NET NEUTRALITY, SMALL BUSINESS, AND BIG GOVERNMENT

Download the .pdf of the full report

We should remember that the current leaders in Internet innovation all began with essentially nothing. Google, eBay, Yahoo! and Amazon all started as simple websites providing limited, but fantastic, services. …That competition would be
threatened by access-tiering.1 — Larry Lessig


The quote by Larry Lessig has become a rallying cry in the debate over
“net neutrality,” a fuzzy concept most generally meaning nondiscriminatory and
open access to the Internet. More specifically, many oppose the notion of
access-tiering, or charging different rates for different services provided.
Proponents of net neutrality seek to expand the Federal Communications
Commission’s regulatory scope over the Internet through new mandates that
would define and regulate access to the internet.

Proponents of net neutrality seek to expand the Federal Communications Commission’s regulatory scope over the Internet through new mandates that would define and regulate access to the internet.   Although advanced under the banner of internet freedom, these proposals require unprecedented new government authority and price controls on internet service.  Open access entails a full suite of regulation, from defining and enforcing nondiscrimination to monitoring and regulating prices to ensure there is no discrimination in the provision of services. In effect, the internet would become a common carrier, replete with all the maladies of a regulated industry.  In practice, net neutrality may yield unintended consequences that generate benefits for some while imposing substantial costs on others.  Most importantly, price controls would push all costs to end users in the name of net neutrality.  Unfortunately, these price controls would have the effect that price controls always have on the workings of the market—less output, less innovation, and less investment in the infrastructure critical for the deployment of broadband.  Much has changed in how the internet is used, yet new mandates threaten to lock in an architecture that may limit the deployment of new technology.

Wither Innovation? 

An important claim in this debate that often goes unchallenged asserts that entrepreneurship and innovation will be stifled without new mandates for net neutrality.  In essence, without government intervention, innovation on the internet will dry up.  Despite the significance of this claim, there has been scant evidence offered in its support.  At the same time, this claim ignores evidence on the dilatory effects of regulation on innovation and entrepreneurship.  Turning the internet into a common carrier—which concerns over tiered pricing and open access ultimately require—is fraught with its own concerns of regulatory delay, federal oversight, and political interference.[2]

 Undoubtedly, corporations such as Google that support net neutrality have been significant innovators and benefit from the current pricing scheme, but locking in a “neutral” network creates costs as well.  If hospitals are willing to pay for guaranteed service at specified speeds, or if gamers are willing to pay for a better managed network to improve the performance of their products, would consumers be better off?  Are Virtual Private Networks, which provide an added layer of security over standard internet communications, violating some sense of net neutrality?  When such experimentation is disallowed on the internet, innovation of a different sort is limited by mandating new content and applications use a specific technology and run in the same network as spam and downloads (legal and illegal) of the latest Harry Potter movie.   

Christopher Yoo, of Vanderbilt University Law School, suggests that network differentiation, rather than network neutrality, could increase consumer welfare.[3]  Already, retailers, content providers, and applications developers are experimenting on the web, pushing technology to its limits.  Just as significantly, they are also experimenting with pricing for content and applications—much of which would be challenged by the stated principles of net neutrality. Moreover, it is difficult to decide the “correct” price for the network without experimentation.  Should broadband consumers pay none, part, or all of the costs related to deploying, maintaining, and improving the infrastructure of the internet?  How much of these costs should be borne by end users relative to content or application providers?  Newspapers, for example charge readers very little for the daily paper, recouping the costs from advertisers.  Would a similar approach be efficient for the internet?

It is a challenge to answer these questions ex ante.  The market system is the most reliable mechanism available for allocating scarce resources and establishing efficient prices.  But net neutrality advocates would discard this system in favor of a regulatory allocation of resources.  In this paper we examine the internet as it has emerged as well as the tenets that underlie mandated net neutrality.  We also examine the impact of new mandates on innovation and entrepreneurship, as well as potential adverse effects on consumers.  Finally, we attempt to identify the concerns that may arise when attempting to operationalize a net neutrality mandate.  Based on regulatory history, current and future web usage, and the costs of government regulation, we suggest that net neutrality mandates would reduce social welfare, harming both consumers and businesses.  

Innovation and the Internet  

The internet is clearly one of the largest transformational events of the last 100 years, generating what some have referred to as an e-revolution, with an economic impact akin to the industrial revolution of the 19th century.  In fact, the internet has touched almost every aspect of the economy, expanding markets, boosting productivity, and enhancing consumer welfare.  One study notes, “The deployment of internet business solutions has yielded to date a cumulative cost savings of $155.2 billion to U.S. organizations. U.S. organizations that are currently deploying internet business solutions expect to realize more than $500 billion in cost savings once all internet solutions have been fully implemented by 2010.”[4]     

It is not intuitively obvious that net neutrality mandates are required to continue this progress.  Indeed, as new technologies emerge and broadband expands to an even greater population, the internet is showing its age.  Streaming video is replacing the static web page, and real quality of service (QoS) issues are developing.  Net neutrality mandates effectively freeze the internet’s development at a stage that may be inappropriate for future use patterns.  In an age of BitTorrent and IPTV, it is not surprising that new tools for network management may be required.[5]  One study notes, “Greater bandwidth and processing power alone will not solve all congestion and QoS problems on the Internet. This is because the Internet involves the use of scarce resources, and when treated otherwise, theory and evidence suggests that congestion will become a problem, undermining convergence and the development of services that require superior QoS statistics.”[6]            

Net neutrality proponents, on the other hand, are wary of new pricing mechanisms for better network management.  Instead, they propose to regulate access, something that may limit innovation.  A simple survey of the marketplace suggests the potential impact of regulation.  Since the deregulation of the market for telecommunications, there has been an explosion of products available for the end consumer.  The regulatory era was marked by a limited set of choices for consumers, with most consumers having access to little more than a simple rotary phone.  However, beginning with the Carterphone decision in 1968, the FCC eased regulatory restraints on third party hardware connecting to the phone network and the result was a rush to market of new products, from fax machines to answering machines to any number of telephones, all of which was marked by significant decreases in price.[7]  

  At the same time, a more dynamic internet creates more opportunities for entrepreneurs.  A faster and more agile internet paves the way for more powerful applications and a larger audience.  It must be remembered that those upgrading the internet can only generate profits by providing a product that consumers are willing to purchase.  Investments in the future of the internet are not made to deter consumers; rather, expanded deployment and better traffic management are efforts to bring more consumers online.  This encourages more participation by consumers and more opportunities for developers of both content and applications.   Far from killing the next Google, then, a more developed internet offers a host of new opportunities for the next generation of entrepreneurs.  Importantly, it would allow entrepreneurs and developers to take full advantage of the latest standards to bring consumers a better online experience.  Indeed, the latest web applications such as YouTube and BitTorrent require the full capabilities of high-speed broadband access.  

What is the Internet?            

