2018 ITU Plenipotentiary Conference: It’s time to change how ITU makes decisions

Consensus by sleep deprivation

Diplomats often joke about the fact that “consensus by exhaustion” is how many resolutions are agreed to. But these days, at the ITU, the joke is souring. People aren’t just exhausted, but positively sleep deprived. Recent research has shown that lack of sleep makes people far more likely to rise to anger. This anger-inducing effect of sleep deprivation has clearly been in evidence in the last two major ITU conferences I’ve attended – the Plenipotentiary Conference 2018 in Dubai and the World Telecommunication Development Conference (WTDC) in Buenos Aires – resulting it becoming even harder for delegates to negotiate agreements to resolutions that were already highly divisive. Delegates blamed each other on the floor for acting in bad faith, and a few times even questioned the conference chair’s competence. Member States may discuss these things among their allies in private, but it’s only the extreme frustration and anger resulting from multiple weeks of general exhaustion and increasing sleep deprivation that leads Member States to utter such thoughts on the floor and available for the world to see on a public webcast.

Delegates can catch up on sleep after conferences end, but those feelings of resentment and anger never quite go away, particularly if you’re delegate to have been publicly accused of bad behavior. And to be frank, I think this ongoing build-up of sleep-deprived resentment is contributing to the worsening state of negotiations at ITU.

A frequent scene of Plenipotentiaries: a huddle of key negotiators at the front and various observers watching from a distance

How bad can sleep deprivation at ITU meetings really be?

They’re pretty bad. For some delegations, working days at ITU conferences begin with delegation meetings at 7:45 am. The first week of Plenipotentiary, negotiations need to end “early” to allow delegates to attend a long series of evening receptions hosted by countries wanting to be elected to ITU Council or various ITU senior management positions. Weekend sessions begin the first weekend, but with the Sunday off to allow the host country to show off their country by bussing participants off on tourist trips. In Week 2, night-time negotiations take the place of evening receptions. These negotiations continue through the second and last weekend of Plenipotentiary, and get even longer during the third week, until we reach inevitable Night of No Sleep. And this excludes the time delegates need, outside all of the formal and informal sessions at the venue, to review an unending stream of updated draft documents and prepare their own updated proposed amendments based on proposed amendments by others that have only occurred during the live negotiations, and that they may not have existing positions on. Like an ICANN meeting, you’re lucky if you get six hours of sleep a night. But unlike an ICANN meeting, which is generally only a week and a half long (various side meetings are often tacked on to the official ICANN meeting), sleep deprivation for the ITU crowd goes on for weeks.

The Night of No Sleep

Hours of the second last day of the Plenipotentiary:

9:30 am, 15 November – 5:25 am, 16 November 2018

The final day of started less than four hours later, at 9 am on 16 November.

Hours of the second last day of WTDC 2017:

8:00 am, 19 October – 4:30 am, 20 October 2017

The final day restarted at 8 am with informal negotiations and, without any food or sanity break, finally ended at 8:54 pm. (The conference was supposed to have wrapped up at midday, so all of the conference infrastructure outside the main room was being dismantled as delegates got more and more frustrated inside the main room.)

Unless you happen to be at a hotel with 24-hour kitchens right next to the venue, you spend the Night of No Sleep without food and without caffeine to keep you awake. Food and coffee that you managed to send your lowliest delegate out to buy at 9 pm isn’t going to sustain you through to 5 am.

When negotiations end just before dawn, the best that delegates can achieve is a nap before they have to be back in the room for the next day’s sessions to start.

The topics that led to sleep deprivation at WTDC 2017 and Plenipotentiary 2018

At the WTDC, it was the cybersecurity resolution that kept delegates up all night.

