IGF 2017 - Day 2 - Room XXVI - WS32 Data Localization and Barriers to Crossborder Data Flows


The following are the outputs of the real-time captioning taken during the Twelfth Annual Meeting of the Internet Governance Forum (IGF) in Geneva, Switzerland, from 17 to 21 December 2017. Although it is largely accurate, in some cases it may be incomplete or inaccurate due to inaudible passages or transcription errors. It is posted as an aid to understanding the proceedings at the event, but should not be treated as an authoritative record. 



>> Good morning.  I co‑lead our center for the fourth industrial revolution, based in San Francisco.  I'm really pleased to be here, not only with all of you, but also with professor William Drake.  We've collaborated together on this particular IGF session, which we've been looking forward to for some time.  And what I'd like to do is make a few contextual comments and introduce our still‑arriving and very welcome key round table participants today.  Just a word or two of context.  We are all aware that the cross-border nature of data flows is one of the most important aspects of this still emerging and deepening digital economy, but it also is one of those dimensions that policymakers and stakeholders are still grappling with.  What are the right approaches? How best to strike a balance between competing legitimate concerns on the economic side or social side, and how, in particular, to maintain the globally interoperable nature of the internet while addressing those legitimate economic, social, and security concerns.  I think the international community is an evolving process of experimenting, learning, trying to develop some common approaches where possible to make the approach to this issue one that works for societies as well as the economies to the optimal extent.  We have been utilizing our multi‑stakeholder platform.  We are one of a number of different kinds of stakeholder frameworks.  In our case, we have been attempting to use our particular platform to encourage stakeholders to have a dialogue, but also to think together and take ‑‑ develop some common bodies of work, and it has been the case in this area as well.  For the past year or so ‑‑ we have been developing dialogue to help bridge different disciplines, particularly the trade community on the one hand and the internet governance community and technical eco‑system on the other, which does not have a long and robust track record of having common approaches to policymaking and norm setting.  We thought because of the interdisciplinary and multi‑stakeholder of our platform, perhaps that's one contribution that we can make in addition to other frameworks and others like the IGF are also seeking to make. 

 So, a couple of years ago, we invited a foundational piece of thinking and research, which was entitled internet fragmentation and overview, drafted by professor Drake, Vint Cerf, and Kleinwächter, to lay out the landscape of potential modalities through which the internet is or could fragment, and assessing the implications of that, whether positive or negative.  And to also provide a taxonomy, a common frame of reference for different parties and different intellectual disciplines to be able to have a common starting point, a resource base, if you will, for a more productive conversation about how to best address these conversations.  I see today's session and the follow‑on piece of work that we're in the midst of developing with professor Drake as a basis for taking the same kind of approach to provide a foundational piece of work looking at the landscape, the history, the taxonomy to provide a more fruitful and informed basis for which to have a discussion on the specific issues.  What is emerging is this notion that we will have as the fulcrum of the discussion today is should we be thinking about this issue as a multifaceted or multitrack way? Not only the traditional approaches of intergovernmental discussions, but also by utilizing a multi‑stakeholder, interdisciplinary process that could encourage better thinking about this issue, but also potentially provide a basis for other geometries, if you will, of parties coming together.  And developing productive approaches, best practices, good practices, guidelines for policy at a national level going forward. 

 What I'd like to do before inviting professor Drake to provide some opening comments based on the research that he has been doing in dialogue with our community is to introduce our round table participants today.  I thank you all for being with us.  I apologize a little bit about the geometry of this room, which is not necessarily optimal, but we'll try to work through that.  I'm going to just introduce you now and then turn it over to Bill.  Fiona Alexander ‑‑

>> If you can raise your hand?

>> Yeah, I think that will be helpful.  Vint Cerf, Vice President and chief internet evangelist at Google.  Professor Bill Drake.  Is Raul here? Vice President of global engagement for internet society.  Thanks for being with us.  Anriette Esterhuysen.  Is she ‑‑ okay.  Very good.  Torbjörn Fredriksson.  He's on home turf here.  Head of technology and logistics here.  Goran Marby, Ricardo Melendez‑Ortiz, who has just made a trip back from the WTO.  Marietje Schaake, Thomas Schneider.  Yes, thank you.  Our host here.  Lee Tuthill, WTO, Mary Uduma, managing director digital solutions, Nigeria, thank you.  And Hong Xue, china. 

 So, we're going to split the discussion into two basic parts.  The first will be just taking a look at the landscape of how is data localization occurring currently? What approaches are being pursued? How effective may they be? And we have some specific information to put on the table, and we'll get a little discussion from the round table participants around that piece.  Then we will look at the suggested approaches about the multitrack notion in the second half‑hour of the discussion, again with round table participants.  For the final third of the segment, we'll open it up for room‑wide discussion.  With that, professor Drake. 

>> MR. WILLIAM DRAKE: Thank you.  Good afternoon, everybody.  So, this is not a very great setting for a round table, obviously.  We've managed to cluster most of our speakers in the center.  As Rick indicated, this is a project that has been running for a while now.  We've had several meetings over the past year with a number of different stakeholders, and we've also had sessions that we did at the forum and now here.  Digital trade is becoming an increasing part of the global policy agenda.  It is being pursued fairly aggressively or with some focus.  Particularly in the industrialized world.  And as that's proceeding, at the same time, we've had an internet governance set of dialogues, and the community slowly catching up and going well, there's things going on in the trade space.  Maybe we should be connecting with this and so on.  So, what we're really trying to do is bring together these two different worlds.  We did an open forum this morning at 9:00 a.m. where we talked about this a bit, and this is sort of building on the conversation that we had there.  A number of you were there, I know. 

