Abstract
The US debate about net neutrality has been unusually contentious for a telecommunications regulatory issue, most recently culminating in a 2017 reversal of a 2015 decision to apply traditional telephone regulations, written for a monopoly era, to internet service providers. This article reviews this history, beginning 1956 when the government first imposed an industry boundary between transmission of information on the one hand, and the creation and processing of information on the other. This regulatory legacy remains embedded in US law and has led to some of the muddle. This article also examines the academic literature relating to net neutrality. On this, the answers found in the literature vary depending on assumptions made about technology, industry structure, and industry practices. When the answer to the question of whether regulations are beneficial is “it depends,” and the scenarios that give different answers are realistic, it would seem that the policy approach should favor applying competition and consumer protection laws that address problems when they occur rather than ex ante regulations, which would be certain to harm at least in some situations.
1 Introduction
The US has engaged in an increasingly contentious debate about net neutrality for more than a decade. The debate was effectively launched by Wu (2003), who made a case for regulatory restrictions on how internet service providers (ISPs) might structure their service offerings. The idea was perhaps given legs by Yoo (2004) and Wu and Yoo (2007), where Yoo essentially took the opposite view of Wu and the visibility of the articles lent the issue legitimacy.
The nature of the conflict on net neutrality has been unusually partisan and combative for an information technology issue. It has included the usual scholarly articles, articles for the general public, regulatory proceedings, Congressional hearings, and social media postings. But there have been unusually intemperate statements, such as Investor’s Business Daily referring to advocates of net neutrality regulations as snake oil salesmen (Investor’s Business Daily 2017) and Senator Ted Cruz referred to the regulations as “Obamacare for the internet.” (Chen 2018) And there are political exaggerations and confuscations, such as the Senate Democrats tweet on February 27, 2018, stating that absent net neutrality regulations, the internet will deliver content one word at a time, implying that the internet would slow down to a crawl.[1] (@SenateDems 2018) Partly as a good-humored response to doomsday-like predictions of the Federal Communications Commission (FCC) voting to 2017 drop net neutrality regulations, the current FCC chairman’s chief of staff, Matthew Berry, tweets almost daily (as of the time of this writing) counting the days since the decision went into effect and noting the continued thriving of the internet.[2]
The back and forth has been uncivil. There have been numerous protests against the FCC (Fung 2018) and two of its chairmen,[3] bomb threats against the agency, (Mak 2017) and threats of violence against the current FCC chairman and his family (Kang 2018). Activism of this nature is unusual for a telecommunications issue, perhaps indicating that some of the oversized rhetoric has been taken literally by some people, that the activism is part of a larger political agenda, or both. Resolving the motives behind these aspects of the conflict is beyond the scope of this paper, but would be an important research topic.
Something that is puzzling about the intensity of the conflict is it appears to be about means rather than goals. Jamison and Layton (2016) conclude that both proponents and opponents of net neutrality regulations have the same stated goal: Robust and open networks so that entrepreneurs can thrive and benefit consumers. If the stated goals are the actual goals, then where the regulatory proponents and opponents differ is the role of regulation. Some believe that ISPs should be regulated in ways similar to those of monopoly telephone companies. Others believe that the FCC should play a more limited role as a resolver of disputes. Still others believe that market pressures will appropriately incentivize ISPs and that consumer harms should be addressed with ex post regulations, such as competition law.
The divide has led to wide swings in how regulators and policy makers have approached the issue and defined it. One extreme has been the effectively hands-off approach adopted first during the administration of President Clinton and now followed by the Trump administration. The other extreme was the Obama administration’s adoption of what is called Title II regulation, which is the regulatory framework contained in the Communications Act of 1934 to regulate monopoly telephone companies. In between these are the light-handed policies attempted during the George W. Bush administration and early in the Obama administration.[4]
This paper’s purpose is to examine how net neutrality policies developed in the US and why, and the relationship between scholarly work and policy development in this issue. It proceeds as follows. Section 2 describes the various regulatory approaches, including how regulation has defined internet services, the hands-off approach of the FCC under former-President Clinton, how the FCC under former-Chairman Michael Powell and former-Chairman Julius Genachowski attempted to address discrimination concerns while maintaining a light-handed approach, the turn to Title II regulation under former-Chairman Tom Wheeler, and the reversal of that policy under Chairman Ajit Pai. Section 3 explores the economic research on net neutrality, emphasizing articles published in leading economic journals. Section 4 is the conclusion.
2 Regulatory Approaches in the US
The FCC has been the locus of activity on how to properly address net neutrality issues in the US. But the agency has struggled to develop a consistent course of action that courts would accept as appropriate given the agency’s authorizing statutes. This is probably reflective of the structure of the relevant US laws, where Title II is the primary statute for regulating telecommunications common carriers and it intertwines common carrier regulation with utility-style regulation, which is designed to regulate legal monopolies.
Cherry (2003) draws clear distinctions between utility regulation and common carrier regulation:
These [common carrier] obligations evolved in medieval England to address numerous situations of economic coercion, exploitation and the illegal wielding of bargaining power. These obligations are: to charge reasonable prices (“just price”); to serve without discrimination; and to exercise their calling with adequate care, skill and honesty.
A public utility is a private corporation that provides a service of public importance, or necessity, under a government grant of privilege. This grant of privilege imposes an affirmative duty to render service demanded by any member of the public. Public utilities were initially created by local government grants of franchises, which were subsequently preempted by the codification of public utility law in state statutes. It should be noted, however, that a public utility’s obligations tend to be greater than those of a common carrier because a public utility typically bears the affirmative duty to extend facilities to serve an entire community, and is also constrained in its ability to discontinue the provision of service. (brackets added and footnotes omitted)
But as Hauge and Jamison (2014) explain, most literature and analyses treat them as nearly the same, if not the same, as does Title II. Perhaps the muddle of court and regulatory decisions that this section describes could be avoided if statutes provided clarity on when a common carrier is and is not a utility, and the FCC’s authority to regulate a common carrier under utility and non-utility conditions.
