Introduction
While John Perry Barlow’s Declaration of Independence of Cyberspace might have suggested that the internet was a magical space that transcended borders and governments, the internet has very real and physical infrastructure. Whatever flavor of wireless access one uses to access the internet, the data we send and receive from our devices eventually runs through physical cables. An examination of the governance of this physical layer of the internet can shed some light on the power and control of the global internet — how is it managed, by whom; to serve what purpose, for whom? This article aims to highlight the problematic nature of a profit-driven approach to internet infrastructure, and the value of treating the internet as a utility, on state and global levels.
Cables
In the early days of the internet, existing telephone infrastructure was used to transmit data. One would make a data call to an Internet Service Provider (ISP) to get connected. These connections would keep the line busy and were easily interrupted and disconnected. This type of connection employed a circuit-switching paradigm, where point-to-point connections were made over a fixed path — a telephone conversation would establish a connection between the caller and receiver, that would remain connected until the caller hung up. Dial-up evolved to DSL, or Digital Subscriber Lines, where voice and data signals would use different frequency bands, allowing one to simultaneously use the same connection for internet and telephone calls, and allowed for faster data rates. DSL employed a packet switching paradigm, where instead of creating a fixed point-to-point connection, data was broken up into packets that specified their destination. This allowed for multiple subscriber lines to be aggregated, which was more appropriate for data transfer, which generally occurs in spurts.
Infrastructure set in place for cable television provides another method for data transfer — cable internet uses the higher bandwidth of coaxial cables for faster data transmission. The latest method of transferring data is through fibre optic cables. Narinder Singh Kanapy, the ‘father of fibre optics’, discovered how data can be transmitted by bending light in glass tubes. Fibre optics have gained popularity due to their reliability, speed and low-latency — data travels at the speed of light.
Understanding the costs
For data to reach your device, cabling runs from your house to your ISP, and then to an Internet Exchange Point (IXP), and then potentially through an undersea cable (depending on where you are the data you are requesting are). (More on that here). The infrastructural costs can be broken down into 4 components (source):
- Local access (you to the ISP): the fixed, mobile or WiFi connection to the national network.
- National backhaul (network to network domestically): the link from the access point to the international gateway.
- International transit (network to network internationally, on land): In some instances, there is no immediate access to the an international gateway, such as with landlocked countries (and occasionally for others); in those cases there is a need to purchase capacity on another country’s network to reach the cable landing station.
- Submarine cable (network to network internationally, undersea): investment in or lease of capacity from the submarine cables that lands in the country.
As a consumer, one is mostly faced with the costs of Local access. “The closer you get to the home, the more investment is needed, averaged per home connected. This applies to all parts of the physical network, like water pipes, sewage pipes, and electricity cables” (source). This is often known as the ‘last-mile’ connection, and there have been numerous cases in the US where residents have been quoted amounts ranging from $5000 to $383,500 to bring fibre to their homes. In India, one of biggest inhibitors to laying cables is earning the ‘right of way’, where cities often use high costs for digging up roads as a source of income (source). A land-locked country like Mali that relies on access to undersea cables through its neighbor Senegal, is subject to rates set by telecom monopolies in Senegal — in 2007 this was twice the cost of transmitting the data from Senegal to Europe (source). Building codes often factor into last mile connections too – in countries like India, lax building codes allow for aerial cables to enter consumers’ houses through windows and balcony doors. In the US, running cables on existing utility poles can also be expensive in cities that don’t control their own poles. (source)
Who pays?
The governance and business models for internet infrastructure differ from country to country, and an examination of a few notable examples can shed light on the different approach and their consequences.
In the US, in 1985 The National Science Foundation took over the development of the internet from the Department of Defense. NSFNET backbone services were provided to support open research and education, and use for other purposed was strictly prohibited. However, in the early 90s, as the internet grew in popularity, it became clear it was an opportunity for profit, and the US began privatizing the internet’s operations to fuel growth through competition. Telephone and cable companies were required by law to rent out their lines to ISPs who could compete to provide better service at lower prices. This separation of physical access and delivery of service is referred to as an open access network. However, in 2002, the FCC passed regulation to allow cable companies exclusive rights on their cables (source), (and telephone companies shortly after), resulting in a telecom cartel — a handful of companies that don’t compete with each other.
This cartel-like structure has led to artificially inflated prices for consumers, where big telecom providers are not incentivized to expand infrastructure to less-dense areas due to low returns on investment.