 Just a background issue quickly, data flows are becoming absolutely essential to the global economy.  There was a study pointing out that from 2005 to 2012, data flows accounted for ‑‑ of global GDP.  Much of the world's data, 90% that's estimated has been created in the past couple years, and data flows are becoming key to the way value chains work in a variety of economic sectors.  It's key to the industrial internet, internet of things, and it's becoming central to growth and wealth creation in a wide variety of contexts.  In micro and small and medium‑sized enterprises are accessing global platforms and markets via the digital environment.  Next slide please? While that is happening, many governments are looking to project territorial authority on to the internet and requiring putting new requirements on data.  Particularly many countries have rules that require that data be locally stored.  It may be broadly or more narrowly gauged.  Sectors sometimes have a macro basis.  Many require ‑‑ and sometimes indeed that it be processed by national suppliers, often in conformity with unique national standards, which are not the national standards.  In some cases, these policies also include restrictions on the cross‑border movement of certain categories of data, or the prior consent of such transmissions or outright bans.  There have been countries that have contemplated reorganized their network architecture.  Their segment of the internet, as the Russian government likes to say, to ensure that data is routed locally within their space as much as possible. 

 So, we've seen a lot of variations in these kinds of policies.  I don't want to go through all the details, but clearly for national security and law enforcement, there have been a lot of regulations, financial, health information, and personal information of various sorts.  Sometimes it's a cross‑sectoral type of approach, and sometimes it's pre‑digital policies that require that data be resident.  Many countries require that companies keep their books on their premises to be checked, etc.  The motivations for these policies vary.  Sometimes, absolutely defensible things like insuring law enforcement access, but sometimes it also goes to broader privacy, but sometimes it goes to broader types of approaches where governments are seeking to surveil their data environment in their country or have cyber sovereignty to advance their interest at the expense of others.  The interest of particular providers.  The incidence of this, it varies again.  Some countries are very prominent examples.  There have been proposals in Nigeria.  Estimates vary.  There have been estimates to try to take stock in this.  The commission counted 21 within the EU.  So, you see the differences there.  Next slide? Data flow, sometimes there are barriers to data flow which are not simply a function of data localization requirements.  Now over 100 countries have got rules put in place.  And very often, these require completely legitimate purpose, of course.  Very often these require an adequacy assessment to make sure that data is protected.  This is a controversial point, but sometimes it may happen. 

 Other forms of restrictions on data flow, obviously censorship, which takes many forms.  And then there's been much discussion of digital protectionism.  There's no agreed definition on this.  There are technical standards and procedures and so on that impose limits on platforms, etc. 

 Estimates of the incidence of this vary, but it seems that 75 countries that have restrictions ‑‑ 75 restrictions in 59 countries based on privacy protection that prohibit the transfer of data.  Next slide please?

 Studies have been done to show that GDP is reduced in a number of countries.  One can also raise concerns and there are concerns about whether or not this fragments the internet.  So, in response of this, there have been proposals in the trade environment.  More recently, a number of these provisions on this language ‑‑ on these issues have been put forward.  Most notable has been the TPP, now the CPTPP, which is quite a mouthful, which requires ‑‑ certain requirements for localization and various data flow that says that any policies that are pursued must not be applied in a manner that would constitute arbitrary or unjust discrimination, restrictions in trade, or that impose restrictions greater than are required to on Chief the objective. 

 The question on the table, though, for us is can trade rules alone suffice to deal with these issues? There are clearly political problems with the views of governments that may not be sold on the idea, that you should have a ban on data localization or bans of any sort.  There are limitations that oppose certain trade provisions.  And there are other issues from the internet governance world who also have concerns based on the kinds of experiences they have in multi‑stakeholder settings that are transparent and open and inclusive, which is very different from what goes on in trade negotiations.  And many have substantive concerns as well.  And finally, there are institutional aspects about using trade here, the very fact of negotiating binding treaty‑based agreements subject to possible sanctions for violations changes the negotiation dynamics.  And there's possibilities of issues being bundled together and linked, people saying I won't sign off unless you give me open concessions on trading bananas or whatever the case may be.  And that's a slow and difficult process. 

 These are the issues that we're putting on the table as background.  What we wanted to do is get views from people.  We've got a very multi‑stakeholder panel of people from all sectors on the nature of these problems and whether they feel trade policy is a sufficient response. 

>> MR. RICHARD SAMANS: Just to open the discussion to our round table participants, how does one feel about the data.  What are the legitimate and might be considered less legitimate motivations for this? It's important to consider why this is happening if you want to address it well? And Bill put the question pointedly.  How fit is the trade regime to address these issues? And I invite anyone who would like to comment on those or other issues.  We'll just take question comments through the round table. 

>> VINT: I'll dive in where angels fear to tread.  Let me start out by saying, as I approach this set of questions, my first question is also, who are the interested parties and what are their interests? What incentives might they have for choosing one policy or another? The second question that I would have what potential hazards might various policies have on the open connection of the internet? And third is whether the analogies one tries to draw from the physical world would turn out to be inappropriate for implementation? There are, as you look at trade manifests and what do you call them? Big ‑‑ sorry.  Going blank.  What are those big boxes that the ships move around? The containers.  I should know that.  Container is also a computer science idea.  I don't know why I'm losing that.  The question is whether the container of the shipping world and the container of the computer science world share only a name and no other characteristic worthy of attention.  So, I raise those not so much as answers, but as a desire to gather information before coming to any conclusions about adoption of one policy or another. 

>> MR. RICHARD SAMANS: Thank you.  Helpful. 

>> LEE TUTHILL: Just looking at the effectiveness issues of the questions you pose, and I have three quick points.  One that I think an awful lot of rules exist already.  I think it's a myth that we don't have rules already.  So, the question is, you know, what else might be needed, if anything? And if you ‑‑ as the panels have determined, they see an awful lot of technology neutrality, but we don't get at the level of detail at the WTO that national laws get into.  So, we have less of a problem of trying to ‑‑ the analog network.  Second, trade is a tiny part of the overall picture.  I don't think it will ever be the whole picture.  If some proponents get what they want, we would move in the direction of an open internet.  And what do internet people need to know about the WTO? I think I've spoken at previous IGFs and say we need to understand for a lot of the important issues, it talks about privacy and it sets the standards for rules that are made that can, if they need to, even break the WTO.  And this is what people in the internet community need to know about trade.  What trade people want to know is the solutions being proposed, how damaging might they be to trade? And would their overshoot the mark? And thirdly, I think from what I've seen, there's a myth that the ideas being circulated in the trade world are all out there to benefit the big multinationals.  Depending on who you put in the acronym, it keeps changing.  From the work we do as our member governments, many of them have small, and many of them developing, my take on this, and it's a personal take is the S & Ds will benefit an awful lot more by the predictability of having basic principles for data barriers of trade that involves data.  The big companies will, because the cost of data localization and the cost of other data varies.  It's much more prohibitive for the small companies in developing countries than it is for the big countries.  Thanks. 

