Over the last three months of 2017, EFF has been representing the interests of Internet users and innovators at three very different global Internet governance meetings; ICANN, the Global Conference on Cyberspace (GCCS), and this week in Geneva, the global Internet Governance Forum (IGF). All of these to some extent or other are held out as representing a so-called multi-stakeholder model of Internet governance. Yet in practice there are such vast differences between them—with the GCCS being mostly government-organized, ICANN being mostly privately-organized, and the IGF falling somewhere in between—that it’s difficult to see what this multi-stakeholder model really represents.

This is one reason why EFF has generally eschewed promoting a particular model of governance by name, but rather has emphasized how fair processes with the characteristics of inclusion, balance, and accountability, can lead to better outcomes. Last month UNESCO issued a report [PDF] with a more detailed list of its own criteria of multi-stakeholder governance processes, according to which such processes should be inclusive, diverse, collaborative, transparent, flexible and relevant, private and safe, and accountable. The use of criteria such as these, rather than merely the application of the buzzword “multi-stakeholder”, enables us to critique how particular global meetings fall short in effectively involving users in the development of policies that impact them.

ICANN’s Multi-Stakeholder Model

For example, although ICANN is the organization with the least degree of control by governments, does that make it the most effective at protecting users’ rights? Not necessarily, because of the way in which its work is organized. As the Association for Progressive Communications (APC) has put it in their briefing on this year’s IGF meeting, “being influential in ICANN requires a degree of effort and consistency which is difficult to sustain.” Consequently, there is a strong tendency for ICANN working groups to be stacked with private sector stakeholders such as lawyers for intellectual property rights holders and the domain name industry, who are able to dominate discussions, to obstruct attempts at compromise, and to push for one-sided outcomes, such as the right for a single company to control a generic word domain.

As a result, ICANN, although notionally multi-stakeholder, in practice fails to fulfil the criterion of balance. Its processes do not place a priority on the facilitation of understanding and consensus between warring stakeholder groups, and this feeds politicking and strategic behavior. Even many industry stakeholders acknowledge this shortcoming; for example Jonathan Matkowsky, who works for a digital threat management company, said in an ICANN mailing list post recently, “It's very sad to see the open Internet breaking down as a result of the multistakeholder process failing to work.”

The Multi-Stakeholder IGF Under Threat

The IGF falls short in different ways. One of these is the criterion of accountability. Management of the IGF is heavily dependent upon the office of the United Nations Secretary-General, which appoints the IGF's Multistakeholder Advisory Group (MAG) in an opaque, top-down process, resulting in a top-heavy group dominated by government and industry. Originally, records of MAG deliberations were kept secret, although meeting minutes and mailing list archives have since been opened to the public.

Another way in which the IGF falls short is in failing to provide a clear pathway for the discussions that occur there to feed into the work that its stakeholders do elsewhere, such as the development of laws and regulations by governments, the development of of terms of service and policies by companies, and the design of software, standards, and tech by coders and hackers. This isn’t a separate criterion in EFF’s model of fair processes, but it is represented in the closing paragraph of our infographic on this topic, which we explain by saying “there is no point in inviting affected communities to help develop policies for the Internet if their recommendations are ignored”.

For example, although the IGF’s grassroots-organized Dynamic Coalitions can and do produce recommendations, such as the resolution on transparency in trade that the Dynamic Coalition on Trade and the Internet issued this week, the IGF itself has never done so, despite a paragraph in its mandate that requires it to be able to make recommendations, where appropriate. This is one factor has led many stakeholders, particularly from government and business, to abandon the IGF for alternative fora, and has made it difficult for the IGF to raise funds. It has even made it difficult for the IGF to find countries willing to host its meetings; in an unprecedented failure, the IGF Secretariat has yet to secure a host for its 2018 meeting, and was only able to hold a meeting in 2017 by hosting it at the UN office in Geneva.

What is Replacing the Multi-Stakeholder Model?

