory and contrary to "technology neutrality." The court would then seek to enquire whether the FCC has cited a good reason for the difference in treatment. In the case of the no-blocking rules, the FCC’s justification seems to be that mobile operators need more latitude to conduct network management. However this latitude has already been accorded under the "reasonable network management" rule. The FCC’s rule in effect permits mobile operators to block services and applications based on reasons unrelated to legitimate network management constraints. And yet the FCC’s justification for the measure appears to be limited to network management constraints. A national court in Europe would identify this as a defect in reasoning and possibly annul the FCC’s decision on this basis.
50. The FCC cites differences in the competitive environment for fixed and mobile as a reason for creating two different sets of rules. According to the FCC, "mobile broadband is an earlier-stage platform than fixed broadband, and it is rapidly evolving."^[38] The FCC states that business models for mobile broadband are evolving quickly. Finally, the FCC points out that most customers have more choices for mobile broadband than for fixed (particularly fixed wireline) broadband.^[39] 
51. If the FCC Order were subject to judicial review in Europe, a court could challenge these two justifications. First, a fixed broadband operator could demonstrate that platforms for fixed broadband access are evolving quickly, and in some cases are quite recent. DSL-based platforms are migrating to VDSL (FTTC) or FTTH platforms. Cable networks are evolving toward DOCSIS 3. Business models are evolving quickly as well, fixed operators experimenting with different subscription models for very high bandwidth offers.
52. As regards the difference in competition between fixed and mobile platforms, a court in Europe could criticize the FCC’s reasoning on the basis that the FCC did not conduct a market analysis to support its assertion that the competitive conditions for mobile networks are more intense than for fixed networks. In Europe, a regulatory authority would have to conduct a market analysis in order to evaluate the level of competition on the retail market for fixed and mobile broadband before the regulatory authority could use the lack of competition as a reason to apply different rules to fixed and mobile platforms. In footnote 47 of the FCC Order, the FCC states that it does not have to conduct a "market power analysis":

Because broadband providers have the ability to act as gatekeepers even in the absence of market power with respect to end users, we need not conduct a market power analysis.^[40]

53. A last reason cited by the FCC for the difference in treatment is that Verizon Wireless has been granted spectrum in the 700 MHz band under open access conditions. In 2008, the FCC licensed a number of blocks of spectrum in the 700 Mhz band and set aside one block, called Block C, as one that would be subject to open access obligations similar to those contained in the FCC’s net neutrality order. Because of these open access obligations, Block C commanded a lower price at the auction than did the other blocks of spectrum unburdened by these restrictions. 
54. The official reason cited by the FCC in its 2010 order for not imposing the full set of neutrality obligations on mobile operators is that the FCC felt it appropriate to observe the operation of the Block C network before imposing an across-the-board net neutrality obligation on all mobile operators. The FCC also said that the market seemed to be naturally moving toward acceptance of open access conditions among mobile operators and therefore it was not yet necessary for the FCC to intervene.
55. These two reasons cited by the FCC seem contestable. As pointed out by Commissioner Robert McDowell in his dissenting opinion, fixed broadband operators almost uniformly apply open Internet policies. The market seems to have settled clearly in favor of open Internet policies for fixed broadband providers, the only exception being a handful of incidents, including Comcast’s blocking of the BitTorrent protocol and the Madison River telephone company’s blocking of Vonage’s voice over IP service. Mobile broadband providers on the other hand have much more routinely applied blocking practices, especially VOIP applications.
56. The real reason for the difference in treatment may be that the FCC could not apply across-the-board net neutrality obligations on mobile operators who purchased spectrum at the 700Mhz auction unburdened by open access conditions.
57. Because the unrestricted spectrum commanded a higher price than the Block C spectrum burdened by net neutrality obligations, operators having purchased the more expensive spectrum could potentially make a claim for refund of a portion of the price they paid for the unburdened spectrum. This reason is not referred to in the FCC Order, but it may have contributed to the FCC’s decision not to apply the full set of net neutrality obligations to mobile operators. Or it was a recognition that, given the dynamically growing amount of wireless video transmissions, which use much more bandwidth than voice or simple data, wireless operators need more leeway to manage video traffic. Although arguably this aspect would be already picked up in the "reasonable network management" rule.