This section reviews the regulatory framework that existed prior to the FCC’s attempts to address net neutrality and then examines the attempts themselves.
2.1 The Development of the Regulatory Philosophy
The regulatory story underlying net neutrality began over 60 years ago. In 1956, the Department of Justice and AT&T entered into a consent decree[5] that prohibited AT&T from serving unregulated markets and confining it to common-carrier communication services, which were transmission only, and government projects. Also in 1956 the DC Circuit Court issued a decision in Hush-A-Phone,[6] which allowed customers to attach a plastic device to AT&T phones, which AT&T had attempted to prohibit. Following up in 1968 the FCC decided in Carterfone[7] that anyone could connect any privately-beneficial equipment to AT&T’s telephone network as long as the equipment was not publically harmful. This triggered the development of technical standards for equipment so that customers could choose any equipment they wanted that met the standards. This became part of the paradigm for net neutrality regulations.
The principles underlying the 1956 Consent Decree, Hush-A-Phone, and Carterfone, combined with rapidly changing computing and telecommunications technologies, led the FCC into a series of decisions collectively known as the Computer Inquiries. These decisions created the regulatory framework that led the agency to define the internet as an information service, not a common carrier service. This designation determined the nature of the agency’s jurisdiction over the internet.
The first of these inquiries, Computer I,[8] created a dichotomy between “basic” and “enhanced” telecommunications services.[9] In its Second Computer Inquiry, the FCC defined basic services as “pure transmission capability over a communications path that is virtually transparent in terms of its interaction with customer supplied information”[10] that were “regulated under Title II of the [Communications] Act.”[11] Enhanced services were “any offering over the telecommunications network which is more than a basic transmission service. In an enhanced service, for example, computer processing applications are used to act on the content, code, protocol, and other aspects of the subscriber’s information.”[12] The FCC concluded that “enhanced services should not be regulated under the Act.”[13]
This line between basic and enhanced found its way into the Modification of Final Judgment (MFJ) that broke up AT&T in 1984. The MFJ distinguished between “telecommunications services” that were “actually regulated by tariff,”[14] and “information services,” which included “data processing and other computer-related services,”[15] and “electronic publishing services.”[16] The Telecommunications Act of 1996 (the 1996 Act) picked up this terminology, defining “information service” the same as in the MFJ.[17] The 1996 Act mapped to the MFJ because the 1996 Act was enacted in part to resolve the litigious process that had emerged since 1984 for unwinding the restrictions that the MFJ placed on the divested Bell companies.
With the 1996 Act, Congress intended to “promote competition and reduce regulation.”[18] In doing so it distinguished between the lightly regulated “information services” and “telecommunications services,”[19] which were regulated as common carrier services. Congress also found that the “Internet and other interactive computer services have flourished, to the benefit of all Americans, with a minimum of government regulation”[20] and made it the policy of the US to “promote the continued development of the Internet and other interactive computer services and other interactive media” and “to preserve the vibrant and competitive free market that presently exists for the Internet and other interactive computer services, unfettered by Federal or State regulation.”[21]
The FCC followed this basic-information dichotomy and also kept to the 1996 Act’s preference for an internet unfettered by Federal or State regulation throughout most of its work on net neutrality, with the exception being the 2015 Open Internet Order.[22] For example, in the Commission’s review the 1996 Act’s definitions as they applied to the internet, it concluded that internet access services were properly classified as an information service.[23] It also concluded that subjecting ISPs to “the broad range of Title II constraints,” would “seriously curtail the regulatory freedom that the Commission concluded in Computer II was important to the healthy and competitive development of the enhanced-services industry.”[24]
2.2 Decisions on Net Neutrality and their Aftermaths
The FCC’s first attempt to address net neutrality was in 2005 – 3 years after Wu’s article – when the Commission adopted an Internet Policy Statement intended “to ensure that broadband networks are widely deployed, open, affordable, and accessible to all consumers”.[25] The statement provided four consumer-centric guiding principles, also referred to as “Four Freedoms” (Powell 2004)
Consumers are entitled to access the lawful Internet content of their choice.
Consumers are entitled to run applications and use services of their choice, subject to law enforcement’s needs.
Consumers are entitled to connect their choice of legal devices that do not harm the network.
Consumers are entitled to competition among network providers, application and service providers, and content providers.
The last principle should probably be read as a statement that regulators should not create barriers to competition: While it is meaningful to say that customers have a right to choose, it is another thing to say that they are entitled to multiple choices because the latter implies that someone has an obligation to create additional providers in otherwise monopoly markets. That is not something the FCC has taken upon itself to do.
If they had been imposed on ISPs, the first three principles would be consistent with the common carrier obligations as the FCC has interpreted them under Title II. But the principles were framed like consumer protection. If the FCC had stopped there and simply (1) pressured ISPs to always offer a broadband service package that included access to all lawful content and apps, and allowed customers use of any device of their choosing, while (2) not prohibiting ISPs from offering other services packages, the net neutrality debate might have ended there in the US because ISPs rarely violated the three principles and most ISPs never violated them. But that isn’t what the FCC did.
In response to a complaint that Comcast was degrading peer-to-peer internet traffic, the FCC chose to impose its principles as obligations.[26] Comcast appealed and the DC Circuit Court rejected the FCC’s decision, stating that the FCC had failed to provide statutory authority for imposing obligations.[27]
With the ball back in its court, the FCC launched a second light-handed attempt to address net neutrality in 2010 when the agency tried to write administrative rules regarding net neutrality.[28] This was early in President Obama’s first term. In this decision, the FCC adopted three basic rules:
Transparency. Fixed and mobile broadband providers must disclose the network management practices, performance characteristics, and terms and conditions of their broadband services;
No blocking. Fixed broadband providers may not block lawful content, applications, services, or non-harmful devices; mobile broadband providers may not block lawful websites, or block applications that compete with their voice or video telephony services; and
No unreasonable discrimination. Fixed broadband providers may not unreasonably discriminate in transmitting lawful network traffic.