Today’s internet is the product of yesterday’s technology, and its age is beginning to show.[8]  The internet was created to move data—disassembling, moving, and reassembling bits and bytes—that did not necessarily require the same demanding standards that current and future applications require.  As many have noted, the internet was not developed as a network per se, rather; it was a means of interconnecting a number of smaller networks.  These early networks were typically Ethernets, and the early internet was fundamentally a project connecting these various networks.       
      
Data is transferred across the internet in packets, with data broken down and moved forward using any route available, only to be reassembled upon final delivery to the end user.  Routing these packets of information required new tools, or protocols, such as TCP/IP as a means of controlling traffic on the internet.  This is the basis for much of today’s network management.  The IP, or internet protocol, moves data packets forward in a “best effort” approach, with no guarantees that they will arrive or they will arrive in order.  The transmission control protocol, or TCP, has the function of creating and then reassembling data packets at the destination in the correct order.    

Since its inception, the internet has evolved into a means of global communications.  Email and text bulletin boards dominated much of the early traffic, and eventually a file transfer protocol became popular as a means of transferring larger amounts of data.  The development of the visual browser and the World Wide Web in the mid-1990s pulled mainstream America online.   More recently, new applications that allow sharing photographs, audio, and video content have attracted even more users who, in turn, run significantly more data-intensive applications.  Today, there are more than one billion users worldwide—almost 17 percent of the world’s population. (See Figure 1.)  Growth continues to be strong, and since 2000, use has increased by more than 208 percent.[9]   

For connectivity, the internet relies on a high-speed backbone that connects internet Service Providers (ISPs) to one another, allowing users to go anywhere on the network.  The ISPs typically use a regional provider that links to the backbone. The backbone, in turn, is comprised of a multitude of high-speed networks and routers, which allow data packets to move across the country and across the globe.  These networks are highly redundant, and internet protocols allow packets to find numerous paths to the end user in real time.             

Owners of the internet backbone negotiate with one another to establish commercial agreements to cover the transfer of data.  Interestingly, many of the transactions between networks are done without fees.  Networks with similar loads of traffic establish “peering agreements” in which they will transfer data for one another at no cost.  If there are significant asymmetries in the data loads, with one network sending far more data across another than it receives in exchange, the network owners will use commercial agreements to come to terms.            

The development of the internet differs in important ways from earlier telecommunications networks.  Earlier networks relied on centralized hubs, which connected individuals in a point-to-point fashion.  The goal was a quality connection between one user and another, and managing the network meant investing in the switches and central stations that made this possible.  The switches remained on as long as one party was connected to another. 

The internet, on the other hand, is not a means of simply connecting one individual directly to another.  It connects an individual user to a vast number of sources at once, anywhere in the network.  The goal is to allow any computer at the edge of the network to send and receive data from any other computer anywhere in the network.  Notably, the internet developed initially to move data and text files, rather than large video or audio files; as such, it did not need the sophisticated load balancing, latency management, and overall bandwidth of today’s internet communications.    

An End Run around the Middle? 

Early on, internet development focused on allowing those on the edge to effectively share data and applications, the center of the network—the pipes and wires—were viewed as secondary, serving only to route data and provide connectivity to those on the edge.  This principle is known as end-to-end connectivity, with the goal of establishing all intelligence at the edge of the network; the pipes providing the connection were viewed as “dumb.”   Advocates for net neutrality uphold the end-to-end principle as a policy goal.  Neutrality means that the internet merely transfers data while applications at the edge manage the data (e.g., spam filters, etc.) and determine how the data is used (e.g., email, web browser, etc.).  They see the creative component of the internet exclusively at the edge; the pipes are simply a mechanism to connect these islands of creativity.  Yet a significant amount of innovation already occurs within the pipes in order to route data packets at higher speeds and lower costs.  Increasing traffic and new applications highlight the need for even better traffic management, and ensuring QoS requires an even greater dose of intelligence for what are often assumed dumb pipes.  

By definition, mandated dumb pipes would functionally operate as a common carrier.  Not only does this limit innovation, but it may also reduce competition in the “last mile,” the very concern that fuels much of today’s debate.  In a competitive market, investments will only be undertaken if there is a reasonable chance to generate a return on the capital outlay.  Innovation and experimentation allow broadband providers to offer new opportunities for networks and deployment.  Dumb pipes, on the other hand, are an undifferentiated commodity, which reduces incentives to deploy in an area already served by an incumbent broadband provider.  In today’s market, it is the new capabilities of high-speed broadband, including faster internet connections, video programming, and internet telephony, that are driving deployment.  Providers are competing in terms of both speed and service, seeking to offer consumers a unique and enjoyable online experience.  WiMax providers offer the additional benefit of mobility, another characteristic of demand that many consumers rate highly.  However, mandates that reduce the network to a commoditized service limit the margins for competition, reducing incentives to invest in critical infrastructure.  Consequently, net neutrality requirements could generate greater concerns over market power than exists in today’s internet, while making QoS and pricing experimentation difficult. 

Given the emphasis on simple emails and file transfers in the early years of the internet, latency and jitter (delays and problems in the data at the end of the pipe) were not significant problems.   However, the shift from moving text files to streaming video and internet telephony has highlighted the importance of quality of service: “As originally envisaged, the Internet was not designed for services with real time requirements…. [A]s opposed to the traditional telephone network's use of switched circuits that are opened for the duration of a telephone conversation, the Internet uses datagrams that may have other datagrams just in front and behind that belong to different communication sources and are destined to terminate on a different host.”[10]  Unlimited bandwidth, once thought to be a solution, cannot address the more fundamental problems of managing packets of information.  Quality of Service (QoS) is a legitimate concern, and an attribute that many consumers demand.           

Although proponents of net neutrality endorse the “end-to-end” view of the internet, it is a view not shared by all.  A joint statement by several leading thinkers endorses the notion of a smarter internet: “The current Internet supports many popular and valuable services. But experts agree that an updated Internet could offer a wide range of new and improved services, including better security against viruses, worms, denial-of-service attacks, and zombie computers; services that require high levels of reliability, such as medical monitoring; and services that cannot tolerate network delays, such as voice and streaming video. To provide these new services, both the architecture of the Internet and the business models through which Internet services are delivered may have to change.”[11]  

What Would Net Neutrality Regulate?            

The internet can be viewed as a series of layers—content, applications, logical, and physical—each performing a distinct function.  Proponents of net neutrality focus on the physical layer that underlies the network, including the wireline, wireless, cable, and satellite services that transport information. They contend that the physical layer should become simply a regulated transport provider, with additional steps taken to ensure that monopoly power is not leveraged into activities that take place in higher layers, such as content or applications.  Those providing the physical infrastructure would not be allowed to discriminate between packets or show any preference for content or applications they provide.[12]  Any service made available to one content provider must be made available to all in a nondiscriminatory manner, which suggests that prices could not vary for different content providers.  The wires, switches, and cables that move information would be regulated to limit market power and ensure access that allows consumers and content providers to exchange information on an equal basis. This physical layer would be further divided into last mile access regulations and internet backbone regulations.  There are numerous backbone providers and there have been few concerns raised about market abuse on the backbone.  Last mile access is at the center of the controversy over net neutrality. 