>> MR. RICHARD SAMANS: Thank you.  Yes?

>> PANELIST: I think this is a useful discussion.  And if the question is can trade rules alone suffice, my answer would be no.  I think we need to begin to address, especially for the people who don't live in the trade world to begin to explain how trade negotiations and agreements generally work or are intended to work.  Because I run into a misunderstanding by a lot of people, not least in Civil Society, that only when there's an agreement, trade is facilitated.  But trade is there with or without rules around it.  And I believe that the public interest is better served when there is a frame of rules around that trade, including increasingly digital trade.  This presents us with new challenges and questions about jurisdiction, globally operating companies, what is the foundation in the law, etc.  So, I do not think that trade agreements will be setting norms as such.  I think they will be translating.  Let's take the EU, where I work on the European level as the example.  We have data protection net neutrality, intellectual property principles embedded in EU law.  But what I do think is useful is to integrate those standards in trade agreements, from my point of view, with the aim to increase those standards, to make them higher in developing economies, to make them leading to a level playing field for the SMEs operating in the global trade -- let's say from the EU to a third country or from a third country to the EU.  So, the way in which we should not understand trade rules is in seeking the middle of the road between two sides, so watering down what's a high standard or lifting up what's a low standard, but rather coming to an agreement of what is a shared set of principles and standards.  And I also don't think trade rules should be understood as leading to the lowest common denominator.  I think the question can trade rules alone suffice, and the answer to that is clearly no. 

>> MR. RICHARD SAMANS: Thank you.  Raul?

>> RAUL ECHEBERRIA: I would like to refer to the impact of the rules of globalization could have in the markets.  More companies, as my colleague pointed out before, not only small companies, but also new commerce.  So, the impact that it could have in the innovation could be ‑‑ today it's very easy for those developing a new business based on the digital world.  Try to put something in the cloud and reaching out to the global community so that anything that that could affect the possibilities is really killing the opportunity of new business to be the next big things.  So, in terms of trading, there is another point to make with regard to economic impacts.  This is the same point that I made yesterday during the discussion on encryption is the data economy ‑‑ the global economy today is mostly based on things that happen in the digital world.  So, anything that we do that can affect the free flow of information and security of the information, that could have a really ‑‑ it's impacts that, we don't anticipate really the impact on the economic stability would be huge. 

>> MR. RICHARD SAMANS: Thanks.  Anriette?

>> ANRIETTE ESTERHUYSEN: I want to make a case for villainizing data collection.  If the opposite of data localization is being imposed on developing countries through trade agreements and the way that certain free trade and intellectual property agreements that have made the promotion of open‑source software difficult for this is being imposed on them, that's not going to work.  I think that many developing countries look at their ‑‑ they are attracted to the idea of data localization because they see opportunity in that.  And I don't think that, even if I might disagree about whether there is opportunity in data localization, the threat of e‑commerce, the threat of a globalized e‑market, platforms that are huge monopolies, very hard for new entrants to enter that.  It needs to be looked at.  So, I think that data localization needs to be discussed, opposed, considered in the context of the real impact of ecommerce and of the digital economy on developing country economies, and they should be given a chance to have some influence and engage.  I think that comes to the procedural concerns.  You said it yourself.  I think the secrecy, the lack of transparency, in these processes being discussed at the moment doesn't give anyone the opportunity to do this in a way that could really benefit developing country economies in a meaningful way as well as respect and consider all the rights implications. 

>> MR. RICHARD SAMANS: Thanks.  Ricardo is on the floor. 

>> RICARDO MELENDEZ‑ORTIZ: It's very important to understand the way the trade systems work.  Today we have the frameworks to address the concerns that we're presented with.  However, as the principles and many of the norms in the frameworks are the point is that countries are not using that.  They're not using it when it comes to protecting or ensuring access to data flows.  And so, the question there is why? And so, the next thing I think, the next level, if you like, is what is it really that we're talking about? What we're talking about is really how critical access, views, and protection of data is, and then in which terms should it be provided? And we don't have clarity, exactly, on what we would like to do.  I think, because we're only now really trying to understand the complexity of the issue ‑‑ what I mean by that is we're in front of an upcoming quantum economy, but one that is very much based on data.  So artificial intelligence, the use of big data, internet of things, all this will affect business models, and will affect really the organizational production of trade that we have today.  It will also affect where the gains go.  All of that requires an illustration between stakeholders to try to understand what are exactly the terms that we need to put in place?

 So, my point, really, is we do have very comprehensive and particularly wise frameworks.  I don't dispute that.  But, are they really adequate to allow for insuring access, use, and protection? The terms that the governments want? I don't think that we do yet. 

>> MR. RICHARD SAMANS: Very interesting.  Professor?

>> PANELIST: Thank you.  I fully agree with my colleague.  Localization should not be imposed on developing countries.  We should think from the perspective of developing countries.  When we think about what is the best interest of developing countries in this data flow process, we need to put it in a bigger picture.  As we just mentioned at WTO, there's a new very important program called "Inclusive Trade," and it is being hotly discussed in the UN.  We should lower the barrier for the MSME and individuals to access the global market.  And actually, this localization requirement really imposed in the local law is very costly to the MSME and to the individual.  Imagine, how can you establish many data centers around the world.  This is not in the interest of developing countries.  If you think of it from that perspective, there would perhaps be a second thought. 