As part of this 2010 attempt to define rules for ISPs, the FCC adopted a light-handed, multistakeholder approach for resolving net neutrality issues. This approach was short-lived, as the FCC never utilized the multistakeholder group (Jamison and Layton 2016).
The agency failed to implement its multistakeholder approach in part because, in 2014, the DC Circuit Court again reversed the FCC on jurisdictional grounds, finding that the 2010 rules were effectively common carrier obligations. This was a problem because the FCC had determined that ISPs were not common carriers and so not subject to common carrier requirements in Title II.[29]
In its ruling, the DC Circuit Court effectively provided the FCC with a roadmap for how to classify ISPs as common carriers and then impose net neutrality rules. The FCC followed the court’s advice in its 2015 Open Internet Order,[30] in which the agency took the heavy-handed approach of adopting the 1930s-era rules for regulating monopoly telephone companies. More specifically, the Commission, under the prompting of then-President Obama, (Obama 2014) classified consumer broadband service as common carrier services under Title II of its enabling statutes and reclassified mobile broadband Internet access service as a commercial mobile service. The Commission also adopted three bright-line rules that had the effect of prohibiting blocking, throttling, and paid-prioritization. It also adopted a general Internet conduct standard,[31] as well as “enhancements” to the transparency rule developed in the 2010 decision. The agency is permitted under its statutes to forbear from regulations that it believes to be inappropriate and indicated that it would do so with respect to some Title II regulations, such as price controls.
The FCC’s stated rationale for adopting the Title II approach included a belief that the country needed more and better broadband; the rules adopted in 2010 appeared to not hinder internet development, so rules under Title II should not either; regulatory rules might even stimulate investment; it needed to use its authority under Title II in order to impose net neutrality rules; and “broadband providers hold all the tools necessary to deceive consumers, degrade content, or disfavor the content that they do not like.” The DC Circuit Court upheld the decision, although Judge Williams issued a strongly worded dissent.[32]
Members of Congress and former FCC chief economists criticized the agency’s 2015 decision and the process used to reach it. The FCC’s chief economist at the time of the decision called the rulemaking process and subsequent order an “economics free zone” because the agency excluded input from its own economists when developing the decision, thus getting its economic analysis wrong (Brennan 2016). Another former chief economist was critical of the decision’s lack of economic rationale, stating that while he appreciated the agency citing his work in the order, that the substance of his articles was the opposite of what the order claims (Katz 2016).
Two important dissenters from the FCC’s 2015 decision were the two Republican commissioners, then-Commissioner Ajit Pai and Commissioner Michael O’Rielly. The Commission’s 2015 split over net neutrality – the three Democrat commissioners voted for the 2015 decision and the two Republican commissioners voted against it – reflected not just a disagreement on regulation, but a partisan divide that had emerged under then-Chairman Tom Wheeler (Wallsten 2016). So it was no surprise that when Republicans gained a majority at the Commission in 2017 that they voted to effectively vacate the 2015 decision, stating, “We reverse the Commission’s abrupt shift 2 years ago to heavy-handed utility-style regulation of broadband Internet access service and return to the light-touch framework under which a free and open Internet underwent rapid and unprecedented growth for almost two decades.”[33] More specifically, the FCC:
Ended “utility-style regulation of the Internet in favor of the market-based policies”;
Reclassified consumer broadband services as an information service and not as a common carrier service; and
Reinstated the private mobile service classification of mobile broadband Internet access service.
In reversing the 2015 decision, the FCC also restored the authority of the “Federal Trade Commission to police the privacy practices of Internet Service Providers (ISPs).” The Federal Trade Commission (FTC) is the federal agency charged with consumer protection and competition jurisdiction in broad sectors of the economy. By statute, the FTC does not have authority over Title II common carriers. Prior to the 2015 decision the FTC had been regulating privacy with respect to ISPs. The FCC’s 2015 caused the FTC to lose jurisdiction to do that, but the 2017 restored the FTC’s authority.
The FCC’s rationale for its 2017 decision included beliefs that laws other than Title II were better suited for protecting consumers and competition, that Title II regulations suppressed infrastructure investment, that classifying internet as a common carrier service was unsound as a legal matter, and that the costs of the 2015 decision outweighed the benefits. In its 2015 decision, the FCC specifically chose to not conduct a cost-benefit analysis.
2.3 Responses to the 2017 Decision
The supporters of imposing Title II on ISPs reacted in opposition to the new FCC decision. In May 2018 the US Senate voted to overrule the FCC’s 2017 decision, but the US House of Representatives failed to go along. Under the Congressional Review Act, Congress can cancel an agency decision within 60 days after the rule is reported to Congress.[34] With a divided Congress effectively siding with the FCC in the 2017 decision, the decision went into effect on June 11, 2018.
Some state governments sprang into action. Governors in Hawaii, New Jersey, Montana and New York issued executive orders stating that their governments would not do business with broadband providers that did not follow net neutrality regulations. State legislatures in California, Oregon, and Massachusetts took up legislation to impose net neutrality requirements on broadband providers. And over 20 state attorneys general, plus the District of Columbia, sued the FCC, arguing that the 2017 decision was arbitrary, capricious, an abuse of procedure, and a violation of applicable laws (Layton 2018). These state actions map to the partisan divide that has come to be part of net neutrality in the US as all of the governor offices, states attorney general offices, and state legislators that have acted are controlled by Democrats.