2. The “no unreasonable discrimination” obligation would likely violate European methodology because no market analysis was performed

58. The most controversial obligation imposed on fixed broadband operators in the U.S. is the “no unreasonable discrimination” rule. According to the FCC, the reason for the rule is to prevent fixed broadband access providers from leveraging their gatekeeper function by extracting payments for priority services from upstream content and application providers. As noted above[41], the FCC is careful to point out that their reasoning is not based on the exercise of monopoly power by broadband access providers and that consequently it is not necessary for the FCC to conduct an analysis of market power. 
59. The FCC’s reasoning nevertheless is based in large part on the market power of broadband access providers with regard to upstream content and application providers. This kind of market power is identical regardless of whether the operator is a fixed or a mobile broadband provider, again highlighting certain incoherence in the FCC’s differential treatment of fixed and mobile operators. The reasoning of the FCC is that an upstream applications provider has no choice but to go through the broadband access provider’s network in order to reach customers of that broadband access provider. Viewed through this prism, each broadband access provider has a gatekeeper function with regard to the access provider’s own customers.
60. The FCC’s assertion may well be true. However, under the European framework it would be necessary to conduct a much more detailed market analysis to determine whether the gatekeeper’s market power is sufficient to justify regulation of this type. 
61. The reasoning used by the FCC is similar to that used in Europe to justify the regulation of tariffs for terminating calls on fixed and mobile networks. Indeed some operators in Europe argue that regulators should impose a data termination tariff for Internet traffic similar to the tariff applicable to the termination of voice calls and SMS messages. The FCC obviously does not go this far. 
62. Indeed the imposition of a data termination rate would radically change the structure of agreements for the exchange of Internet traffic. Not going so far as to create a data termination rate, the FCC nevertheless imposes a non-discrimination obligation on broadband operators vis-à-vis upstream content and application providers.
63. In Europe the non-discrimination obligation is provided for in article 10 of the Access Directive, and can only be imposed on operators holding significant market power. As noted above[42], the FCC justifies its ruling on the fact that fixed access providers enjoy a kind of market power with regard to upstream content and application providers for each customer – customers can’t change their ISP easily. But the FCC’s assertion does not seem to be supported by a detailed market analysis of the kind one would need to conduct in Europe. 
64. In Europe, a regulatory authority would have to first define the relevant market using a hypothetical monopolist (SSNIP) test and then determine whether one or more operators can behave independently of its competitors and ultimately of consumers on that market. This would determine whether the operator holds significant market power ("SMP"). In conducting this market analysis, the regulator would have to analyze the likely reaction of content and application providers to any attempted discrimination by a broadband access network, analyze the role of content delivery networks (CDNs) and of the possible reaction of consumers to discriminatory practices by their access provider.
65. This analysis is not simple. It may or may not lead to the conclusion that the broadband access provider enjoys a high degree of market power. The market analysis may show that discriminatory behavior of this kind by a broadband access provider would immediately be sanctioned by unfavorable reactions by the broadband access provider’s own retail customers. If that were the case, the market would function as a sufficient safeguard against discriminatory behavior and regulatory intervention would be unnecessary.
66. Another curious aspect of the FCC’s decision is the definition of what constitutes “unreasonable discrimination.” Some operators argued in the FCC proceedings that the test should be one of “anti-competitive discrimination.” Operators pointed out that the cases of discrimination cited by the FCC were essentially cases of anticompetitive conduct. By referring to anticompetitive discrimination, operators would have a better idea of what forms of discriminatory behavior are prohibited. 
67. The FCC declined to use this approach and imposed a non-discrimination obligation that would in effect prohibit discrimination of a kind that is permitted by competition law. The FCC stated that its unreasonable discrimination rule would most likely apply to any kind of paid prioritization arrangement between a fixed broadband access provider and an upstream provider of content or application. In other words, such an arrangement would likely be prohibited under the FCC rule, even if it were not prohibited by competition law.
68. Based in large part on competition law, the European framework is less prescriptive with regard to upstream paid prioritization arrangements and would not accommodate an across the board prohibition of the kind imposed by the FCC. In Europe there is no prohibition of discrimination per se, but national regulatory authorities are able to intervene in individual dispute resolution proceedings between content and application providers on the one hand, and broadband access providers on the other.
69. When deciding individual dispute resolution proceedings regulatory authorities must take into account the objectives of article 8 of the Framework Directive, which include the objective of ensuring that end users have access to the content and applications of their choice. Under the European approach, ex post intervention through dispute resolution would be less intrusive and less potentially harmful to innovation and competition than would an across-the-board non-discrimination rule. 
70. In competitive markets, discrimination can lead to innovation and efficiency. Discrimination can also lead to distortions of competition, particularly when the firm discriminating occupies a dominant position. Because the European regulatory framework is based in large part on competition law principles, and on the principle that sector-specific regulation should disappear once markets become competitive, European lawmakers would have had difficulty imposing a non-discrimination obligation without the identification of a clear market failure.
71. This is particularly so since under the European Framework, the non-discrimination rule is closely linked to regulation of operators with significant market power. Consequently, there would be an inherent contradiction if European regulators were to impose a non-discrimination rule outside the context of a market analysis and a determination of significant market power. The FCC Order would fail under this test.