In the Plenipotentiary, there were multiple topics that kept delegates up through their last night:

  • Cybersecurity (Resolution 130)
  • Artificial intelligence (no resolution produced in the end)
  • International sanctions (Resolution 64)
  • References to Internet-related organizations (Resolutions 101, 102, 133 and 180)
  • A possible merger of two ITU Council working groups (a draft recommendation that in the end didn’t happen)
  • A proposed reference to small and community network operators in a connectivity resolution (Resolution 139)

Factors contributing to the ever-increasing difficulty of reaching consensus

The factors I’ve identified below are my perceptions, and mine alone, of what is causing difficulties. Others may dispute the groupings. There are probably also additional factors that are contributing to negotiation deadlocks that haven’t jumped out at me, but may have jumped out at those who have participated in the ITU and intergovernmental circles for longer than I have.

  1. The growing importance of ICTs to the world means more governments want a say in how ICTs are developed and governed
  2. A shift in ITU’s focus from basic infrastructure and connectivity issues to complex ICT applications and services deepens the fault line between intergovernmental and multistakeholder models of ICT governance
  3. The inclusion of topics that are not within the expertise of the telecommunications and ICT ministry representatives who (rightly) dominate the ITU
  4. The ITU conference structure was not designed to handle the increasingly large number of highly contentious topics on its agenda
  5. The oft-invoked negotiating tactic of “ask for vastly more than you want, in the hope that you get the small amount you really want”
  6. Increasingly bloated resolutions that make easy for delegates to get lost in the weeds
  7. A change in leadership style at ITU’s highest level

These seven factors are described in detail below:

1. The growing importance of ICTs to the world means more governments want a say in how ICTs are developed and governed

ICTs increasingly raise wicked policy problems, which, in theory, are best tackled by getting as many perspectives on the multi-dimensional issues as possible, leading to solutions that, if not perfect, are far better than potential solutions that only look at a problem from a single perspective. In practice, however, the more perspectives that are brought to the table, the harder it can be to find a way forward. Only Member States make the decisions during Plenipotentiaries, WTDC, and similar high-level conferences, but with an increasing number of ITU’s 193 Member States feeling the need to have a say on how ICTs have an impact on their country, it’s harder to find ways forward that suit everyone’s needs. I’m not suggesting that we should go back to the Cold War days when Russia and the USA were the main protagonists, their allies lined up behind them, and a few independents lurked in the margins. But we do need to recognize that incorporating a more diverse range of needs and opinions from more governments will inherently slow down the decision-making process at ITU (and other intergovernmental forums).

2. A shift in ITU’s focus from basic infrastructure and connectivity issues to complex ICT applications and services deepens the fault line between intergovernmental and multistakeholder models of ICT governance

Today’s governments and intergovernmental organizations (IGOs) are mostly designed for a world in which change happens at 19th and early 20th century speeds. Governments and IGOs can find it difficult to keep up with the sheer speed of ICT developments that are mostly emerging from the private sector in developed countries and to manage the resulting social, cultural, economic, political and environmental impact that ICTs are increasingly having on the world. Developing countries, in particular, can feel overwhelmed by the deluge of emerging ICT issues and, not having the resources to participate in various issue-specific bodies, want the ITU to be a one-stop-shop for all of these issues.

Other governments, however, are concerned about ITU duplicating work already being performed by more specialist organizations and about ITU risking underperforming in its core functions by trying to cover more ground with a finite set of resources. Governments holding this view tend to be those with the resources to participate in multiple venues, or who practise a multistakeholder view of ICT governance that means they, as governments, don’t feel the need to participate in all venues specializing in emerging ICT issues because they’re confident that the non-government specialists who do participate are those best qualified to be making decisions about the development of standards for those emerging ICTs.

The difference of opinion on whether an intergovernmental governance model—where processes about emerging ICTs can be centralized and accessible to all governments, regardless of resource limitation—or a multistakeholder governance approach—where there is a decentralized, informally cooperating web of specialist processes that utilize the inputs of interested stakeholders—has existed since the World Summit on the Information Society (WSIS) era, and earlier, and has played out at previous Plenipotentiaries via the Internet-related resolutions (101, 102, 133 and 180). But at the most recent Plenipotentiary, the introduction of proposals related to ICT applications and services, such as big data, artificial intelligence and Over-The-Top (OTT) services, have extended the battle field.