The major concerns of net neutrality advocates center on the fear that a monopoly broadband provider in the physical layer could charge application providers for connecting them to consumers, or that the broadband providers could expand into providing content and set prices to favor their content over others.

The logical layer, which is just above the physical layer, is the code, or architecture, of the network, such as TCP/IP and other protocols, and would engender its own regulations, such as whether data packets can be prioritized. Above this is the application layer, which is the software that provides users an interface, such as e-mail or a web browser. The final layer is the content layer, which provides the information viewed on the web page, or the video or data stream relayed to the end user.  The major concerns of net neutrality advocates center on the fear that a monopoly broadband provider in the physical layer could charge application providers for connecting them to consumers, or that the broadband providers could expand into providing content and set prices to favor their content over others.  To avoid such scenarios, there are two key provisions included in net neutrality mandates.  First, broadband providers would not be able to tier access to the internet.  All content and applications would pay the same fees to move across the wires; no “fast lane” could be established to move certain data at more quickly.  Second, broadband providers would not be able to prioritize data, or move certain packets to the head of the queue when transporting data.  All packets of information are to be treated equally as they are routed to the end user.  

Where’s the Monopoly?            

Net neutrality supporters typically point to the threat of monopoly in this last mile as a primary reason to adopt new regulations. Yet, studies by the FCC suggest broadband internet usage is increasing at a significant pace.  In the first half of 2006, for example, the increase was 26 percent, bringing the total number of lines in use to more than 64 million.[13] (See Figure 2.) Most of the nation live in zip codes that have 2 or more broadband providers, and more than 75 percent of the population has three or more choices, and these numbers are improving.[14]  Figure 2Total High Speed LinesSource: Federal Communications Commission, “Local Telephone Competition and Broadband Deployment,” available at http://www.fcc.gov/wcb/iatd/comp.html.  

Additionally, emerging technologies should further allay fears of anticompetitive behavior among broadband providers.  New providers offer alternative avenues to consumers, with WiMax creating the potential of wireless networks covering entire towns or cities.  Other technologies under development, such as broadband over power lines (BPL), also hold promise for future deployment and new competitors.  In fact, DirecTV has announced that it is already experimenting with this new technology.[15]            

Blocking particular websites or limiting web access, common assertions of made by net neutrality advocates makes little sense in a dynamic and expanding market.  Competition requires providing consumers with what they are seeking: the best online experience possible.  For broadband providers, then, blocking websites or limiting access does not increase broadband penetration.            

Moreover, it must be remembered that large content and application providers may exhibit a degree of market power themselves.  For, example, if Verizon tried to block a popular site such as Amazon or Google because they could not come to terms on pricing, consumers may abandon Verizon in favor of a broadband provider that does connect to sites that consumers demand, which raises an important issue.  Namely, where does the monopoly threat end?  If, in fact, Google, with more than half the market for internet searches, does have market power, is regulation required in the name of net neutrality?  Similarly, even content providers may possess market power, with a provider such as ESPN able to extract surplus from ISPs.   

            A firm in any layer can expand to provide new services on the internet, as has been witnessed by Google’s jump from the application layer to the content layer with the $1.65 billion purchase of YouTube.[16]  The major concern is whether a firm can leverage market power in one layer into market power in another layer.  Proponents of net neutrality assert that broadband providers have monopoly power in the last mile that could be used to create market power in the content layer.  But Google’s search engine had a market share of over 64 percent; indeed, in March 2007, the two top search engines have a market share just over 75 percent, according Nielsen/Net Ratings, but it would be difficult to demonstrate that this warrants new regulations.[17]  All of these situations can be addressed by existing antitrust laws should abuses occur.  These are phenomena that are general to any market and no legislation or special rules are warranted.     

Google has leveraged itself into the content market with YouTube and other services.  With respect to advertising, the search giant is clearly engaged in a two-sided market, matching advertisers to consumers using algorithms to extract as much surplus as possible.  While this may be an example of an application that is expanding into the content market, it does not seem to suggest that Google is abusing market power.  Similarly with broadband providers, in the absence of evidence of abuse, new regulations are not required.  The existing antitrust laws are more than ample to address any such concerns.   As the online world emerges, questions of market power may arise that cut across all market participants.  Net neutrality has, to date, focused primarily on the physical layer.  Yet market power can exist elsewhere.  The diversity of potential sources of market power suggests that the existing antitrust laws embody a degree of flexibility that makes them more aptly suited to addressing any problems than would an ex ante, proscriptive net neutrality regulation.  

Network Effects, Pricing, and Two-Sided Markets            

To properly understand broadband fees and assess the existence of anticompetitive practices, it is important to understand how the market for broadband services operates.  Broadband is an example of what recent economic literature has described as a two-sided market.[18]  In a two-sided market, two different groups interact with each other through a shared platform.  For example, a newspaper provides a platform for both readers and advertisers, and a broadband provider offers a platform for internet users and content providers hoping to attract customers. 

Unlike the standard market of economic analysis, the two-sided market raises new questions about costs and pricing.  In a standard market, prices are set to cover the costs of all inputs, as well as normal returns for the entrepreneur whose capital is at risk.  But the two-sided market adds a new element to the analysis.  Namely, how does the platform allocate costs between the two markets?  In the newspaper example, the reader pays a very low price for the product, while revenues from advertisers are a more important stream of income.  However, what the newspaper can charge the advertisers is a function of how many readers can be claimed.  In effect, the question of pricing becomes an exercise in joint maximization, where both sides of the market must be considered simultaneously.            

This interaction between the two sides of the market is the result of what economists call “network effects” and makes the optimal pricing strategy a more difficult issue than the standard market.  In short, the value of the network increases as the number of people who use the network increases.  Consider the simple example of a phone, which is only valuable because it allows the user to connect with others on the network.  The more users, the more value the phone provides.  It is the same with high-speed broadband internet service.  The greater the access to content, the greater the benefits created for the end user.  But in this two-sided market, the reverse is true as well: the greater the number of end users, the more attractive it becomes to provide content on the network.  The fundamental question then becomes how to maximize the overall value of the network to both content providers and consumers.              

Prices on each side of the market affect the other side, which means the costs of the network must be allocated across both sides of the market in a way that increases the value for everyone.  As Rochet and Tirole note, “Platforms [such as high-speed internet providers] must perform the balancing act between the two sides along various policy dimensions and not only with respect to the price structure.  They therefore often regulate the terms of transactions between end-users, screen members in non-price related ways and monitor intra-side competition.  In all instances, they sacrifice profit by constraining one side to boost attractiveness for and recoup losses on the other side.”[19]            

From the broadband provider’s perspective, this means balancing both sides of the market and establishing prices to optimize the overall value of the network.  These prices are tempered by competition between broadband providers and with alternative forms access to content.              
Competition occurs on both sides of the market.  Consumers are looking for providers offering universal access to the worldwide web at the best price.  Content providers, on the other hand, are looking for access to the largest number of end users at the lowest price.              