>> MR. RICHARD SAMANS: Thanks.  Mary?

>> MARY UDUMA: Data localization has many facets.  We have infrastructure side of looking at it.  We have economy.  We have the social angle of looking at it.  We are concerned that platform economy is increasing the joblessness of young people.  If everything is done there and with data we don't protect in any form, we run the risk ‑‑ we're already running the risk of unemployment for our youth.  That's for the economy. 

 We have, also, the issue of security.  For instance, as we speak data from my country is being profiled.  If you want to pay with your credit card when you buy on e‑commerce, and some competition will not accept a credit card from Nigeria.  What should we do? It will not help.  Okay?

 We talk of ‑‑ we don't have data protection laws in place.  Meanwhile, there's huge trade from the internet.  And young people and old people are buying their goods and transactions online.  So, the barriers are not yet there.  We are just thinking to put in place local content, with what we say is local content.  What if I want to trade in a country, there should be a local context to it.  Why do we say this? Because of what is already happening.  First, we have image problem.  So, data from Nigeria will be profiled.  We have infrastructure problem.  So, to manage the process.  There comes the part where my country has signed some aspect, but we have not given enough thought to these aspects of trade.  My colleagues in Nigeria who are doing everything without anybody interfering.  There are some that will develop organically.  But when we come to the full process with the nation, the access, the youth, and the protection, we are still grappling with that.  So, anything that's going to be done should be done in such a way that these ‑‑ our own situations should be taken consideration and barriers lowered so we're able to play in the same level as others.  That's what I can say for now, and as we continue to talk, we will raise other things. 

>> MR. RICHARD SAMANS: I suggest that we take one more comment on this segment of the conversation from Thomas Schneider, and then invite professor Drake to present the second part for the basis of our discussion today, and then we'll have another round among our round table participants.  Mr. Schneider?

>> THOMAS SCHNIEIDER: I think many people, and this is not just developing countries, but also an issue in all of our European countries, depending on where and what segment of income and education and so on you are.  So many people are afraid of losing control, of losing their economic basis, of losing the world that they grew up in.  And of course, that causes fears.  And politicians are afraid of losing control.  Those who are looking for control, at least, are afraid of losing control.  So, there's a mix of reasons, some of them are legitimate, some of them are maybe less legitimate.  Some people have a tendency to wanting the old times back, the good old times back.  Look for old models, like nations looking back at national protectionist regimes.  This is supported by the security issues that we are facing.  So, there's a new argument of people who say we have to go back to national systems because the world is not secure, the internet is not secure.  We can protect you only if you trust us as government within our borders.  And of course, that creates a new dynamic, also, to this.  I would say this is probably not fake news, but fake security that people are thinking that they get when they go back to old models, and they will probably sooner or later wake up and see that they can't make the world moving forward. 

 So, the question is, I think the underlying question is how to create trust that people move on in the process and don't try to block it or to step out of it? And the easiest way is people need to see a perspective, working class people, entrepreneurs, and also politicians, they need to see a perspective of the process and to carry it along.  One of the problems is, those who have seen the recent reports about growing inequalities in almost all our countries, of course the SMEs also benefit somehow, but other people benefit much more than the lower half of society.  And this is, of course, a ticking time bomb, if you want.  If these inequalities continue, the stability of any political economic system as much as the internet, it's actually our whole societies that are getting more fragmented or more risky, if we don't provide the perspectives for everybody.  The debate about leaving no one behind is going in the same direction.  There are some elements, of course, interesting elements on Bill's report that he's probably going to present us.  And trade rules are one element.  There are many others and many of them are actually at least alluded to in Bill's report. 

>> MR. RICHARD SAMANS: With that, why don't we invite you, professor Drake, to describe a little about some of the new suggested approaches to building broader dynamics that can build trust?

>> MR. WILLIAM DRAKE: So, my suggestion is to consider whether or not ‑‑ I don't think we're going to change what governments do in the trade policy community very directly in terms of whether they go forward or don't go forward, but we can try to think of ways to open up the conversation more.  And I think that the multitrack approach could be useful in this context to help address concerns that are driving day‑to‑day localization and the barriers to data flow, and identify what kinds of objectives are legitimate.  And trying to figure out, can we identify a basis for mutual understanding and building some consensus around what works and what doesn't work from a developmental and other kind of perspective to try to nudge national policies towards good practices.  And perhaps feed into the trade community's work by helping to shape the larger discourse within trade policy and efforts are embedded. 

 So, the suggestion is to consider a three‑track approach with not tracks that are completely disjointed, but rather on decentralized basis, different people will be doing different things.  But if they could be a bit more structured and interacted between those tracks, we might be able to get some cross pollination of ideas, concerns and so on.  The paper that will come out next month is to be guided by certain principles.  So, the first principle to me? Put the internet first.  Not to play on Trump, but how do we avoid doing damage to the fundamental internet that we're all relying on going forward for economic and social development? Other principles are focusing more on the jurisdiction over and access to data rather than where exactly it's located.  Can we try to ensure that countries have access to the data when they need it? Emphasizing the diversity of stakeholders that are impacted by restrictions.  This cuts across all sectors of society when you inhibit, restrict, retard development of data flow and so on, etc.  Engaging the concerns that drive restrictive policies, and drawing from the full menu of global governance models.  There's lots of way of pursuing cooperation. 

 So, the suggestion is we could do three tracks.  One track would involve non‑binding intergovernmental processes, such as the work that's already been doing in the G‑7, G‑20, but at a high level of generality.  Trying to institutionalize the agenda within the different governmental ministries and other processes and NGOs and private sectors and so on.  Orchestrating some coordination between the intergovernmental spheres, promoting regulatory cooperation among data protection people, consumer protection people, and so on.  Importantly, establishing, monitoring, and reporting mechanisms to enhance conformity with legally non‑binding mechanisms.  If we report on departures from shared standards, and everybody is aware of what's being done, this is encourage greater conformity with whatever agreement there may be.  And capacity development efforts. 