3. The inclusion of topics that are not within the expertise of the telecommunications and ICT ministry representatives who (rightly) dominate the ITU

The old saying goes that when you’re holding a hammer, everything looks like a nail. This also applies to governments, and how they frame issues. For example, the long reign of neoliberalism, where everything began to be seen through a purely economic framework of immediate financial profit versus loss, has deeply affected how many governments are approaching climate change: those governments in countries that have been highly influenced by neoliberal economics are reluctant to do anything that will limit economic growth in their country now, to the detriment of their children and grandchildren’s potential for survival in mere decades.

Similarly, for governments struggling to understand and manage the impact of ICTs on a wide range of sectors and issues in their countries, the fact that ICTs can feel so new and alien means it can be an attractive proposition for them to view ICTs as an issue in and of themselves, rather than as enablers within existing sectors or as digital manifestation of issues that already exist in the real world. The tendency to view everything with an ICT-related dimension as an issue to be solved by a single ICT-related hammer leads many governments to want ITU to take on a broader and broader range of issues and activities. Unfortunately, this is causing the traditional structure of Plenipotentiary decision-making to strain under the weight of everything on the agenda.

Some ITU resolutions are about issues that would be better handled by the experts who participate in the various UNGA committees (two examples: international sanctions and the sovereign rights of occupied territories) or probably better addressed within other UN bodies that have the mandate over a broader issue in which ICTs is a supporting element (examples: Ebola – this resolution has now been suppressed – and human trafficking).

4. The ITU conference structure was not designed to handle the increasingly large number of highly contentious topics on its agenda

In less complicated times, the ITU Plenipotentiary structure of having a Plenary that allocated work to a number of specialist committees as well as a Working Group of Plenary, which in turn delegated more specific negotiations to Ad Hoc Groups and the less formal Informal Coordinations, Groups and Consultations made sense.

The basic structure of how decision-making works at ITU Plenipotentiary Conferences

But in 2018, with up to half a dozen Ad Hoc Groups having to meet in parallel, and many of those Ad Hoc Groups spinning off their own informal consultations and coordination groups to work on the most contentious parts of proposals, the system is breaking down.

To demonstrate how overburdened the Plenipotentiary 2018 decision-making structure was, I put together the graphic below. It’s based on the Ad Hoc Groups and other informal groupings listed in the daily schedule for the Plenipotentiary, as well as the final reports of the Working Group of the Plenary and Committees 5 and 6 to Plenary. The graphic doesn’t include the various very informal drafting efforts that happened for specific pieces of text. These very informal efforts were generally only between a handful of parties and didn’t need a room, so weren’t listed in the daily schedule for the meeting. Click on the image to see it at its full size:

How decision-making was delegated to a myriad of sub-processes at  Plenipotentiary 2018

The above graphic reveals that the Working Group of Plenary spun off 19 different sub-processes, which in turn spun off an additional 13 processes – one of which was an additional Ad Hoc Group. In total, the Working Group of the Plenary, Committee 5 and Committee 6 spun off 49 direct sub-processes in Dubai. Despite Member State asking that there not be parallel session during Plenipotentiary, as you can see from this very large number of sub-processes, there was no option except to have parallel sessions in Dubai.

Let’s remember, however, that most Member States at Plenipotentiary don’t have the luxury of assigning a single delegate to represent a single topic. Most delegates – and this isn’t just a problem for developing countries, but for many developed countries, too – have to cover multiple issues, meaning they can’t be in every meeting discussing a topic of importance to them, and even being in the sessions they can cover means they’re working late nights and weekends. In Dubai, as in Buenos Aires last year, this led to tired, cranky, and increasingly unwell delegates who were less willing to compromise on issues that they suspected the “other side” of having negative ulterior motives about. It also made it more likely that agreements reached in one of the many parallel sessions would not include all interested parties, resulting in fragile agreements falling apart when an interested party who’d been unable to attend the earlier negotiation turning up for the next session, and objecting to the agreement reached in their absence.