Clearly, both sides of the market have different demands that must be addressed.  Prices are set by broadband providers based on the economic characteristics of demand by both consumers and providers.  One important component for establishing the appropriate rates is the elasticity of demand on both sides of the market.  This is nothing more than a measure of how sensitive both sides of the market are to price changes.  Using this joint methodology, broadband fees could be set at a point where more revenues are generated from content providers rather than end users. 

            In the end, the fee should attempt to reflect the value of the service being offered to content providers and end users, who both clearly receive a great number of benefits from the availability of a broadband network.  However, some of the largest providers of content, such as Yahoo and Google, oppose attempts to optimize prices across a two-sided market.  In effect, they propose the costs of broadband deployment and maintaining the network be collected exclusively from end users.  At a hearing on net neutrality, Vinton Cerf, of Google, testified before the Senate: "The broadband carriers are fully compensated by their residential customers for their use of the network. These companies can charge their own customers whatever they want, in order to make back their investments."[20] In an analysis of various pricing strategies, Gregory Sidak notes that content providers are proposing a particular pricing policy for the web that focuses on end users. [21]  He looks at pricing bandwidth according to two variables: bandwidth and priority.  Tiered pricing can use either of these.  There is little debate over the former, as all parties generally agree that all users—content providers and consumers—should pay for the bandwidth they use.  Yet tiered pricing may, in fact, be a beneficial activity and should not be viewed as a form of discrimination.  Importantly, Alfred Kahn notes that such measures are not, in fact, price discrimination in an economic sense.  Price discrimination entails charging different prices for the same good, based on individual elasticities.  But in this instance, they are not the same good.  Charging more for a higher tier of service is not price discrimination, no more than charging different prices for apples and oranges.[22]            

Fees charged by broadband providers attempt optimize the value of the network.  However, quantifying the benefits of the network to content providers is often difficult.  This has led many, who would prefer to pay less for access to end users, to call for cost-based regulation, which is nothing more than price controls for broadband providers.  Both tiered access and packet prioritization would be prohibited.  Historically, price controls have performed poorly, generating market distortions and market inefficiencies.  These dangers are even greater in a two-sided market, where the economic theory is still unsettled.    

Increasing Loads Create New Challenges for the Network 

As the internet has become more ubiquitous, how it is being used is also changing, making concerns over network management even more important.  The shift from email to BitTorrent has put new strains on the internet that will only increase as more users and more bandwidth-intensive applications become available.  One recent study found that BitTorrent, a peer-to-peer network application, was already responsible for more than 30 percent of internet traffic by the end of 2004.  Overall, peer-to-peer traffic is currently responsible for more than 60 percent of internet traffic and poses real concerns for internet service providers, and can affect QoS for all subscribers.[23]  Internet telephony and IPTV also loom on the horizon, and as many have noted, the amounts of data being moved across the internet raise significant concerns over carrying capacity.       

Moving forward, then, two tasks are required to create the networked world of the future.  First, broadband providers must continue to invest in the fundamental infrastructure of the internet.  Doing so requires investors to allocate billions of dollars to extend networks to reach rural and other customers.              

In addition to a wider deployment of broadband, there is also a growing need for improved network management.  Bandwidth-intensive applications and content are placing new strains on the internet that threaten to diminish the benefits for everyone.  Not only is more bandwidth required, but that bandwidth must be managed more efficiently.  New technologies for traffic shaping and deep packet inspection provide new management tools for the internet, allowing broadband providers to more efficiently allocate network resources.  Net neutrality mandates, however, would limit the applicability of these tools.            

Net neutrality mandates would generate a typical commons problem.  That is, whenever prices are set too low, there is a predictable outcome.  Economics and history tell us that a good whose price is set below a market-clearing level will lead to a shortage in which the quantity demanded will exceed the quantity supplied.  By eliminating tiered pricing and packet prioritization, net neutrality mandates would lead to a world of over-used, clogged pipes.  For example, net neutrality mandates could lead to a world in which spam e-mail has equal access to the clogged pipes as a patient’s EKG required by a heart specialist.  As in any situation with an under-priced good, there is little, if any, incentive to innovate, expand, or upgrade.

            This is especially true when examining the emerging applications such as IPTV, gaming, telephony, distance learning, and medical technologies, all of which require more data to be moved with greater levels of effort to ensure quality of service.  And it is not as if these applications are replacing many of standard uses of the internet, such as email and file sharing.  These activities remain popular and will actually expand as broadband deployment brings more people online.  The new applications and social networking under development are an expansion on top of the existing uses of the internet, requiring even more data to be delivered to an expanding set of applications and increasing number of users.           

  In fact, the term “exaflood” has been coined to describe the crush of information that is coming to the internet.  An exabyte is one billion gigabytes, and the exaflood raises the challenge of how the internet will manage the surge in information.  One study found that “in 2006 some 161 exabytes of digital information were created and copied.  That amount is equal to three million times all the books ever written.  According to the IDC [International Data Corporation], the amount of information created and copied in 2010 will surge more than six fold to 988 exabytes—a compound annual growth rate of 57 percent.”[24]   

Net Neutrality, Defined  

Despite the highly public debate over net neutrality, there is perhaps little consensus over its meaning among its loudest proponents.  Some couch the debate in terms of internet freedom, claiming the debate is about limiting individual access to various sites on the web.  Others emphasize the role of tiered access, and whether broadband providers should have the ability to market higher speeds or better QoS for specific content or applications.   

The unifying element of all these definitions, however, is a larger role for the FCC.  Interested parties are requesting the FCC impose new standards on pricing and access to the internet.  What net neutrality ultimately means, therefore, would be determined through a rulemaking procedure, with federal regulators attempting to meet the demands of the various interested parties.             

The mandates advocated to keep the net neutral all lock in some idealized notion of the internet that would mark a significant expansion of the FCC’s regulatory authority.  Although the Internet, up to this point, has evolved largely in a market-based setting, the new mandates pose a potential threat to all businesses using the Internet, particularly with respect to broadband deployment and upgrading the Internet to handle new data-intensive applications.            

From a regulatory perspective, federal policies toward the internet can be traced to a speech by then FCC Chairman Michael Powell where he discussed net freedom.  In this speech, Powell noted that the internet was free, and it was in the self-interest of broadband providers to maintain its openness.  He also noted that there was little evidence of market power problems: Based on what we currently know, the case for government imposed regulations regarding the use or provision of broadband content, applications and devices is unconvincing and speculative. Government regulation of the terms and conditions of private contracts is the most fundamental intrusion on free markets and potentially destructive, particularly where innovation and experimentation are hallmarks of an emerging market. Such interference should be undertaken only where there is weighty and extensive evidence of abuse.[25]   Since that speech, there has been little evidence to suggest that “net freedom” is in jeopardy.  However, Chairman Powell challenged broadband providers to ensure that the internet would remain free in the future, and he outlined the principles by which that freedom could be measured.  With some clarification, the Federal Communications Commission, under new Chairman Kevin Martin, released a policy statement on internet freedom in 2005 that included four main principles: 

  • Consumers are entitled to access the lawful Internet content of their choice.
  • Consumers are entitled to run applications and services of their choice, subject to the needs of law enforcement.
  • 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.[26]
  • Requiring firms to share new innovations with their competitors would significantly reduce investments in new technologies.  For consumers, this means a delay in expanding the size of the network, and a delay in adopting new technologies.