As of the time of this writing, these state responses have not been resolved. They all have legal problems. The governors’ actions might violate the Supreme Court’s marketplace-participant doctrine, which holds that a state cannot use procurement practices that are tantamount to regulating an industry (Cooper 2018). Actions by state legislatures might be viewed as regulating interstate commerce, which states do not have authority to do. In order to avoid this problem, states would likely have to separate internet traffic that has its beginning and end points within the state from interstate and international traffic, something that might be impossible to do in any practical way (Lyons 2018). The legal appeals of the FCC’s 2017 decision face the problem of judicial deference: Courts in the US largely accept regulatory agencies’ claims regarding jurisdiction based on interpretation of their statutes, and the FCC in 2017 appeared to map its decision both to the evidence it had and to its enabling statute. So while it may seem strange that an agency would make a decision and then reverse itself within a span of 2 years, courts in the past have upheld their authority to do so (Hurwitz 2017).
3 Academic Research on Net Neutrality
Most of the substantive arguments offered in the net neutrality debate are about the economic effects of ex ante regulation – regulations restricting conduct and normally imposed by sector regulators – versus those of ex post regulation – responses of competition regulators to anticompetitive conduct. Ex ante regulation makes certain conduct illegal whereas ex post regulation makes conduct illegal only under certain conditions, such as when consumers are harmed. These issues have been researched in the scholarly economics literature, so this section reviews that literature.
The emphasis here is on what has been demonstrated in articles that use explicit economic models and that appear in higher-level, peer-reviewed economics journals.[35] These articles’ findings are examined in subsections 3.1 through 3.7, which are organized into sub-questions: (1) How regulations restricting ISPs from offering enhanced network features (primarily fast lanes) might affect (a) total welfare, (b) network investment, and (c) the variety of content on the internet and content provider investment; (2) How prohibitions on network termination fees affect total welfare; (3) How prohibiting ISPs from blocking content affects total welfare; (4) How prohibiting ISPs from blocking access to rivals affects total welfare; and (5) How price regulations affect welfare and investment. Subsection 3.7 reviews articles that appear to be frequently cited, but that are in non-economic journals.[36] Subsection 3.8 summarizes conclusions in the literature.
3.1 Effects of Ex Ante Restrictions Prohibiting Features Such as Fast Lanes on Welfare[37]
The literature finds that the welfare effects of prohibitions on features such as fast lanes depend on market conditions, such as whether ISPs are monopolies, how charges might be implemented, network engineering, customer valuation of content, and the types and variety of content provided on the internet. Most articles find that the ex ante regulations decrease welfare.
One of the earliest papers, Hermalin and Katz (2007) addresses the situation of a monopoly ISP that does not provide its own content, but provides network services to competing content providers and to consumers. (In contrast to most other papers, this paper analyzes transmission differences as quality differences as opposed to explicit fast lanes or paid prioritization.) The paper finds that ex ante regulations can either increase or decrease total welfare, but are most likely to reduce it: The restrictions lower total welfare when content providers who would prefer low-quality connections are excluded from the market because ISPs are not allowed to offer lower quality services, or when content providers who would prefer high-quality service have to settle for lesser quality service. The paper finds similar conclusions for a duopoly ISP market.
Choi and Kim (2010) examines a situation where there is a monopoly ISP and two content providers that differ in the quality of their content. It allows only one content provider to purchase a fast lane. The research finds that ex ante regulation lowers welfare when content providers greatly vary in what they provide.
Njoroge et al. (2014) examines a situation where there are two ISPs and neither provides its own content. There is a large number of competing content providers that vary in the quality of their individual services and that receive money only from advertising. The paper assumes a content provider might connect directly to one ISP and then reach consumers that are connected to a second ISP, and that the second ISP might charge the content providers for reaching its consumers. The paper finds that if an ISP imposes fees on content providers that connect through another ISP, ex ante regulations lower total welfare.
Njoroge et al. (2014) finds that ex ante regulations lower total welfare, but can increase welfare if content providers vary greatly in the types of content that they provide. These results are different from those of Choi and Kim (2010) regarding the effects of content differentiation in part because Choi and Kim allows only one content provider to purchase the fast lane. Choi and Kim also assumes a monopoly ISP and that the ISP offers fast lanes by offering paid prioritization. Njoroge et al. assumes that the ISP offers separate fast and slow lanes. This is an example where technology choice matters to model results.
Most economic analyses assume that ISPs can vary the amount of bandwidth that they provide, which appears to align with actual ISP practice to choose network capacity based in part on the amount of traffic. In contrast, Economides and Hermalin (2012) examines a situation where there is a single ISP whose total bandwidth is fixed, the ISP does not provide its own content, and there is a large number of content providers, but each is a monopoly, which is an unusual assumption. The paper states that ex ante regulations improve total welfare in monopoly situations, but appears to omit industry profits from the welfare calculation.
Economides and Tåg (2012) also makes an unusual assumption, namely that content providers may not necessarily benefit from advanced network features such as fast lanes. The paper considers both monopoly ISP and duopoly ISP situations where there are many content providers that obtain revenue from advertising and that differ in their costs of setting up their services. The model omits considering how ISPs might alter their investment and requires that all consumers purchase ISP service. The paper finds that the effects of ex ante regulations on total welfare depend on a variety of market conditions, but the paper restricts its welfare calculation to only the value of content. Economides and Tåg also finds that consumers are always worse off with ex ante regulations. In a related article, Caves (2012) uses the Economides and Tåg model and finds that such regulation decreases total welfare under the most common market conditions.
In a more generally applicable paper, Bourreau et al. (2015) finds that ex ante regulations lower welfare. They assume there are two competing ISPs, which aligns with the situation in many US markets, and content providers vary in their preferences for fast lanes, which also seems appropriate.