Why, you might ask, does it matter if a fairly obscure, 12 year old Internet governance forum loses support and goes away? Well, that really depends on where the IGF’s participants go instead. If this means that governments and business flock to less inclusive institutions such as the International Telecommunications Union (ITU) to develop rules and policies for the Internet, that could end up being profoundly dangerous for users.

More or less, that seems to be what is happening, as governments are increasingly bypassing civil society and concluding agreements directly with companies. The increasing treatment of Internet public policy issues in closed, opaque trade negotiations at regional levels and at the World Trade Organization (WTO) is one example of this. There are also governments pushing at the Working Group on Enhanced Cooperation of the Commission on Science and Technology for Development (CSTD) for the establishment of a new, government-led multilateral mechanism for the oversight of Internet-related public policy development.

In November the Council of Europe concluded agreements with large tech companies and associations on human rights, democracy, and the rule of law. The Secretary-General of the Council of Europe Thorbjørn Jagland said in his speech at the ceremonial signing, “it is the first time the Council of Europe is also giving a formal, institutional role to the private sector, one which is open-ended allowing other companies and representative associations to join in the future.” The text of this agreement is not publicly available—at least, EFF requested a copy of it from multiple parties, and was told a month ago by the Council of Europe, “we’ll look into it”.

China’s Alibaba, now the world’s largest retailer, is also taking a larger role in global Internet governance, partnering directly with governments, but leaving civil society in the cold. It recently launched a pilot Digital Free Trade Zone as a a public-private partnership with the Malaysian government, and its CEO Jack Ma was also at last month’s WTO ministerial meeting in Argentina to announce a partnership with the WTO to create an Electronic World Trade Platform (eWTP).

True Believers in Multi-Stakeholder Models

That’s why there is merit in continuing to strive for the development and improvement of truly inclusive, balanced, and accountable global fora for the discussion of Internet policy issues, rather than allowing governmental and industry-only fora to dominate. This might mean a reinvigorated and improved IGF, or it might mean something new.

Microsoft has proposed this year that there should be a new Digital Geneva Convention on cybersecurity, and during this year’s IGF it gave further details of how it sees the initial draft of this document emerging from a multi-stakeholder dialog, although it would be finalized by governments in the same manner as a conventional international treaty. The proposal has received a mixed reception here in Geneva.

The Internet Society is incubating a project that aims to bring the multi-stakeholder model to the development of other policy issues, in an outcome-oriented fashion that has eluded the IGF to date. The project, which was approved as a pilot by the board of the Internet Society in November, aims to undertake three key activities:

• Convening stakeholders to solve concrete problems and develop norms on a consensus basis,
• Training stakeholders on how to be effective in multistakeholder discussions, and
• Building and promoting academic research and writing on the multistakeholder approach.

Meanwhile a French civic enterprise called Missions Publiques is promoting its proposal for a Global Citizens Debate on the future of the Internet to be piloted across the world during 2018. The project would involve ordinary citizens coming together to actively deliberate on a concrete policy issue, the results of which could then feed back to policy makers at the IGF and other venues. The project is currently seeking support from governmental, private sector and civil society partners.

For our part, we are chairing a group that is developing an option paper for the IGF’s own Multi-year Strategic Work Plan Working Group, to investigate whether there are any such multi-stakeholder processes that the IGF itself could use, possibly incorporating one or more of the above external initiatives or partners, to improve its own ability to generate useful and actionable policy recommendations, while avoiding the problems of capture that have beset ICANN, or the democratic deficits of intergovernmental text negotiations.

The important thing is not whether a particular global policy forum such as the IGF lives or dies. None of the existing Internet governance forums is perfect, or close to it. But such fora will always be part of the global governance ecosystem, and whether they are inclusive, balanced, and accountable matters. The flaws of particular self-identified multi-stakeholder fora should be identified and addressed, using user-focused criteria such as those developed by EFF and UNESCO. And we should also remain open to the idea that new innovations in global governance could emerge that would fulfil these criteria better than existing processes and institutions do.