 Multi‑stakeholder level, we could also do more, I think, to facilitate the development of the global community of expertise of practice that is interdisciplinary that can bring together the different parties that have interests and expertise here, and perhaps establish expert grouping that could be a focal point of contact for intergovernmental processes.  There have been various task forces and commissions and so on, on other topics.  It's possible that this could be useful here.  And conducting expert analysis and dialogue on questions of exceptions and what is necessary or more restrictive than necessary.  And finally, the digital trade policy processes themselves.  You know, I think you'd be useful to clarify the applicability of existing disciplines and related national commitments.  Lee has pointed out that the GATS framework is already applicable.  What issues really need to be done through trade mechanisms versus other mechanisms? Clarify the relationship between privacy and data protection.  The fear that trade disciplines will somehow come to trump ‑‑ sorry to use that term ‑‑ but privacy is something that's been driving a lot of opposition to digital trade work.  And I think some country could be, perhaps, established there.  Craft digital trade norms in a more transparent and open manner.  I think the process by which we devise the fundamental rules of the game within the trade process could be opened up a bit.  And encourage participation of stakeholders and national trade consultations.  And ultimately, I think, from my standpoint, if there is going to be digital trade agreements, they should be ‑‑ I would argue, they should be stand‑alone rather than intermingled with other trade issues, because, again, this business of having trade ‑‑ having the opens of the internet traded off for concessions in unrelated markets or ideas that companies should be able to sue governments for their profit margin, etc.  This, I think, makes it very difficult to get agreement around measures that could help to preserve a more open internet.  I'll stop there. 

>> MR. RICHARD SAMANS: Okay.  Thanks very much.  What I'd like to do is invite the round table participants to comment on this.  These are ideas in ferment, and they're being tested with our community.  It's a really good chance to test it with you.  And then we'll take a segment of our time and open it up to the room as well as online as well.  I think first, do you still want to come in here? Yes? Go ahead.  Torbjörn first. 

>> There are rules requiring that some data be kept, and at least kept and copied in the country.  Even if it's just to certain sectors.  It attaches to certain segments, and really points to the need to have a multi‑stakeholder approach to looking at the best optimal solutions to protecting data and also regulating the flows of data across borders.  We are a world of silos.  So, we have, during the IGF, a tremendous presence of internet expertise in various areas.  Last week we had the WTO with a tremendous amount of trade expertise, and seldom they meet, these two worlds.  So, the challenge is always to see how do we bridge these groups? Internet, trade, maybe human rights, and other dimensions.  I think the work has been trying to capture all of these different perspectives in developing its policy levels.  So, I think we don't have the answers to all of this, but we would like to suggest that we need to explore every trusted shops that are available.  We have we have the annual e‑commerce meeting where we discuss these issues.  We have several intergovernmental expert groups that deal with e‑commerce and economy and deal with consumer protection and deal with competition policies.  They all have a bearing on these topics and implications of the data flows.  We should not forget that the forum is a platform that many can use.  In the end, it's up to member states to have this dialogue.  Since we are here in Sweden, I would like to express that we are ready to support this process.  Thank you. 

>> MR. RICHARD SAMANS: Thank you.  Sir?

>> PANELIST: I want to take Bill's number one point, which is to make sure we don't destroy the internet in the process of trying to introduce these ideas, let me remind you of one of the most important features: The packets are uninterpreted as they cross the borders.  It's very important that we do it that way, because the interpretation is done at the ends, the edges of the net, nowhere else.  Changing that could rapidly destroy the way in which the internet works.  Certainly, its speed and responsiveness.  That's one place where we want to be careful.  The second thing is if we really care about protecting the traffic from exposure and modification, then cryptography is actually our friend here.  I hope the people trying to wrestle with these problems don't forget that there are certain key properties that make the internetwork the way it does now.  Please be very careful not to destroy those in the course of trying to achieve some other objective. 

>> MR. RICHARD SAMANS: Thank you.  Marietje?

>> Marietje Schaake: I think rules can help protect the internet from destruction.  I want to caution on data flows when we're talking about trade.  There's a lot more going on.  Vint just mentioned encryption.  It's extremely important.  But encryption is still part of dual use regulation.  We in parliament have said it should be removed from that and completely uncontrolled, because it is so important for data protection, so we have to think about all the trade rules.  Intellectual property applies to offline goods.  Net neutrality.  We still have it in Europe.  We're proud of it.  We would like to see how it can actually continue to be part of a discussion of global standards of fair competition in the global digital economy.  Data protection we just discussed.  So those things considered, you know, I'm trying to wrap my head around it, but I'm not yet convinced that stand‑alone agreements are the best agreement.  Part of industry, production, services, etc., that I just don't know how you can divorce the two.  I just want to put that question back to everyone else.  Is this really feasible or not? And what could it lead to? One more comment on the notion of multi‑track approach and bringing people in to the discussion.  I think Civil Society should not underestimate the impact that they've had on trade discussions and engagement over the past couple of years.  I think if you boil it down, the impact of Civil Society mostly protests, I should say, but still Civil Society, but still has been extraordinary over the past couple of years.  What I would hope we can now move to is a much more constructive and inclusive discussion on what trade rules should look like.  And there's another point that I would want to put back to everybody.  If we stick with non‑binding on a number of issues, because it feels safer, right? Because it feels easier to agree on than binding rules at the moment, because it can be more multi‑stakeholder, we do have to think about what are the consequences of violations to the norms.  Because the whole idea of rules or norms is really that you agree on something and there are consequences if you violate.  So binding or non‑binding, but let's not avoid the question of what the consequences should be in case of a violation, otherwise everyone will say fine.  Scouts' honor, we agree, until they violate.  And then you have nothing to put back to these violating actors.  I think with agreements, there should be consequences in the cases of violation as well.  And otherwise, if the hard decisions are being put off to the future, I'm worried there will be more confrontational court cases and legislation to try to deal with some of the lowest points of anything from tax avoidance to non‑compliance with laws in other countries.  Uber is an interesting case study here to look at what laws they're running into and who, then, enforces those in the absence of global agreements on some of those issues. 