Choi et al. (2015) find that the effects of ex ante regulations on welfare are sensitive to the degree to which content providers receive revenue from advertising versus payments from consumers for accessing content. This paper is more general in that it allows that content providers are heterogeneous in the value of their content and vary in how much they value fast lanes. Also, content providers receive revenue from advertising and may charge customers for accessing content. It examines monopoly and duopoly ISP markets and finds that it makes no difference as long as the ISPs determine interconnection charges and subscribers pay a single subscription price. In general the paper finds that ex ante regulations reduce welfare when either consumers receive all of the surplus or content providers do, but the regulations may be welfare improving depending on how surplus is shared between the two.
Reggiani and Valletti (2016) examines a situation where there is a single ISP and multiple content providers. Unusually, there is only one large content provider, which is defined as one that sells multiple applications. All other content providers are small and provide only a single application. An important feature of this paper is that the ISP charges a fixed price for prioritization and that prioritization is not a substitute for any other input that a content provider might use. These mean that buying prioritization is more profitable for the large content provider than for the small content providers. Therefore the effects of fast lanes are driven by the response of the large content provider. The finding is that ex ante regulations lower welfare if they lower content innovation, which the paper defines as the amount of content provided.
Greenstein et al. (2016) offers a summary of the economics literature with respect to net neutrality and provides basic analyses to address the core issues. The paper shows that general ex ante restrictions on ISPs charging content providers (that depend on advertising for revenue) for access to consumers may increase or decrease total welfare, but the restrictions do result in higher consumer prices for ISP services. The paper also explains that if content providers charge customers for services, such as in the cases of Netflix and Amazon Prime, then the regulations decrease market efficiency if ISPs are able to charge different service fees to lower-value content providers than to higher-value content providers. The regulations can increase efficiency if ISPs are unable to discriminate between content providers. Finally, the paper finds that recent theoretical contributions generally support the idea that ex ante net neutrality regulation on competing platforms is welfare decreasing, but states that this is an under-researched area.
3.2 Effects of Ex Ante Restrictions Prohibiting Features Such as Fast Lanes on Investment
The literature gives mixed results, showing that the investment incentive is sensitive to how content providers and consumers respond to prices and to how consumers value content. Choi and Kim (2010) find that ISPs’ incentives to invest depend on the value that customers place on network services relative to the value that some content providers might place on faster delivery speeds, so the effects of ex ante regulation are ambiguous.
Several models find that ex ante regulations lower ISP investment. These include Economides and Hermalin (2012), Njoroge et al. (2014), and Bourreau et al. (2015). Reggiani and Valletti (2016) find that the regulations lower network investment if fast lanes stimulate content provision by a large content provider enough to overcome any decline in the number of small content providers, which the ex ante regulations would cause.
3.3 Effects of Ex Ante Restrictions Prohibiting Features Such as Fast Lanes on the Variety of Content on the Internet and Content Provider Investment
The literature gives mixed results. Most articles find thatex ante regulations lower content value when some content providers value the features that ISPs would offer absent the regulations.
Hermalin and Katz (2007) finds that ex ante regulations reduce the number and variety of content providers because regulations eliminate content providers that would prefer a lower quality service. It also finds that the regulations induce some content providers to buy a higher quality service than they would prefer, and others to purchase lower quality service than they would prefer. This lowers economic efficiency for those content providers.
Bourreau et al. (2015) finds that ex ante regulations result in lower content innovations. Reggiani and Valletti (2016) finds that ex ante regulations lower the amount of content provided by a large content provider if its advertising fees are low relative to its content production costs and high relative to the smaller content providers’ production costs. Greenstein et al. (2016) describes how an ISP might price the enhanced service in such a way as to exclude some lower-end content providers if lower-end and higher-end content providers would both pay for terminating their content.
Peitz and Schuett (2016) examines a situation where a monopoly ISP serves two types of content providers: Some that offer content whose quality suffers if delivery is delayed, while others whose quality is not sensitive to delay. The paper assumes that content providers can make investments to improve traffic management. The paper finds that ex ante regulations cause content providers to underinvest in technology that could efficiently manage the amount of traffic they put on the network. As a result, strict ex ante net neutrality restrictions often lead to socially inefficient allocation of traffic and traffic inflation in this paper.[38]Ex ante net neutrality regulations that would allow some differences in traffic treatment by ISPs do not necessarily improve the situation.
3.4 The Effects on Total Welfare of Regulations Prohibiting ISPs from Charging Content Providers for Terminating Traffic
This question has received little attention in the literature. The two papers reviewed find that the effect depends upon how ISPs charge consumers and the value that customers obtain from content.
Musacchio et al. (2009) examines a situation with multiple ISPs of equal size, but each is a monopoly for its customer base. There are multiple content providers. It finds that ex ante regulations against such charging for terminating traffic lower total welfare in situations where consumers’ price sensitivity to prices for clicks is low or high (as opposed to nearly the same) relative to content providers’ charges for advertising. It also finds that the regulations can improve total welfare in situations where consumers’ price sensitivity is nearly the same as content providers’ charges for advertising.
Njoroge et al. (2014) finds theex ante regulations lower total welfare in situations where an ISP imposes fees on content providers that connect through another ISP. Greenstein et al. (2016) finds that restricting ISPs from charging termination fees reduces total welfare.
3.5 The Welfare Effects of Regulations Prohibiting ISPs from Blocking Content
Few economists have investigated this issue. In general the papers find that blocking content providers that customers want to access decreases welfare. Ex ante prohibitions on blocking improve total welfare in Economides and Hermalin (2012). Greenstein et al. (2016) confirm this result.