    More importantly, the Policy Statement also asserted that the FCC had the authority to ensure that the internet was operated in a neutral manner.  Nonetheless, advocates of net neutrality find these principles wanting, and are seeking legislation to strengthen these principles.  In particular, they hope to add a fifth principle on “nondiscrimination” that would prohibit distinctions between data packets.  Nondiscrimination is most clearly described in the Network Neutrality Act of 2006, introduced by Rep. Ed Markey (D-Mass.).  The legislative language adds a precision to the debate that gives far more detail to the concept than the generic slogans of the movement’s advocates.  The legislation notes, “if the broadband network provider prioritizes or offers enhanced quality of service to data of a particular type,” then the broadband provider has the duty to “prioritize or offer enhanced quality of service to all data of that type (regardless of the origin of such data) without imposing a surcharge or other consideration for such prioritization or quality of service”[27] Despite assertions that net neutrality is about access and not network management, this language would have significant implications for the internet.  Indeed, Richard Bennett, an early pioneer of the internet, notes, “On the technical side, my objection to the ‘Net Neutrality’ bills (Markey, Snowe-Dorgan, Sensenbrnner [sic], Wyden) is the ban on for-fee Quality of Service (QoS).  QoS is a legitimate offering, especially in the day of BitTorrent and what’s to follow it….The End-to-End is fine for error recovery in file transfer programs, not so fine for congestion control in the interior links of the internet.  For the latter, we need QoS, MPLS, and address-based quotas.”[28] Perhaps the most notable effect would be the disincentives created for investing in new technologies or expanded broadband deployment.  Requiring firms to share new innovations with their competitors would significantly reduce investments in new technologies. The 1996 Telecommunications Act provides a useful analogy. In the spirit of promoting competition, the law required incumbent service providers to unbundle their networks and allow their newly created competitors—the competing local exchange carriers (CLECs)—access at rates determined by the FCC.   Unfortunately, this did little to create new competitive networks.  Rather than invest and build facilities to compete with the incumbents, the CLECs simply used the incumbent’s networks at the regulated prices, which were often set at below-market rates.[29]  As a result, CLECs relied increasingly on regulated access rather than investing in new facilities to compete with the incumbents. At the same time, the incumbents have little incentive to expand their networks under such conditions, because any investments in the network are immediately available to competitors at prices that do not necessarily allow the incumbents to recoup their investments.  For consumers, this means a delay in expanding the size of the network, and a delay in adopting new technologies.

    Without an opportunity to recoup such investments, tomorrow's internet may look very similar to today’s, from a technological perspective, despite the very real changes in how the internet is being used.  

    In much the same way, when mandates require any updates that enhance or prioritize the ability to move information across the internet to be immediately available to anyone using the network, there is a reduced incentive to invest in better network management.  Enhanced quality of service or tiered access that prioritizes the delivery of specific packets of data may no longer be available as an option, which can have important consequences for the operation of the internet.   Proponents of net neutrality are always quick to point out that the new requirements do not prohibit network management.  This may be true, but the more fundamental issue is whether the incentives for improving network management exist in a world where any such changes are immediately available to everyone on the network at the same price.  Without an opportunity to recoup such investments, tomorrow’s internet may look very similar to today’s from a technological perspective, despite the very real changes in how the internet is being used.  To date the internet has evolved relatively free from federal regulation.  This flexibility has created an important resource that continues to evolve to meet a growing demand among consumers and businesses. 

    Moreover, that demand has generated new products that may require new tools for network management.  It is not in the public’s interest to give the government the job of controlling this evolution.  Whether pipes are dumb or smart should be determined by those using the network, as should decisions on handling the burgeoning flow of information over the internet.   It is the users that will determine the most efficacious and efficient structure for the internet—consumers, broadband providers, internet service providers, and applications and content providers.  These relationships have been defined in the marketplace, and given the absence of market power, there is little reason to interfere.              

    It should be remembered that federal antitrust laws continue to provide significant regulatory oversight of the market, even in the absence of new net neutrality mandates.  In the most significant example of anticompetitive behavior relating to internet access, it is worth noting that the FCC immediately intervened, as did Canadian authorities.  An ISP, Madison River, was blocking Vonage on its network, and regulatory authorities promptly issued a fine and forced the ISP to allow access to its network.[30]   

    Despite the availability of federal and state antitrust laws to address anticompetitive behavior, many still advocate ex ante regulatory authority to promote net neutrality.  Proponents of this view suggest antitrust enforcement is burdensome and slow, making it difficult to apply to a fast-changing technological world.[31]  It must be remembered, however, that regulatory lags can be just as slow, if not slower, as interested parties challenge regulatory oversight at both state and federal agencies and within the courts.  Significant components of the 1996 Telecommunications Act are still shrouded in legal uncertainty.  Cases have been challenged up to the Supreme Court before resolution, and the case of unbundled access, for example, was only resolved in 2005.  Given current antitrust enforcement and the structure of the market, the burden of proof lies with those advocating new mandates.            

    The FCC has already issued its Policy Statement on internet freedom—something that will guide its behavior in future rulemakings.  In addition, the agency has used its enforcement powers to curb market power on the internet, as evidenced in the Madison River case. Moreover, given the fact that broadband is no longer treated as a common carrier, the Federal Trade Commission has also asserted its ability to address concerns over market power.  Nothing has changed in the broadband market to suggest additional federal action is required.  Indeed, if anything, broadband deployment has expanded, prices have fallen, and new entrants have joined the market.  These are hardly signs of market failure, or a need for new legislation or additional government intervention.  

    Small Business and the Internet

    Technology and innovation have allowed small businesses to challenge big business in ways never possible before.  Mass customization has strengthened niche markets, offering alternatives to products generally dominated by big businesses.

                 
    The technology revolution has been a driving force in boosting productivity and expanding economic output in the United States.  Currently, more than 91 percent of American business have broadband.[32]  One study notes that, “…in the United States IT was responsible for two-thirds of total factor productivity growth between 1995 and 2002….and virtually all of the labor productivity growth.”  Perhaps one of the most significant beneficiaries of the internet revolution has been small business.  The internet has provided new opportunities to expand the market and reach new customers, create new niche markets, and outsource traditional activities such as payroll and accounting.  New opportunities for using the internet are constantly being developed.  In many ways, technology and innovation have allowed small businesses to challenge big business in ways never possible before.  Mass customization has strengthened niche markets, offering alternatives to products generally dominated by big businesses.              