>> MR. RICHARD SAMANS: Raul, did you want to come back in?

>> RICARDO MELENDEZ‑ORTIZ: Thank you.  I want to be brief for the sake of the conversation.  I wanted to make a point that there could be some cases in which data is kept local.  But they are not forced to.  We have to look at the differences.  For quality of services, this is the reason because we promote ISPs around the world.  Also with growing of internet of things, there are many implications.  If I have my censors connected to a server, I prefer the server to be near, so it doesn't defile the information coming to my house.  Or services.  If I have a heart attack, I have something connected so that I want the doctors to come as soon as possible to treat me.  There could be many cases in which the localization of that is important, and it's a business opportunity.  And the conclusion we have to take is don't force data localization.  It is impossible.  Something that Vint said before that localization becomes impossible without compromising the internet's global reach, and this is one of the main values that we have to protect.  With regard to the things that you propose, Bill, and the pieces you propose on dialogue and discussion, I think that all of this is important.  The fate of localization is exactly opposite of that that the people who proposed it.  For example, security.  Some people use the reason that it increases security if you keep it local, having all the information in one single place increases the risk in terms of security.  Or even on jobs, and I think that's ‑‑ data localization, I don't think that it has an effect on creating jobs.  It has the opposite effect, because it renews the opportunities of people to participate in the digital economy.  As I said before, one of the main, the biggest companies in Latin American unit is a very light market.  They work with all the regional community.  If they have to be in one country in order to have the money for deploying the infrastructure as pointed out before, that could be almost impossible.  And this is a big thing.  So, we have to trust the experiences of people in the digital economy in developing countries and create avenues for them to include their barriers. 

>> MR. RICHARD SAMANS: Thank you.  Fiona and then we will open it up?

>> FIONA ALEXANDER: All three tracks are good tracks, and pursuing any and all of them would be good to do.  None are mutually exclusive.  It's research intensive, so there has to be priority setting.  And maybe to react to things that I heard colleagues say, but from my perspective and experience, I think track two, having these dialogues is still necessary before we can actually flesh out track one and three.  Because I do still think there's a huge disconnect between traditional trade stakeholders and traditional internet governance people.  That's evidenced by several of the venues that are referenced where we can have the conversations with other places.  Having a shared space to understand, I still feel like there's very much two distinct worlds ‑‑ internet people and trade people.  I feel like there's a big disconnect there. 

>> MR. RICHARD SAMANS: Okay.  Just before opening up, let me make one comment myself on track two.  If our colleague could put track two up.  I personally think, and I'll pose this more as a question for discussion.  Aren't we being a bit modest, particularly sitting here in the IGF, about the scope of what's up there.  If you look at what's up there, it's foundational.  It's dialogue and research.  But the internet has had multi‑stakeholder decision‑making about norms.  So rather than presume, it seems to me, that only intergovernmental processes really would be the decision‑making frameworks, it seems to me that fundamentally, that second track could actually make quite a contribution to beginning to shape norms.  They may not be formal multilateral intergovernmental forum, but a rough consensus or dynamic coalitions that take form a common view and take common practice and influence the behavior of others in this regard.  That would be my one question mark framework with regard to that.  With that, let's open it up to the room and online for other comments and discussion, and we'll see if we can bring a couple of round table participants in at the rear. 

>> AUDIENCE: Just to follow up, there is a dynamic coalition on trade in the internet at the IGF.  And we held our first meeting this morning, in fact.  One of the aims is to bridge the two worlds between trade community and the intergovernance community.  We would very much encourage everyone here interested to join the dynamic coalition.  You'll see that we endorsed a statement on the transparency, the improvement of the transparency and inclusivity of trade negotiations today.  In about an hour from now, you will be able to read a blog about that.  I notice there's also been a blog published on the intellectual property watch website as well.  Anyone who is interested in learning more, please don't hesitate to contact myself or my colleague here as representatives of the dynamic coalition.  Thanks very much. 

>> MR. RICHARD SAMANS: Lady in red near the rear of the room?

>> AUDIENCE: Sandia from Third World Network and I wanted to comment on the WTO's privacy exception.  Governments have only succeeded once out of 44 attempts and the privacy pattern is even harder to use.  You can only use it for laws that are consistent with the WTO rules.  Even the EU saw it as a problem.  Insisted on a better exception, not just copying it in.  And yesterday, the European District Attorney data supervisor said that it should not be subject to trade negotiations because the types of exceptions in trade agreements are not good enough, and that data localization rules could water down data protection rules and open them to challenge. 

 And as other people have already said here, this is a lot of internet experts, but developed countries require local data storage for things like effective tax regulation, financial regulation, privacy, security, and these experts are not here in the room.  So those other ministries including health ministries and so on should be consulted in any process going forward. 

>> MR. RICHARD SAMANS: Thanks.  Gentleman here?

>> AUDIENCE: Thanks.  I was interested in Bill's slide where he said that we should focus on jurisdiction over data and access to it rather than its physical location.  I'm curious to hear what he recommends to do that.  One of the main problems is that while digital economy considers data as a resource, the digital policy is not considering that as a resource.  It only looks at the human rights angles, which are very important.  But the fact is that data is a personal resource and a collective resource of a community.  At school, the data is collected of the students of the school district and of the country.  As a collective resource, it is said that it should not be enforced localization, there are just people trying to keep their data resources to themselves.  We should read the positions that WTO has got, and that's why they're interested in having data ownership of data talks and not only free flow of data.  Why do we not talk about free and equal access to data for everyone? These are important questions.  Countries are not dumb that they don't understand it, but they have a framework to bring to the table which is not being accepted.  So, data ownership, free and open access to data is as important.  Real inclusion takes place, and then we can actually discuss the issues.  But these frameworks do not include the issues, and that's the problem of the people who don't want to discuss the issues at global levels.  Thank you. 