3.6 The Welfare Effects of Regulations Prohibiting ISPs from Blocking Access to Rivals
Broos and Gautier (2017) examines a situation where there are up to two ISPs that compete for customers. The ISPs offer traditional phone service and internet service, and the phone service competes with an app that offers voice services over the internet (VoIP). It finds that the ISPs have incentives to exclude VoIP competes with the ISPs’ own services. A monopoly ISP may want to exclude a competing internet app if the app is of inferior quality relative to the ISP’s own service and if the ISP cannot ask for a surcharge for its use. However, this is not always profitable. In both the monopoly and duopoly situations, prohibiting the exclusion of the app and surcharges for its use has ambiguous effects on welfare.
3.7 The Effects of Regulations on ISP Prices
Regulation of ISP prices is rarely considered a net neutrality issue in the US. Indeed the FCC in its 2015 decision explicitly stated that it was not proposing to regulate prices. However, the agency did assert its authority to do so.
Gans (2015) and Gans and Katz (2016) address the pricing issue in situations where consumers pay for content. The 2015 paper finds that if ISPs can price discriminate on the basis of content in setting prices consumers, then ex ante restrictions do not affect the distribution of welfare. The 2016 paper explains that this conclusion may hold even if all prices are regulated, but identifies conditions under which price regulation does affect profits and consumer surplus, namely when the regulator sets caps on fees that an ISP may charge content providers and when consumers vary in their content preferences. In general the imposition of price controls lower economic efficiency.
3.8 Selected Literature from Non-Economic Journals
This subsection examines often-cited literature from non-economic journals.
Krämer et al. (2013) provides a survey of literature in economic, policy, and computer journals, examining issues of market power, traffic management, differentiated quality of service, content delivery networks, and termination fees. The paper concludes[39]
Irrespective of the regime, the majority of the papers that conduct an economic analysis find that strict NN [net neutrality] regulation is warranted only under very special circumstances. The main assumptions that drive pro NN results are that (i) innovation at the edge is more important than innovation at the core (e.g. van Schewick 2007), (ii) the introduction of QoS [quality of service] tiering will inevitably lead to a re-congestion effect (e.g. Economides & Hermalin, 2012) and that (iii) CSPs [content service providers] have less market power than ISPs. With respect to the latter, the extant literature that models the Internet as a two-sided market has thus far ruled out that large CSPs may also be the recipient of additional revenues in a termination fee model (e.g. Economides and Tåg 2012), because no access ISP would be able to attract customers without access to these CSPs.
Krämer and Wiewiorra 2012 consider a situation with monopoly ISPs and multiple content providers that receive revenue only from advertising. The content providers do not compete as each receives the same amount of consumer traffic regardless of the content provided. The paper examines the effects of an ISP offering different qualities of service to content providers and finds that doing so may improve efficiency in the short run relative to an ex ante prohibition, because it better allocates the existing network capacity. Differentiated service quality may also improve efficiency in the long run by incentivizing greater investment.
Kourandi et al. (2015) consider a situation with two competing ISPs and two competing content providers. They examine exclusive arrangements between ISPs and content providers (called fragmentation) and find that exclusive arrangements reduce competition over ads among content providers. Incentives to engage in such arrangements are greater when competition over ads among content providers is more intense, revenues from advertisements are very low, content provider services are highly complementary, or termination fees are high. Consumers always benefit from an absence of exclusive arrangements, but the effects of ex ante prohibitions on total welfare are mixed. The paper concludes that “regulatory interventions may be ineffective or even detrimental to welfare and are only warranted under special circumstances.”
3.9 Conclusions from Academic Literature
In general, economic research on net neutrality has found that the impacts of regulations depend on the conditions in the marketplace. Under various conditions such regulations can be harmful to consumers, harmful to network providers, harmful to content providers, or hinder investment. But there are also conditions under which opposite effects might occur. Most but not all of the articles conclude that regulatory restrictions on ISPs offering advanced features can lower economic efficiency. A notable exception is blocking customer access to content that customers find valuable. Such blocking is generally found to be harmful, based on the assumption that consumers do not want blocking.
What should readers make of studies with conflicting findings? Each study relies on specific assumptions about technologies and markets. The assumptions drive the findings. The challenge for readers who want to apply these studies is to determine which conditions are most applicable in their situations. Because circumstances can change and vary across markets, ex post regulations should generally be preferred when the welfare effect of an action depends upon circumstances.
4 Conclusion
The US debate about net neutrality has been unusually contentious for a telecommunications regulatory issue, especially given that all sides appear to have the same stated objective. The substantive difference appears to be over whether to apply ex ante or ex post regulations. On this, the economics literature gives varying answers regarding how ex ante and ex post regulations differ in their effects on welfare and investment. The assumptions of the models are key. But when the answer to the question of the effects of ex ante regulation is “it depends,” and the scenarios that give different answers are realistic, it would seem that the policy decision should be in favor of addressing problems when they occur because ex ante regulations would clearly do some harm and ex post regulations, properly conducted, would do little harm. That appears to be an unacceptable answer for regulatory advocates.
Perhaps the persistence of contention arises from the way the issue entered the regulatory landscape. The US has a long history of dealing with the anticompetitive conduct of AT&T’s regulated monopoly. The company took what could be viewed as extreme measures to protect that monopoly, including denying customers the opportunity to place even a plastic Hush-A-Phone cup on their telephone receivers to block noise. The body of regulatory activity against AT&T provided the intellectual foundation for that Wu (2003) and others built upon for net neutrality, including device connection standards, drawing bright lines between telecommunications and enhanced services, and regulating wholesale pricing.
The advocates for an ex post approach, beginning with Yoo (2004), appear to adopt a post-monopoly mindset, emphasizing the dynamics of rapidly changing technologies, the value of customer choice, and diversity of situations.