These quasi-monopolies don’t just opt-out of laying fibre — they also exert their power to stop municipal efforts to build their own networks. The American town of Chattanooga in Tennessee has become known for having some of the fastest internet speeds in the world, connecting the entire town. By using the existing electric infrastructure, the government-owned electricity company, EPB, helped wire the entire town with a loan from the government, at no additional cost to taxpayers (source). The internet is also amongst the cheapest in the US, being free from the inflated prices dictated by telecom cartels. This has not gone unnoticed by telecom providers however — they lobbied for laws to prevent municipalities from building out internet. “This year, the Tennessee state legislature was finally considering a bill that would have let EPB expand its coverage (without providing it any special tax breaks or grants; EPB is profitable and doesn’t rely on taxpayer money). Rather than pass that bill, Tennessee has just passed the Broadband Accessibility Act of 2017, which gives private telecom companies—in this case, probably AT&T and Comcast—$45 million of taxpayer money over the next three years to build internet infrastructure to rural areas. …. Tennessee will literally be paying AT&T to provide a service 1000 times slower than what Chattanooga could provide without subsidies.” (source). Tennessee isn’t an exception — the below map (source) shows the 25 states where municipal-fibre is restricted [as of 2018 California was no longer on the list, see the updated report here].
The flyer below illustrates Comcast’s efforts to appeal to the citizen of Batavia, Illinois to stop municipal broadband efforts (source).
Strong telecom monopolies are not just in the US — Africa has more ‘consolidated markets’ (ie, lacking competition) than every other continent globally, “the average price for one gigabyte of data costing 7.12% of the average monthly salary—much higher than the defined affordability benchmark of 2% of average income.” (source). A 2018 report by Ecobank Research shows that countries with two operators tend to have higher prices for mobile data than countries with more competitors.
On one hand, it can be argued that competition is necessary to provide cheap access to the internet. However, there is a distinction between lowering prices and increasing penetration, and the huge upfront capital costs in building infrastructure can be prohibitive for private entities, leaving little room for competition in expanding access. The infrastructural requirements for the internet closely resemble that of utilities such as electricity and water, which are often co-financed and regulated by the state to ensure they serve the public good.
Susan Crawford, in her book Fiber, defines the idea of a utility — “the basic idea of utility is a service the relies on physical infrastructure of some kind, and is a basic input into both domestic and economic life. A utility is not a luxury. Can be sold by private or public entities, but they are always subject to public obligations to reach everyone at a reasonable price with a service meeting public quality standard”.
The principle of reaching everyone is at the core of a utility, and it is why some countries have excelled at connecting their populations, regardless of the number of competing telecom players. A 2017 report that examines fibre deployment in European countries, identifies some of the factors leading to high-penetration:
In Estonia, the “Tiger Leap” project (1995) set to “wire every school to the Internet and to provide PCs to each classroom, all across the 50,000-square-mile territory” (source). A 2009 public-private partnership — the EstWin project — set out to connect all Estonians with a 100Mbit/s connection, by building of a country-wide middle-mile-network, that connected the existing fibre backbones with local access points such as schools and libraries. Private companies then built the last-mile connections and were granted exclusivity in their respective areas. 75% of the project was funded by the European Commission and the State of Estonia. This initiative was part of Estonia’s “Digital Society Strategy 2020” – contributing to the country’s status as the most modern digital state in the world, with 99% of public services available online (source)
In Sweden, a 2000 action plan to create “an information society for all” delegated private companies responsible for the rollout of fibre. However, with the Dotcom bubble collapse, they were unable to finance the construction, and the state stepped in with large investments to lay down a national backbone. As of 2017, municipalities own 60% of the fibre market, with just three private entities owning the rest. The infrastructure and service layers are separated, following an open access network approach, where public bodies will often lay down passive fibre that service providers can then lease. This model works since public bodies can afford longer time-frames to refinance their investments, as opposed to private enterprises. As cited in the report, “In Sweden, municipalities have transferred the concept of”services for the public” to broadband Internet: As with water or electricity, the provision of broadband Internet is considered to be a public responsibility in Sweden (Mölleryd 2015)”
In the period between 2012 and 2016, Spain went through a “fibre miracle” — an extraordinary increase in the rollout of fibre-to-the-premises, from ~10% to ~50%. In the cities, this was realized through regulatory holidays for the incumbent Telefonica, who during this phase, rolled out highly successful bundled plans that prompted competitors to lay out their own fibre networks in order to compete. Telecommunication lines are often laid above ground in Spain, which facilitated this massive fibre rollout. Outside of the cities, fibre deployment was co-financed by the state and often managed by municipalities, working towards a coverage goals defined by the Spanish government in 2013 . Once the regulatory holiday was lifted, private companies had to open their infrastructure to allow for competition at a service level.
Through private-public partnerships, the active role of municipalities, open access networks and ambitious coverage goals, these European countries have managed to increase high-speed internet coverage.