    Small business has the potential to be a large winner due to increased broadband deployment.  In his book Long Tail, Chris Anderson argues that the internet allows for extreme niche markets that small (and large business) can exploit.[33]  Through lower search costs, both retailers and consumers create markets not limited by geography.  Amazon, eBay, and many others have exploited this opportunity very well.  But it is not just the large firms who can profit from the long tail markets; small firms are serving these markets as well.  For instance, independent music bands are able to reach potential fans who can listen to music free (or for a small fee) that would never be heard without the ability to share audio files over the internet.             

    Such niche markets are a good example of Adam Smith’s famous proposition more than 200 years ago that “the division of labor is limited by the extent of the market.” Broadband deployment has allowed connections between disparate buyers and sellers to create markets in virtually anything, which is why, for example, listeners can hear many music artists that they may not hear otherwise.  The wide reach of the market and cheaper distribution costs have provided numerous small businesses an expanding set of opportunities.            

    Yet these opportunities require a fast and reliable internet, especially when transferring content such as music, videos, or online gaming.  While such uses clearly require more bandwidth than other internet transactions, they compete for space with all data transfers and must queue together to reach end users.  More bandwidth and better network management are enhancements that would help all small businesses—not just those with audio or video content—and foster greater internet commerce.  Net neutrality mandates may inhibit such improvements, to the detriment of all business.

    More bandwidth and better network management are enhancements that would help all small businesses—not just those with audio or video content—and foster greater internet commerce.  Net neutrality mandates may inhibit such improvements, to the detriment of all business.

               
    Small businesses do not have an easy time competing with larger firms, but the internet provides new avenues for that competition.  As Anderson notes, firms like Netflix and Amazon, with “heir huge libraries of less-mainstream fare set them apart, but hits still matter in attracting consumers in the first place.  Great Long Tail businesses can then guide consumers further afield by following the contours of their likes and dislikes, easing their exploration of the unknown.”[34]  Again, these businesses are filling in voids that larger businesses do not capture, but doing so requires a reliable network.            

    One example of this is Pandora, a startup online company that asks consumers to list bands they enjoy and then offers users similar, unknown music they might enjoy. How does Pandora “know” what kind of music listeners like? Pandora runs what they euphemistically call the Music Genome Project. They have listened to tens of thousands of songs from thousands of bands, which they break down by musical attributes, such as melody, arrangement, lyrics, vocal harmony, and hundreds of other attributes.  Based on the bands and songs listeners select, Pandora starts recommending songs with similar musical attributes on “radio stations” the user sets up.  Listeners give feedback for each song Pandora thinks they will enjoy, allowing Pandora to learn what music to provide to different listeners.             

    It is likely that the biggest winners will not be famous Top 40 bands, but more obscure artists that have little marketing or corporate muscle behind them.  Pandora creates an opportunity in the Long Tail to be profitable, and there are similar opportunities for other audio and video sites.  Already, there is competition in this market through firms like last.fm, and, of course, people can still go to Amazon, iTunes, and many other outlets.  But a fast and reliable internet makes real competition possible between the corporate giants and smaller upstarts that can take full advantage of the latest technology on the web.             

    Although there have been efforts to mobilize small businesses in favor of net neutrality mandates, regulations that inhibit the deployment of broadband and new technologies may disproportionately harm small businesses.  Indeed, surveys suggest that a far greater concern for small business with respect to the internet is a reliable connection.  In fact, one survey of small businesses found that “84 percent said that a less reliable Internet with slower speeds and periodic outages would have a negative impact on their business—including 56 percent who said it would have a major negative impact.”[35]            

    Another 2007 survey by Wells Fargo shows the enormous benefits that small businesses gain from the use of the Internet.  Not inconsequential is the response of small business owners when asked what would occur if they lost internet access: More than one-third of the firms thought the loss of Internet connectivity would have a major negative influence on their firm, with 18 percent stating it would put them out of business. [36] 

    From the survey results it is clear that small business finds Internet connectivity important for success, with one-fifth expecting to fail because of a loss of internet access. The implication is that, at a minimum, the reliability of the Internet is a main concern to the small business person.             

    This massive increase in broadband use will put enormous pressure on the backbone of the Internet. Capacity will be strained to the limits with users requiring streaming video, internet telephony, potential medical usage, and so forth. This demand will require both more bandwidth and better management of what is becoming an increasingly scarce resource.  Even more importantly, newer applications and content require constant “jitterless” internet connectivity. The price controls and mandates inherent in net neutrality would have a detrimental effect on investment in broadband and inherently limit its value to small business users.    

    Myth of Net Neutrality Today

    The internet is home to businesses of all sizes, and small businesses compete head-to-head with some of the largest corporations in the world.  And while it is true that the internet has benefited all businesses, those benefits are not necessarily distributed evenly among all firms.  In fact, there are many aspects of today’s internet that may provide more benefits to larger corporations rather than small businesses, which suggests that even on the current internet, neutrality is a relative term, creating benefits for some and costs for others.            

    Under the end-to-end principle of the internet, as mentioned previously, intelligence is located at the edge of the network.  All the applications and content required to run a business are located at the edge.  This means that the costs of doing business on the internet are located at the edge as well.  Businesses, large and small, must acquire and maintain the content and applications necessary for online business.  Not only does this include accounting, marketing, payroll, and other business operations applications, but also security and data management.  In this regard, large enterprises can utilize applications that improve the internet in ways that many small businesses cannot afford. 

    >

    There are many aspects to today’s internet that may provide more benefits to larger corporations rather than small businesses, which, suggests that even on the current internet, neutrality is a relative term, creating benefits for some and costs for others.

    A reconfigured internet could move some of these activities to the network itself, reducing the number of applications an individual company must acquire to do business online.  Security, for example, is an issue that all firms must address, and small businesses are particularly vulnerable to malicious attacks on the internet, because they often cannot afford to hire a full-time information technology staff.[37]  Worms, viruses, and malware pose a constant threat to business, but most large corporations have the resources to defend themselves against such potential problems.              

    For a smaller firm, allowing the broadband provider to perform these tasks may be more cost effective.  Building these tasks directly into the structure of the internet may benefit consumers as well.  But larger enterprises, with in-house IT staff may be able to more effectively address these concerns on their own.  Determining the appropriate solution is something best left to a market, which provides the flexibility to address a number of different configurations.   But net neutrality mandates could limit the levels of security that broadband providers are allowed to offer and ultimately raise the costs of doing business on the internet.           

    Another application used by many big businesses allows them to store data closer to end users, speeding the delivery of data over the internet.  Content delivery companies such as Akamai cache information throughout the globe to enhance the delivery of data and improve QoS.  Rather than have all data query go back to the corporation’s server, the information can be accessed much more quickly from a server located closer to the user.  While very effective in improving internet data requests, the service is not inexpensive.  Most small businesses cannot afford the service, leaving bigger businesses another advantage in the internet marketplace.  This is not to say that such practices should be banned; rather, they have been a valuable contribution to the internet.  But broadband providers may be able to provide customers similar traffic management solutions; arbitrarily prohibiting such behavior does not improve the internet.            