Given the partisan nature of the issue, what happens next is likely to depend more on elections than on analysis. Regardless, further research could inform later policy actions. Of particular value would be research on rules of thumb for when ex post interventions should occur. Some spokespersons for small content providers express concern regarding the cost and time delays of FCC ex post interventions. Research that provided the agency with insights as to when it should investigate, the evidence that it could expeditiously review, and how it can proceed on a timely basis would seem valuable. Also of interest would be how the vacillations and politics of regulatory actions since 2014 have affected industry and consumers. Finally, guidance on how the country could achieve a more substance-driven approach to net neutrality issues could be the most important contribution of all.
References
@matthewberryfcc (2018) August 22, 2018, https://twitter.com/matthewberryfcc/status/1032365032225497088 (accessed August 24, 2018).10.1044/leader.PPL.23082018.22Search in Google Scholar
@SenateDems (2018) https://twitter.com/SenateDems/status/968525820410122240. February 27, 2018 (accessed August 24, 2018).Search in Google Scholar
Bourreau, M., F. Kourandi and T. Valletti (2015) “Net Neutrality with Competing Internet Platforms,” Journal of Industrial Economics, 63:30–73.10.1111/joie.12068Search in Google Scholar
Brennan, T. (2016) “Is the Open Internet Order an ‘Economics-Free Zone’?” The Free State Foundation: Perspectives from FSF Scholars, 11(22). http://www.freestatefoundation.org/images/Is_the_Open_Internet_Order_an_Economics_Free_Zone_062816.pdf (accessed August 27, 2018).Search in Google Scholar
Broos, S. and A. Gautier (2017) “The Exclusion of Competing One-Way Essential Complements: Implications for Net Neutrality,” International Journal of Industrial Organization, 52:358–392.10.1016/j.ijindorg.2017.03.003Search in Google Scholar
Caves, K. W. (2012) “Modeling the Welfare Effects of Net Neutrality Regulation: A Comment on Economides and Tåg,” Information Economics and Policy, 24:288–292.10.1016/j.infoecopol.2012.07.002Search in Google Scholar
Chen, A. (2018) “Why is Net Neutrality a Political Issue? Democrats & Republicans Won’t Settle For A Compromise,” Bustle. May 16, 2018, https://www.bustle.com/p/why-is-net-neutrality-a-political-issue-democrats-republicans-wont-settle-for-a-compromise-9099979 (accessed August 27, 2018).Search in Google Scholar
Cherry, B. A. (2003) “Utilizing ‘Essentiality of Access’ Analysis to Mitigate Risky, Costly and Untimely Government Interventions in Converging Telecommunications Technologies and Markets.” CommLaw Conspectus, 11:251–275.Search in Google Scholar
Choi, J. P. and B. Kim (2010) “Net Neutrality and Investment Incentives,” RAND Journal of Economics, 41:446–471.10.1111/j.1756-2171.2010.00107.xSearch in Google Scholar
Choi, J. P., D. Jeon and B. Kim (2015) “Net Neutrality, Business Models and Internet Interconnection,” American Economic Journal: Microeconomics, 7:104–141.10.1257/mic.20130162Search in Google Scholar
Cooper, S. L. (2018) “State Executive Orders Reimposing Net Neutrality Regulations Are Preempted by the Restoring Internet Freedom Order,” The Free State Foundation: Perspectives from FSF Scholars, 13(5). http://freestatefoundation.org/images/State_Executive_Orders_Reimposing_Net_Neutrality_Regulations_Are_Preempted_by_RIF_Order_020218.pdf (accessed August 24, 2018).Search in Google Scholar
Economides N. and B. E. Hermalin (2012) “The Economics of Network Neutrality,” The RAND Journal of Economics, 43:602–629.10.1111/1756-2171.12001Search in Google Scholar
Economides N. and J. Tåg (2012) “Network Neutrality on the Internet: A Two-Sided Market Analysis,” Information Economics and Policy, 24: 91–104.10.1016/j.infoecopol.2012.01.001Search in Google Scholar
Fung, B. (2014) “Protesters Descend on FCC Chairman’s House Over Net Neutrality,” The Washington Post. November 10, 2014, https://www.washingtonpost.com/news/the-switch/wp/2014/11/10/protesters-descend-on-fcc-chairmans-house-over-net-neutrality/?utm_term=.31ad2ee3b82d, (accessed August 27, 2018).Search in Google Scholar
Fung, B. (2018) “Why the Internet is Suddenly Protesting on Net Neutrality all Over Again” The Washington Post. May 9, 2018, https://www.washingtonpost.com/news/the-switch/wp/2018/05/08/senate-democrats-have-been-pushing-a-big-vote-on-net-neutrality-all-year-its-here/?noredirect=on&utm_term=.d451194aefda (accessed August 27, 2018).Search in Google Scholar
Gans, J. S. (2015) “Weak Versus Strong Net Neutrality,” Journal of Regulatory Economics, 47:183–200.10.3386/w20160Search in Google Scholar
Gans, J. S. and M. L. Katz (2016) “Weak Versus Strong Net Neutrality: Correction and Clarification,” Journal of Regulatory Economics, 50:99–110.10.1007/s11149-016-9305-7Search in Google Scholar
Greenstein, S., M. Peitz and T. Valletti (2016) “Net Neutrality: A Fast Lane to Understanding the Trade-offs,” Journal of Economic Perspectives, 30:127–150.10.3386/w21950Search in Google Scholar
Hauge, J. and M. Jamison (2014) “Do Common Carriage, Special Infrastructure, and General Purpose Technology Rationales Justify Regulating Communications Networks?” Journal of Law, Economics, and Organization, 10:475–493.Search in Google Scholar