Non-monetary consequences
The consequences of a profit-driven approach to internet infrastructure stretch beyond financial burdens on consumers and digital divides in internet access. In the US, telecom companies own several media/content brands (illustrated in the image below). The 2018 roll back on net-neutrality allows these private actors to control information flow by employing zero-rating practices — ie, allowing certain content to be available for free. As articulated by Ingrid Burrington, “There’s something stomach-twisting about a company offering free video streaming but charging for the data used when filling out an online job application.” (source). Studies have shown that countries that allow for zero-rating have higher costs for access, by charging premiums to access all content (source). The decoupling of infrastructure and service (ie open access networks) could create healthy competition where it would be disadvantageous for companies to provide selectively free content — or at the very least, put the choice in consumers hands.
It is not uncommon for ISPs to collect and store data flowing through their cables, as well as metadata on customers’ activities, governed by data retention laws that differ from country to country. While some countries declare it as unconstitutional, the US currently does not have data retention laws, and in 2017, privacy protections were repealed, allowing ISPs to sell data to third-parties — compromising the privacy of consumers for the profit of the ISPs, who also run advertising networks. “AT&T has AdWorks, while Verizon has the blandly-titled Relevant Mobile Advertising and Verizon Selects, which it melded with its AOL Advertising Network in 2015” (source).
A company more frequently associated with profiting from consumers data – Facebook – is capitalizing on weak infrastructure in developing countries by rolling out their own zero-rated internet, known as free basics. As of 2019, it has been rolled out in 65 countries (source). It has been referred to as an
“African Dictator’s Dream” (source) by creating a way of controlling the internet, and has put little emphasis on local content. The image below highlights a search on Free Basics in Ghana, where the government’s website cannot be accessed. (source)
“Facebook is not introducing people to open internet where you can learn, create and build things,” said Ellery Biddle, advocacy director of Global Voices.“It’s building this little web that turns the user into a mostly passive consumer of mostly western corporate content. That’s digital colonialism.” (source). Zero-rated practices such as these can create a tiered internet, which can amplify digital divides on a global scale. The global impact of content services providing internet access, as seen with Free Basics, is even more concerning given the shifting landscape of undersea cable ownership. As of 2018, Google, Facebook, Amazon and Microsoft owned or leased more than half of the undersea bandwidth in 2018 (source). Would the same problems of profit-driven control of infrastructure by content providers seen on a national level in the US expand to a global scale?
Conclusion
While the high capital costs of internet infrastructure can act as a deterrent for private companies to expand coverage, public-private partnerships and open access networks can facilitate the deployment of high-speed internet. Though it can be argued that treating the internet as a utility would hinder healthy competition, the infrastructural similarity to utilities cannot be ignored. “The fact that the infrastructure is being built with public money does not mean that the competition principle is abolished. To the contrary, Open Access Networks enable competition on the service level in areas where previously no competition was possible because there were no networks available.” (source)
The prioritization of private interest and profit set the US apart from countries like Estonia and Sweden, where nation-wide access and coverage are priorities, dictated by ambitious coverage plans. By viewing internet access as a utility rather than a luxury, these countries have excelled in connecting their populations, and avoiding digital divides within their borders.The prioritization of private interests leads to digital divides not just in terms of connectivity, but also information, through practices such as zero-rating; as well as privacy issues resulting from data-retention policies. Efforts such as Facebook’s Free Basics serve the interests of local telecom companies by reducing their financial burden to create infrastructure, while creating further information divides and fueling a form of digital colonialism. With this precedent, the shifting landscape of undersea cables towards mostly American content providers, is even more concerning.
Tim Wu, in The Master Switch, argues that information empires have always followed a cycle — “from somebody’s hobby to somebody’s industry; from jury-rigged contraption to slick production marvel; from a freely accessible channel to one strictly controlled by a single corporation or cartel– from open to closed system“. However, the global nature of the internet sets it apart from past information empires, and poses a new set of challenges in global governance. The landscape is not all bleak however, as multiple players have made efforts to address fair and open access to the internet on state and global levels. The multi-stakeholder Internet Governance Forum was set up in 2006 to address issues such as governance of undersea cables (source). Mozilla’s Internet Health Reports provide state and global level information of the health of the internet. ICANN (The Internet Corporation for Assigned Names and Numbers), that is responsible for managing and coordinating the unique identifiers (ie domains, or the URLs you put into a browser), was globalized in 2016, ending a 20 year contract with the US Department of Commerce, to better serve the global nature of the internet. In 2015, regulatory authorities in Southern Africa pushed “to open up access to undersea cables to all countries in the region in order to bring down the high cost of communications in the area” (source). The European Commission recently laid down a goal to have the entire EU connected with high speed internet by 2025. Tim Berners-Lee, who is considered the inventor of the world wide web, recently put forth a Contract for the Web, a set of principles designed to “save the web from political manipulation, fake news, privacy violations and other malign forces that threaten to plunge the world into a “digital dystopia” “ (source).
[This article was written in January 2020 — things have probably changed since then. Perhaps The coronavirus crisis proves the internet should be a public utility after all]