    Tools that allow businesses to optimize their use of the internet clearly add value.  Net neutrality mandates, however, place tight strictures on who can add what value to the internet.  In the absence of evidence of market power abuse, these constraints are unnecessary and may prove costly for different users.  Current federal and state antitrust laws already provide the power to address any concerns of market abuse, without limiting the pattern of innovation on the internet.  

    Non-Net Neutrality On-Demand

    Net neutrality mandates, however, place tight strictures on who can add what value to the internet.  In the absence of evidence of market power abuse, these constraints are unnecessary and may prove costly for different users.

                Advocates of net neutrality argue that broadband providers should not have the right to charge for improved access or special handling that moves specific packets closer to the front of the queue.  But the Japanese NTT’s DoCoMo service suggests consumers and content providers may find real value in more active network management.  In fact, DoCoMo is a clear example of a non-neutral service filling a valuable market niche.  Like other wireless providers, DoCoMo’s i-mode is a mobile internet platform that restricts content, offering what some refer to as a “walled garden” version of the internet that provides access and is customized to work with mobile devices.  Launched in 1999 with approximately one million customers, i-mode today has made numerous technological innovations and counts almost 50 million customers in Japan.   Yet DoCoMo has flipped net neutrality on its head by providing a very non-neutral service. Why would customers and content providers flock to such a gated environment, unless it added value? Although there are some variations, DoCoMo basically charges consumers a fixed fee and also a “packet communication charge” that is based on the volume of data transmitted. If subscription-only content, such as entertainment sites, is accessed, an additional fixed monthly fee is charged for the subscription and a per-unit fee for specific usage.[38]                       

    Although not neutral, DoCoMo plays a valuable role by simplifying the billing process as a “middleman” for transactions between content providers and consumers.  DoCoMo collects fees from consumers for content use that are then transferred to the specific content providers, while charging the content provider a percentage of revenue generated.  By being the middleman to all dealings, they lower the transaction costs for both consumers and content providers.  Establishing prices in a two-sided market made DoCoMo i-mode service successful—yet charging consumers and content providers different prices for different services is explicitly rejected by net neutrality advocates.    

    Markets or Mandates?            

    At the most fundamental level, the debate over net neutrality is about allocating scarce resources.  As with any good, there are two options, the market or the government.  Markets allow providers and consumers opportunities for innovation, sending important information signals that coordinate the activities of everyone involved.  These signals can promote investment in existing networks as well encourage consumers to utilize scarce resources more efficiently. 

    As Nobel laureate Friedrich Hayek demonstrated, the knowledge available in a market is dispersed and beyond the reach of any one individual.[39]  It is the interaction of all individuals that effectively generates the knowledge guiding behavior.  Government planners cannot capture this knowledge, making it difficult for them to replicate the efficiencies of the market.  Consequently, in regulated markets, shortages or gluts are common as regulators overshoot or undershoot their targets.  Similarly, artificial restraints on network development can have unintended consequences that distort the development of new internet technologies.            

    Providing the FCC new authority to manage the internet makes a clear choice about the allocation of resources.  The government, rather than the market would make critical choices, and regulatory rulemakings would replace prices as a means of allocating resources.  This is not only a lengthier and more involved process, it is also subject to political influence that may distort the outcome.  As Alfred Kahn states, "The process has inevitably reflected a complex mixture of political and economic considerations. Governmental price-fixing is an act of political economy."[40]            

    Net neutrality proceedings would need to establish operational definitions of what constitutes nondiscrimination and packet prioritization.  To ensure neutrality with respect to QoS, regulators may have to review and adopt standards for acceptable levels of latency and jitter; this would ensure that each broadband providers “best effort” for moving data forward is nondiscriminatory.  In effect, new mandates would require a federal agency to determine which product should be available at what price, paving the way for a long and active regulatory chapter in the development of the internet.  

    The plain choice for the future is whether to continue to allow the internet to evolve in a market-based setting, with the flexibility to address new concerns that arise about performance and deployment, or to cede oversight of the internet to a federal agency that will use the blunt tools of common carrier regulation to guide its future development.

    There have been several studies demonstrating that economic regulation actually reduces consumer welfare, in markets such as transportation, energy, and telecommunications.[41]  Such regulation may pose particular problems for the internet, a dynamic and continually evolving technology that must respond to the latest improvements in computing technology.  Regulations evolve in a much slower framework, with important decisions about resource allocation being transferred to the political process.  Rent-seeking becomes more important than innovation and entrepreneurship in the marketplace, and the political process can divert attention from concerns over efficiency to other social goals.  For example, in telecom, universal service has become an important policy driver, generating significant distortions in the market.[42]           

    There is no doubt that innovation at the edge has provided an amazing array of applications and content that has benefited consumers and businesses alike.  But we have reached the point where the activity at the edge is testing the limits of the entire structure.  The plain choice for the future is whether to continue to allow the internet to evolve in a market-based setting, with the flexibility to address new concerns that arise about performance and deployment, or to cede oversight of the internet to a federal agency that will use the blunt tools of common carrier regulation to guide its future development.  Based on the evolution of the internet to date and the history of regulation, the prudent path may be to allow the internet to continue its development with minimal regulatory interference, with an understanding the federal antitrust laws provide more than ample authority to address any market-power concerns, wherever they may emerge.

    * Wayne Brough (wbrough@freedomworks.org) is vice president and chief economist, FreedomWorks, Washington, D.C.

    ** Arthur Fleisher (fleishet@mscd.edu) is a professor of economics, Metropolitan State College of Denver, Denver, Colorado.


    [1] Larry Lessig, Testimony before the Senate Committee on Commerce, Science, and Transportation, February 7, 2006.

    [2] See. Alfred Kahn, The Economics of Regulation, (MIT Press, 1988) for a discussion of the federal regulatory process.

    [3] Christopher S. Yoo, “Beyond Network Neutrality,” Harvard Journal of Law and Technology, vol. 19, no. 1, Fall 2005.

    [4] See, for example, Hal Varian, Robert Litan, Andrew Elder and Jay Shutter, ”The Net Impact Study,” available at http://www.netimpactstudy.com/nis_2002.html[5] BitTorrent is a peer-to-peer file sharing protocol that allows the distribution of very large amounts of information across the internet.  One study found that “it only takes about 10 BitTorrent users bartering files on a node (of around 500) to double the delays experienced by everybody else.”  Cited in Leslie Ellis, “BitTorrent’s Swarms Have a Deadly Bite On Broadband Nets,” Multichannel News, May 8, 2006, available at http://www.multichannel.com/article/CA6332098.html

    [6] The Wik Consult, The Economics of IP Networks—Market, Technical and Public, Policy Issues Relating to Internet Traffic Exchange, Study for the European Commission, May 2002, p. 162.