Hermalin, B. E. and M. L. Katz (2007) “The Economics of Product-Line Restrictions with an Application to the Network Neutrality Debate,” Information Economics and Policy, 19:215–248.10.1016/j.infoecopol.2007.04.001Search in Google Scholar
Hurwitz, G. (2017) “The Legal Road Ahead for Net Neutrality and the Restoring Internet Freedom Order” AEIdeas. http://www.aei.org/publication/the-legal-road-ahead-for-net-neutrality-and-the-restoring-internet-freedom-order/ (accessed August 24, 2018).Search in Google Scholar
“Net Neutrality Advocates Are Modern-Day Snake Oil Salesmen,” Investor’s Business Daily. November 22, 2017. https://www.investors.com/politics/editorials/fcc-ajit-pai-net-neutrality-repeal/ (accessed August 27, 2018).Search in Google Scholar
Jamison, M. A. and R. Layton (2016) Beyond Net Neutrality: Policies for Leadership in the Information, Computing, and Network Industries. Washington, DC: American Enterprise Institute.10.2139/ssrn.3008914Search in Google Scholar
Kang, C. (2018) “Man Arrested for ‘Threatening to kill’ Ajit Pai’s Family Over Net Neutrality,” Independent. June 30, 2018. https://www.independent.co.uk/news/world/americas/net-neutrality-threaten-to-kill-ajit-pai-family-fcc-us-justice-a8424256.html (accessed August 27, 2018).Search in Google Scholar
Katz, M. (2016) Comments at “The Future of the Internet Ecosystem in a Post-Open Internet Order World: Part 1,” Technology Policy Institute symposium January 12, 2016. https://www.youtube.com/watch?v=Gz4 W0sDxkfs&t=834s, at 13:40 (accessed August 27, 2018).Search in Google Scholar
Krämer, J. and L. Wiewiorra (2012) “Network Neutrality and Congestion Sensitive Content Providers: Implications for Content Variety, Broadband Investment, and Regulation,” Information Systems Research, 23:1303–1321.10.1287/isre.1120.0420Search in Google Scholar
Krämer, J., L. Wiewiorra and C. Weinhardt (2013) “Net Neutrality: A Progress Report,” Telecommunications Policy, 37:794–813.10.1016/j.telpol.2012.08.005Search in Google Scholar
Kourandi, F., J. Krämer and T. Valletti (2015) “Net Neutrality, Exclusivity Contracts, and Internet Fragmentation,” Information Systems Research, 26:320–338.10.1287/isre.2015.0567Search in Google Scholar
Layton, R. (2018) “Are Crafty Tactics the Way to Make Internet Policy?” AEIdeas. http://www.aei.org/publication/are-crafty-tactics-the-way-to-make-internet-policy/ (accessed August 24, 2018).Search in Google Scholar
Lyons, D. (2018) “California’s Net Neutrality Law: Will it Survive Judicial Review?” AEIdeas. http://www.aei.org/publication/californias-net-neutrality-law-will-it-survive-judicial-review/ (accessed October 6, 2018).Search in Google Scholar
Mak, A. (2017) “The FCC Had to Pause Its Net-Neutrality Hearing Because of a Bomb Threat” Slate. December 14, 2017, http://www.slate.com/blogs/future_tense/2017/12/14/fcc_vote_on_net_neutrality_was_interrupted_by_a_security_threat.html (accessed August 27, 2018).Search in Google Scholar
Musacchio J., G. Schwartz and J. Walrand (2009) “A Two-Sided Market Analysis of Provider Investment Incentives with an Application to the Net-Neutrality Issue,” Review of Network Economics, 8: 1–18.10.2202/1446-9022.1168Search in Google Scholar
NDTV (2018) “In Net Neutrality Protests, US Regulator Ajit Pai Bombarded With Pizzas,” NDTVhttps://www.ndtv.com/world-news/protest-outside-us-telecom-regulator-head-ajit-pais-home-over-net-neutrality-1780769, (accessed August 27, 2018).Search in Google Scholar
Njoroge, P., A. Ozdaglar, N. E. Stier-Moses and G. Weintraub (2014) “Investment in Two-Sided Markets and the Net Neutrality Debate,” Review of Network Economics, 12:355–402.10.1515/rne-2012-0017Search in Google Scholar
Obama, B. (2014) Statement on Net Neutrality. November 10, 2014, https://obamawhitehouse.archives.gov/the-press-office/2014/11/10/statement-president-net-neutrality.Search in Google Scholar
Peitz, M. and F. Schuett (2016) “Net Neutrality and Inflation of Traffic,” International Journal of Industrial Organization, 46:16–62.10.1016/j.ijindorg.2016.03.003Search in Google Scholar
Powell, M. K. (2004) “Preserving Internet Freedom: Guiding Principles for the Industry” Silicon Flatirons Symposium. February 8, 2004, https://apps.fcc.gov/edocs_public/attachmatch/DOC-243556A1.pdf.Search in Google Scholar
Reggiani, C. and T. Valletti (2016) “Net Neutrality and Innovation at the Core and at the Edge,” International Journal of Industrial Organization, 45:16–27.10.1016/j.ijindorg.2015.12.005Search in Google Scholar
van Schewick, B. (2007) “Towards an Economic Framework for Network Neutrality Regulation,” Journal on Telecommunications & High Technology Law, 5:329–392.Search in Google Scholar
Wallsten, S. (2016) The Partisan FCC. Technology Policy Institute. https://techpolicyinstitute.org/2016/02/16/the-partisan-fcc/.Search in Google Scholar
Wu, T. (2003) “Network Neutrality, Broadband Discrimination,” Journal of Telecommunications and High Technology Law, 2:141–180.10.2139/ssrn.388863Search in Google Scholar
Wu, T. and C. Yoo (2007) “Keeping the Internet Neutral? Tim Wu and Christopher Yoo Debate,” Federal Communications Law Journal, 59:Article 6.Search in Google Scholar
Yoo, C. S. (2004) “Would Mandating Network Neutrality Help or Hurt Broadband Competition? A Comment on the End-to-End Debate,” Journal of Telecommunications and High Technology Law, 3:23–68.Search in Google Scholar
©2019 Walter de Gruyter GmbH, Berlin/Boston
