Does the Internet Need More or Less Regulation?

Blog | July 08, 2019


The Internet is perhaps the most contested field of global power, and we are now at a critical juncture of its evolution. In the last two years, calls for stricter Internet regulation have intensified across the political spectrum. Government representatives, civil society leaders and, more recently, CEOs insist that new measures are needed for Internet markets, services, intermediaries and even for content and online behavior. Geopolitics plays a key role in this, but authority and legitimacy in this field has multiple (private) sources.

Given the shifting dynamics, increasingly observers are debating—would more regulation solve the problem? Of course, the question of imposing direct rules as an effective way of controlling the undesirable effects brought by the Internet on modern society does not capture the whole story.  

50 years ago, when the first packet-switching network, ARPANET, was built as part of a U.S. government military project, few formal rules constrained the work of the small group of researchers contracted by the Department of Defense. As Elizabeth Feinler, the creator of ARPANET’s first Network Information Center, recalls, despite some inherent political sensitivities, community members often established protocols on an ad-hoc basis:

“In the early days we put out the directory, which was sort of a phone book of the internet. And there were a lot of military people, there were a lot of graduate students, so there was a spectrum of users and developers. In the 1970s, there were [...] lots of strong feelings about the Vietnam war and what not. So I took it upon myself not to put anybody’s title in the directory, so that meant that everyone was talking to everybody and they didn’t know whom they were talking to.”

From early on, the Internet was governed—though not necessarily regulated—by power struggles, an aspect often neglected by Internet historians. The daily practices of influential community members (in particular, Internet pioneers like Steve Crocker, Jon Postel, Robert Kahn and Vint Cerf) became as important for Internet governance as the formal rules gradually put forward by governments and international organizations. In the technical community, disputes were alleviated via the establishment of anchoring routines such as the Request for Comments, multistakeholder participation or ad-hoc expert groups, which became a sine qua non for the evolution of the Internet.

But that did not solve the geopolitical tensions. The privileged role of the U.S. administration in the management of the network became a source of ongoing tension, in particular for its unilateral imposition of rules in international fora, its oversight over the Internet Assigned Names Authority functions (relinquished in 2016) and its surveillance capabilities, as revealed by Edward Snowden in 2013. Similarly, the overrepresentation of American interests in different technical and political communities gradually became a concern; at different paces, international organizations and Internet communities started enhancing the representation of developing countries and of other interests. Non-state actors, both for-profit and not-for-profit (sometimes working together), found ways to participate in international discussions, to put pressure on other players and to form supportive coalitions.

Today, more than 160 organizations are involved in Internet governance at the global level, designing more than 300 governance instruments covering the technical, economic, legal and developmental aspects of the Internet. Regulation, used extensively in telecommunications, was complemented by other important governance mechanisms in the move from deploying generic approaches to designing specific rules exclusively applicable to the network—in areas such as domain names, privacy and more recently online speech and disinformation.

The Three Mechanisms of Internet Governance

Over the past four decades, three mechanisms of governance were at play in the evolution of the Internet: legal enshrinement, institutional solidification and modelling. Their leverage varied over time, diffusing authority and diversifying the bases on which norms could be introduced. 

First, the traditional way of regulating telecommunications via legal enshrinement (treaties, conventions) had little appeal for the Internet after the mid-1990s. Yet, the role of states, courts and law enforcement bodies has grown over the years in spite of the tacit consensus on soft law approaches and self-regulation; case law and legislation applied at the national and regional level returned as a dominant regulatory approach, in particular for protecting civil liberties.

Second, institutional solidification—or efforts to build institutions and specialized bodies—remained a constant in the history of the Internet. This meant introducing strategic frameworks, global agendas and action plans, as well as creating monitoring and benchmarking tools, alongside efforts to build internal expertise. For the longest time, there was a near-absence of sanctioning mechanisms in the field, apart from the occasional reputational costs. This is starting to change with stronger stances taken by bodies such as the European Union (EU), which introduced considerable fines for non-compliance with its General Data Protection Regulation (GDPR). In the long run, strong public institutions provide a counterbalance to the power concentrated in too few private hands.

Third, modelling continues to shape Internet governance in unexpected ways. Modelling aims to shape the behavior of relevant actors without imposing direct restrictions on how they operate—either by promoting particular values to be endorsed by a larger group or by pre-empting regulation. Discursive actions (guiding principles, codes of conduct, resolutions, high-level statements, etc.) and operative guidance (recommendations, toolkits, model laws, etc.) have intensified in the last two decades, stressing the high political stakes of this policy field. CEOs calling for the introduction of norms and policies is part of this.


Twenty-five years after the privatization and globalization of the Internet, skepticism and caution dominate the public discourse around its regulation. The power of Internet giants—Google, Apple, Facebook, Amazon, Microsoft (GAFAM) on the American side and of Baidu, Ali Baba and Tencent (BAT) on the Chinese side—has crystallized in a highly consolidated market. Two features of this market give us a pessimistic outlook: the buy-out of innovative start-ups before they become competitive; and the cross-sector expansion of Internet giants and investments in developing their own infrastructure. The five GAFAM companies are now regarded as corporate monopolies, and regulators around the world have started to investigate ways to take them down. Proposals for breaking up the tech giants in the U.S. follow a number of efforts in the EU to reduce the power of these companies. However, some forms of government-mandated policing that companies implement consolidate, rather than weaken, their position on the market.

While scholars note stronger national approaches in areas such as cybersecurity and privacy, many areas on which the Internet economy thrives remain unregulated. Some are decades old, like connectivity, and others are more recent, such as cloud computing or artificial intelligence. Whether it’s harmful content or data portability, the battle for imposing norms of behavior online is more intense. To understand these dynamics, we need to explore the many ways in which these spaces are privately governed, even in the absence of regulation.

The current imbalance of power in Internet governance has deeper, systemic roots that regulation is not likely to address on a large scale. It is time to ask the uncomfortable questions, raise the distributed authority issue and question the “one action at a time” logic. It is time to act on more than one level to bring about change. 

Roxana Radu is a Postdoctoral Researcher at the University of Oxford’s Programme in Comparative Media Law and Policy, working on Internet regulation, algorithms and knowledge production in the public sphere. She is the author of the book Negotiating Internet Governance.

The views expressed in this publication are solely those of the author and do not necessarily reflect the views of the EastWest Institute