>> AUDIENCE: Very quick comments on the model.  I heard that last week as WTO minister in Argentina, the initiative has been concluded between WTO, in the press release, this is a public‑private dialogue.  I want to know whether this is a first step for the multi‑stakeholder model, because this is only two stakeholders involved, IGO, and private sector.  Can Civil Society join this initiative for further discussion and research?

>> MR. RICHARD SAMANS: The answer to that is yes.  It's intended to be a multi‑stakeholder process, multi-company, multi‑stakeholder, international process. 

>> AUDIENCE: With all the respect to what was developed in the '60s and '70s, but I don't think we should put the internet first.  Not just my country, but all the people.  I can't repeat the leaving no one behind is fundamental.  I think this is something we should put first.  And a consequence of this may be that we should keep the internet unfragmented and at the disposal of everyone.  I'm with you in the consequences.  I quite like this, as I already alluded to, I like these three tracks.  We may discuss to what extent, really, multi‑stakeholder decision‑making procedures can be developed.  We have ICANN as an example that sometimes works better and sometimes works less, but it works in general.  It was possible based on a rough consensus to agree, it is something that should be picked up and looked at again more closely.  That was the beginning of a new way of multi‑stakeholder diplomacy.  I think there's lots of relevant points that you have collected in the three tracks and also in the principles, but the most important one, try and be as inclusive as possible.  And if people decide that they don't want to be a part of everything that is possible in terms of digitization and globalization, self‑determination is also a fundamental right of people and people in my country and culture, you can't force people to make progress.  If they are happy with the way their houses look and they want to keep it like that, it's their right to keep it like that as long as they don't disturb anybody else.  In that sense, this is something that we need to keep it people‑centered, but of course try to use the benefits that the new technologies offer.  Thank you. 


>> AUDIENCE: Thank you very much.  Richard Hill.  I want to start by noticing that the panel has been heavily stacked on one side of the debate, which is okay.  We ran a panel which was completely stacked on the other side of the debate, but be aware that we have heard mostly one view.  I think it is framed wrong.  It's not about data localization, but about whether or not we should have free flow of data.  Some is legitimate.  For example, tax records and financial records, otherwise you don't have enforcement.  If they're outside of your jurisdiction, you can't do anything about that.  As Richard said, the question is how do we strike a correct balance between free flow of data and other interests.  And my take on that is definitely not in the WTO.  It's not transparent or inclusive.  And Fiona, it's not that we didn't know what was going on in Buenos Aires, but that we got locked out.  And why? WTO never protested about that and didn't make much noise.  How can we create trust? Certainly not by doing what WTO has done.  In fact, we should do the opposite.  Don't negotiate in secret for the benefit of a few big companies.  Anybody remember AFTA? It was Civil Society activism that led to the death of that particular monstrosity.  And I think we were influential in what happened or didn't happen in Buenos Aires.  I think it's interesting to hear that WEF is building itself as a multi‑stakeholder process.  I have never heard one where WEF would be in that category, because as far as I can tell, and it's perfectly legitimate.  It's promoting a particular version of neo‑liberalism, which is, in my view, unabashed corporatism. 

 The first real issue is that we need a global anti‑trust mechanism.  Then we need global data protection norms, and we had a meeting where basically everybody agreed on the convention 108 of the council of Europe could be a good starting point of that.  And then we need a fundamental reform of the WTO if it's going to survive an implementation of the Doha agenda.  That's what's leading to what some of us don't like and some of us like, like Brexit and the election of presidents like Trump.  I'm going to finish, because it was so unbalanced.  Data localization may be the only way to protect citizens' rights.  If I own oil, why should I give it away for free to everybody else and why should the country not get royalties on its oil.  It's not just that.  It's a human right.  My personal data is about me as a person.  Who's going to protect that? You say yeah, if you localize it in some evil place, then they're going to take advantage of it.  But by not getting localized, some evil companies are stealing my data or using it in ways that I never agreed to? Why am I getting advertising? We know the answer, because they are taking my data and using it.  Small and medium enterprises have not benefitted on the whole from WTO agreements and they will not.  We will have the same repeat in manufacturing, which are leading to the election of protectionist people. 

 Somebody said that trade should not set norms, but what about TRIPS? AFTA we managed to fend off.  It's about negotiating binding agreements.  Why do you want to bring spam into WTO? Do you like it? It's about imposing regulation and a certain type of regulation.  WTO was supposed to be a level playing field, and the results are just the opposite.  It's favored big companies.  Guess where it's coming from? Globalization, and totally incorrect norms negotiated in the WTO.  If you ask what they want, they don't want what's being proposed.  They want other things.  They want common platforms and reduction in certain bureaucratic things and that's been well documented.  UMTAT is the only logical body to discuss these issues.  Nothing else makes any sense.  Thank you. 

[ Applause ]

>> MR. RICHARD SAMANS: Thank you.  Yes?

>> MARY UDUMA: Can data be really localized in the face of the internet, the online, the opens of the internet? The issues of the one internet that will not be broken? Can data, can it be localized? Why I ask this question is that if we pass a law and say localize your data in your country, people will say e‑commerce.  They will buy from Amazon.  They should buy from Ali‑baba.  And they will continue to operate in our country.  So, I am a bit ‑‑ I'm happy with this.  I want to support what Fiona has said.  There's a disconnect as far as I'm concerned at the local level.  At the global level we might be doing something, but we need to bring it down to the level so that those gaps will be bridged between the techies or the internet, and those developing IOTs and the rest of them, and the business people.  This is the starting point, as I said.  And I believe this, and I also want to encourage that we'll take this home to our countries.  We'll be able to talk to those as well as about the online businesses.  There should be some meeting point.  And then build capacity between the two.  Thank you. 