    [7] Robert Crandall, After the Breakup: The U.S. Telecommunications Industry in a More Competitive Era, The Brookings Institution, 1991.  It should be noted that some have pointed to the Carterphone decision as a means of instituting net neutrality (for instance, Tim Wu, “Wireless Net Neutrality: Cellular Carterfone on Mobile Networks,” New America Foundation, Working Paper no. 17, ver. 2.1, February 2007.  However the analogy is inapplicable here.  The initial decision was issued in a market defined by a regulated monopolist.  Carterphone, in effect, eased regulatory restrictions for entering the market.  In today’s broadband market, such a decision, would impose a new regulatory burden on a market that the FCC views as competitive.

    [8] See, for example, Mark Handley, “Why the Internet Only Just Works,” BT Technology Journal, vol. 24, no. 3, July 2006.  Available at http://www.cs.ucl.ac.uk/staff/M.Handley/papers/only-just-works.pdf.

    [9] Internet World Stats, “World Internet Usage and Population Statistics,” available at http://www.internetworldstats.com/stats.htm.

    [10] The Wik Consult, supra, note 5, May 2002, p. 36.

    [11] David Farber, Gerald Faulhaber, Michael L. Katz, and Christopher Yoo,” joint statement on net neutrality, June, 2006, available at http://www.interesting-people.org/archives/interesting-people/200606/msg00014.html.

    [12] See, for example, Richard S. Whitt, “A Horizontal Leap Forward: Formulating a New Public Policy Framework Based on the Network Layers Model,” An MCI Policy Paper, March 2004, available at http://global.mci.com/about/publicpolicy/presentations/horizontallayerswhitepaper.pdf.

    [13] FCC, Chart 1, Industry Analysis and Technology Division Wireline Competition Bureau, “High-Speed Services for Internet Access: Status as of June 30, 2006” January 2007, Table 1.  Available at http://hraunfoss.fcc.gov/edocs_public/attachmatch/DOC-270128A1.pdf.

    [14] FCC, Industry Analysis and Technology Division Wireline Competition Bureau, “High-Speed Services for Internet Access: Status as of June 30, 2006” January 2007, Chart 15.  Available at http://hraunfoss.fcc.gov/edocs_public/attachmatch/DOC-270128A1.pdf.

    [15] See “DirecTV May Try Broadband on Power Lines,” Reuters, May 14, 2007, available at http://www.reuters.com/article/technologyNews/idUSN1433448320070514.[16] Google Press Center, “Google To Acquire YouTube for $1.65 Billion in Stock,” October 9, 2006, available at http://www.google.com/press/pressrel/google_youtube.html.

    [17] Cited in ZDNet IT Facts, Avaliable at http://blogs.zdnet.com/ITFacts/?cat=27.

    [18] Jean-Charles Rochet and Jean Tirole, “Two-Sided Markets: An Overview,” Mimeo, IDEI, Tolouse, France, March 12, 2004.

    [19] Ibid, p. 41.

    [20] Vinton G. Cerf, Testimony before the U.S. Senate Committee on Commerce, Science, and Transportation, February 7, 2006, p. 6.

    [21] J. Gregory Sidak, “A Consumer-Welfare Approach to Network Neutrality Regulation of the Internet,” Journal of Competition Law and Economics, 2006.  Available at http://ssrn.com/abstract=928582.

    [22] Alfred Kahn, “Network Neuatrality,” AEI-Brookings Joint Center for Regulatory Studies, March 2007, p. 3, available at http://aei.brookings.org/admin/authorpdfs/page.php?id=1374&PHPSESSID=a3701a64cc6321f33088accb799e16d7

    [23] Andrew Parker, “P2P in 2005,” CacheLogic Research, August 2005, available at http://www.cachelogic.com/home/pages/research/p2p2005.php

    [24] David P. McClure, “The Exabyte Internet: An Analysis of the Growing Flow of Billions of Gigabytes of Digital Data and Its Impact on Public Policy for the Internet,” US Internet Industry Association, May 1, 2007, p. 6.

    [25] Michael Powell, “Preserving Internet Freedom: Guiding Principles for the Industry,” speech delivered in Boulder, Colorado, February 8, 2004, p. 4, available at http://www.cdt.org/speech/net-neutrality/20040208powell.pdf.

    [26] Federal Communications Commission, Appropriate Framework for Broadband

    Access to the Internet over Wireline Facilities et al., FCC 05-151, CC Docket No. 02-33, available at http://www.publicknowledge.org/pdf/FCC-05-151A1.pdf, August 5, 2005.

    [27] Network Neutrality Act of 2006, introduced by Rep. Ed Markey, available at http://markey.house.gov/docs/telecomm/Markey%20Net%20Neutrality%20Act%20of%202006.pdf

    [28] Quoted in Andrew Orlowski, “How ‘Saving the Net’ May Kill It,” The Register, July 17, 2006, available at http://www.theregister.co.uk/2006/07/17/net_neut_slow_death/.

    [29] See, for example, Wayne T. Brough, “What’s Wrong with this Picture?” FreedomWorks, April 9, 2004, available at http://www.freedomworks.org/processor/printer.php?issue_id=1735.

    [30] Alfred E. Kahn, “A Democratic Voice of Caution on Network Neutrality,” Release 2.24, Progress and Freedom Foundation, October 2006, p. 2, available at http://www.pff.org/issues-pubs/ps/2006/ps2.24voiceofcautiononnetneutrality.pdf.

    [31] See, for example, Nicholas Economides, “’Net Neutrality,’ Non-Discrimination and Digital Distribution of Content Through the Internet,” AEI-Brookings Joint Center for Regulatory Studies, March 2007, p. 19.

    [32] Gary Kim, “As Broadband Access Reaches a Saturation Point, Sales Strategies Must Be Remade,” Extreme Makeover, Feb. 22, 2007, available at http://channelvisionmagazine.com/showArticle.php?id=55.

    [33] Chris Anderson, Long Tail, Hyperion, 2006.

    [34] Qutoed in Chris Anderson, “The Long Tail,” Wired, available at http://www.wired.com/wired/archive/12.10/tail_pr.html

    [35] Survey conducted by the Association for Competitive Technology, available at http://www.actonline.org/press-releases/060315-KeepInternetReliable.html.

    [36] https://www.wellsfargo.com/press/20070410_SmallBizTech?year=2007

    [37] See, for example, http://money.cnn.com/2006/06/05/smbusiness/security_threats/index.htm

    [38] See http://www.nttdocomo.co.jp/english/service/imode/about/charge.html

    [39] Friedrich A Hayek, “The Use of Knowledge in Society,” The American Economic Review, vol. 35, no. 4 (Sept. 1945), pp. 519-30.

    [40] Alfred Kahn, supra, note 2, vol. I, p. 42.

    [41] See, for example, Clifford Winston, “Government Failure Versus Market Failure,” AEI-Brookings Joint Center for Regulatory Studies Washington, D.C., 2006.

    [42] Raymond L. Gifford, “Universal Service for the Digital Age,” in A Digital Age

    Communications Act: Essays on the Need for Communications Policy Reform, Progress and Freedom Foundation, 2005, pp. 17-23.