5. CONCLUSION

72. The difference between the European and U.S. regulatory approaches to net neutrality is due in large part to the underlying principles that form the basis for the European framework for electronic communications. Those underlying principles include, for example, the principle of technology neutrality, which is largely absent from the U.S. regulatory framework. The U.S. Communications Act is still divided into silos based on the kind of network and technology being used to provide the service. 
73. The principal reason for the European reform in 2002 was to eliminate these technology silos. Consequently, technology neutrality is firmly entrenched in the DNA of the European framework and it would have been difficult if not impossible to create a separate body of net neutrality rules for fixed and mobile networks as was done in the U.S. The European regulatory framework will naturally acknowledge technological differences between fixed and mobile services, but this would be taken into account in the notion of “reasonable network management,” as opposed to embedded in the rules from the outset. 
74. By contrast, the FCC’s approach creates different sets of rules for fixed and for mobile networks, while in addition stating that reasonable network management will take account of the technical differences between the two kinds of networks. The FCC seems guilty of a form of double counting when justifying the difference in treatment between fixed and mobile networks.
75. The other main difference between the US and the European approaches relates to the non-discrimination rule. Here, too, we see one of the philosophical underpinnings of the European framework. The European framework divides regulatory problems into two categories: problems that relate to lack of competition, and problems linked to more general issues of consumer protection or compliance with technical standards. The methodology for treating competition-related issues is to conduct a market analysis, and to identify specific market failures and firms that hold significant market power. It is only after going through this process that NRAs may impose asymmetric remedies such as a non-discrimination obligation. 
76. The purpose of this European methodology is to ensure that regulation is imposed only where it is absolutely necessary and that regulatory measures are rolled back as soon as the market is effectively competitive. For issues related to general consumer protection and compliance with technical standards, regulators in Europe can more easily impose obligations on operators of all kinds. NRAs in that case are not required to conduct a market analysis. This dichotomy in the EU methodology no doubt explains why it would have been difficult to impose a non-discrimination obligation on all operators in Europe. The non-discrimination rule targets competition related problems, such as vertical leveraging. 
77. Just as distinguishing between fixed and mobile operators would have run against the DNA of the European regulatory framework, so would have the idea of imposing an across the board non-discrimination obligation without linking it to some form of market analysis. Some say that the non-discrimination rule imposed by the FCC also runs against the regulatory DNA of the U.S. Communications Act insofar as non-discrimination is a remedy generally associated with “common carriers” under Title II of the Communications Act, and the FCC has not attempted to argue that Internet access providers are common carriers. 
78. Consequently, some observers have pointed out that there is an inherent contradiction between the non-discrimination obligation in the FCC Order and the FCC’s qualification of Internet access providers as information service providers (ie. non common carriers).
79. As a general matter, the European Framework requires Member States to eliminate regulation where it is no longer necessary[43]. Consequently the imposition of a general non-discrimination or no-blocking rule would at a minimum have required a showing of an actual market failure linked to upstream discrimination by Internet access providers, which could only be addressed by a general non-discrimination obligation. In the current state of the Internet, a market failure of this kind is difficult to show. As pointed out by the Commission in a recent communication[44], evidence of blocking is currently lacking.