>> RICARDO MELENDEZ‑ORTIZ: I wanted to respond to Neil's proposal, but just before that, one of the issues that I think is coming up is the legitimate question of why negotiate frameworks and rules under trade? Is it trade agreements that you need? The vehicle you need for governance of cross border data flows? I think the three‑track approach, you have already a mechanism that would have ensured, hopefully, that you separate what belongs on trade agreements and what belongs somewhere else. 

 Also, I don't think that anybody's doubting, for instance, content's role.  It plays a large role, particularly dealing with capacity issues.  There are economic divide and digital divide, and all of those things that need to be taken into account.  You have the simple response that basically all companies in the world today are digitally enabled.  And they participate in one way or another in the digital economy.  They do require certain particles and they record certain principles to be able to participate in those global markets, but also to be able to avail themselves of opportunities of the digitalization.  The reason why, I think, in Buenos Ares, the intention to start talks towards negotiated rules in the WTO was signed by 27 developing countries on all levels of development. 

 You have to first clarify what's the coverage today, what are the issues with that coverage as well? For instance, in the GATS agreement.  Then you need to find what other issues you need to really put on the table and discuss.  That's what negotiations are about through negotiations that include different levels of development.  And bringing the enabling provisions that would enable countries to participate.  Then the question of whether this is going to lead to multilateral solutions.  We need to see what's the content.  Now again, I think in the real world, what we have today is that word, program, at the WTO, a moratorium on the imposition of duties on transactions, which was extended for two more years.  And the statement of intent from 17 countries to start the negotiations.  I think that provides an opportunity to start this three‑track approach and to contribute to that approach. 

 One quick anecdote is that it was important for Buenos Ares needed to see the impact if the moratorium to not be extended.  And then the discussion for about a day, we had a very rich discussion with all sorts of stakeholders about what it would mean to live in a world without that moratorium.  Basically, the instability that it would bring, this possibility that electronic transactions would be taxed without people understanding, including from the technical point of view, what exactly was it going to ‑‑ what was it that it made happen? And what would the consequences be principally for issues related to tech, and also for the regular use of internet?

>> MR. RICHARD SAMANS: Thanks.  Lee?

>> LEE TUTHILL: Yeah, I wanted to speak up to agree with your points and some of my own.  I will not take Richard's bait.  However, he gets there, I do agree with his conclusions 20 to 30% of the time.  Just quickly, I think that one of the misconceptions I seem to be hearing as I listen today and throughout the day, actually, is that if the internet community has a problem with what governments are doing that might break the internet, I don't think the problem is the WTO at all.  I think that might be a misconception, because the proponents of doing something on electronic commerce on the WTO are also wanting to promote within this. 

 Second of all, I agree principles to be worth anything, I'm very old school.  I started working on GAT in 1983.  I believe in binding principles maybe a little less, but the approach helps that.  Where is the role of the experts who know what they're doing on these things even if you have principles? WTO never tells governments what to do and how to do it.  They have the guiding principle to guide trade and we leave it to governments to work within their own infrastructure and cultures.  That's where the expertise of people who know what they're doing on cyber-crime, cyber security, how the internet could be managed a better, that's where the experts come in is helping governments implement things in a way that meets the principles.  I would be very wary in conclusion to setting up new and additional layers of processes.  I think the forums that have the expertise are out there.  What I'm not seeing are the links.  You need more of a clearinghouse. 

 For example, our friends from EFF this morning? Are you going to be sending a copy to the WTO and asking the director general? I don't see this happening.  Because I'm an IT person at the WTO, I see all these initiatives, and I don't see the communication happening. 

>> MR. RICHARD SAMANS: Anriette has asked for the floor.  We need to close the session. 

>> ANRIETTE ESTERHUYSEN: I think there's a lot more to be done.  I want to commend electronic frontiers for starting the dynamic coalition, but I think for it to have real value, it has to include developing country actors and government actors as well, which is sometimes a challenge for us in the IGF.  It is going to have to deal with diversity of views from Civil Society.  And to flag one area, which as a rights person I feel we're not looking at yet.  We're looking at data protection in relation to trade.  We're not looking at RTI and freedom of information legislation and what the implications are.  The new age of freedom of information legislation gives access to private sector information.  And how can that be ‑‑ how can you solve that not through data localization but actually through agreements about how those rights can be enforced in the context of distributed storage and cloud computing?

>> MR. RICHARD SAMANS: Okay.  We've reached the end of our time.  If you can make it very quick?

>> AUDIENCE: I want to make it short.  I'm on my way out of the IGF.  I've been a frequent visitor here, and I value this community.  But being the parliamentarian, walking around in policy circles and government circles as well, I really encourage you, especially in Civil Society to think about which stakeholders are not here and why.  I also encourage you to think about positive agendas.  I encourage you to think about the global dynamics that are going on that are not in favor of opens or Democratic principles or in favor of universal human rights.  If we do too much naval gazing, and if we start to look too much at the imperfections amongst ourselves and the arguments between ourselves, we may be losing the bigger picture.  I would hate for that to happen, but there's a lot at stake.  It's not only what rules do we need? It's also that we're up against forces that are happy to not have any rules at all.  Make the internet the domain of the fittest who are going to be principle‑less, and sweep aside every right, whether it's a human right or digital right.  I wish that the bigger picture and strategic thinking and bringing on board people who are not in the seats here would be part of the process here.  I'm happy to help, in case people think I could contribute to that. 

>> MR. RICHARD SAMANS: Thank you.  Professor Drake, why don't we give you the last word? See what you've taken out of the conversation?

>> MR. WILLIAM DRAKE: You're already helping with the work you're doing through the European parliament, so thank you for that.  I don't want to change Thomas's house.  I think that the Swiss houses are very charming, and if people choose to have their houses look a certain way, and they're willing to bear the full consequences of that, then that's fine.  The suggestion here was to create a process to promote dialogue and interaction, and to get people focused on understanding the concerns that each other have expressed, and they are motivating some policies as a way of trying to move towards greater consensus.  I think that's antecedent to anything happening in the trade space and hopefully we will continue to have dialogues like this going forward.