In addition, none of the sub-processes held below the level of Working Group of Plenary and the Committees have simultaneous interpretation or live transcripts. This means that delegates who don’t have English as a first language are very much at a disadvantage in the decision-making processes.

5. The oft-invoked negotiating tactic of “ask for vastly more than you want, in the hope that you get the small amount you really want”

The tendency for Member States to inflate what they’re asking for led to many stalled negotiations while delegates played out an elaborate ritual of defending every last piece of their proposals for a good week or more, only dropping the parts they never really expected to be adopted “in the interests of compromise” in the closing days (and sometimes hours) of the conference.

Everyone knows this happens. Member States know what other Member States really want because Member States tend to push for what they really want in multiple processes, while the inflated additions tend to be random pop-ups that will never be seen again. But the price of carrying out this elaborate charade is that multiple days are wasted, putting greater pressure on everyone to find compromise solutions in the final Night of No Sleep, when people are at their most exhausted and crankiest.

6. Increasingly bloated resolutions that make easy for delegates to get lost in the weeds

I have written about this in detail in my previous blog post, Resolution bloat at ITU Plenipotentiary Conferences.

7. A change in leadership style at ITU’s highest level

The previous Secretary-General, Hamadoun Touré, was a consummate negotiator. The current Secretary-General, Houlin Zhao, is more of an administrator. Just as Touré would do, Zhao also performs “we need to come together” speeches when it seems a negotiation is intractable. But unlike Touré, Zhao hasn’t laid extensive groundwork first, which means his last-minute inspirational speeches are less successful at getting Member States to resolve their differences. Well before he gave any rousing “come together” speech, Touré would have identified the issues most likely to be contentious and have embarked on informal meetings with the key parties as soon as the conference started, all in an effort to make sure that there was a positive outcome for the conference. It seems that in the absence of that mediation activity previously performed as a personal initiative of Touré, Member States have yet to find a workable solution for bridging large differences between them. Member States can’t expect Secretary-Generals having negotiating skills, either, as each person will bring their own unique set of talents to the job. So Member States need to be able to find their own ways of negotiation directly between each other, rather than relying on the Secretary-General to knock heads together. This is even more vital in the current geopolitical climate, where some governments are taking a more nationalistic approach to their place in the world and others are testing the boundaries of what it means to be a pariah on the international stage.

The perseverance of committed delegates saved the day in Dubai, but unless decision-making at ITU changes, there is a breakdown point looming in the future

With all the eight factors above colliding, decision-making in Dubai slowed down to the point where it’s a miracle that the conference was able to produce 53 updated resolutions, 11 new resolutions and to suppress (remove from the list of currently active resolutions) 11 existing resolutions.

The outcomes of the conference demonstrate a marvel of endurance by utterly committed – if totally exhausted – delegates, highly accomplished chairing and a general recognition by all that many of the issues covered by ITU resolutions really did need to move forward in a positive manner.

Unfortunately, however, the stamina and goodwill of ITU participants will not be able to overcome the structural and political issues that are leading to increasingly bloated and contentious conferences. Most delegates from the Buenos Aires meeting last year were ill after the meeting ended – a consequence of immune systems being worn down by the long nights and weekends of negotiations. No doubt, the same has happened to delegates of the recent Plenipotentiary in Dubai.

With ITU holding at one major decision-making conference every year, with a series of preparatory meetings associated with each one of those meetings, not to mention numerous other events that ITU holds each year, it is unsustainable for both the ITU’s and Member States’ resources to continue on this path of expansion of activities and resolutions without doing some hard rationalization of pre-existing activities and resolutions that may no longer be as relevant to the world. Overstretched Member States will only continue to view each other with increased suspicion, leading to ever more difficult negotiations. In a world where misuse of ICTs can severely hamper reaping the benefits that ICTs can offer, ITU and its Member States need to find a new way of working, to ensure it becomes an organization of and for the future. If it doesn’t, ITU risks becoming irrelevant, smothered by its adherence to 19th and 20th century methods of negotiating.