IGF 2017 - Day 4 - Room XXVI - WS188 Let The People Speak: Using Evidence From The Global South To Reshape Our Digital Future


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. 



>> MATTHEW SMITH:  Good morning, everyone.  Just one or two or minutes while we set up the PowerPoints and then we'll get started. 

(Standing by)

>> MATTHEW SMITH:  Hi, again.  Thank you for your patience.  It seems like we got everything sorted.  Welcome to the session Let the People Speak:  Using Evidence from the Global South to Reshape Our Digital Future.  This is a nice cozy little group.  It's too bad we can't come together and sit around in a circle but that's fine.  It is what it is. 

I'm Matthew Smith, your last‑minute substitution moderator.  And if anyone watches TV, I'm not that Matt Smith, I'm another Matt Smith.  I work at Canada's International Development Research Center. 

We lost time and I'm not going to get into a big introduction.  We have been proud to support this work that's going to be talked about today.  Today we have panelists and discussants.  I'm going to really quickly introduce them and then we are going to get right to it. 

So where is Alison?  Alison Gillwald from Research ICT Africa, University of Cape Town; Helani Galpaya, from LIRNEasia; Alina Gurero (phonetic), Darcy (phonetic).  And for discussants we have got Daniel Abadie from Argentina, the Digital Government Office Ministry of Modernization, which I look forward to hearing more about; Adeel Sudamen (phonetic) from the Information Society Division, African Union Commission; Anriette Esterhuysen from APC; and Scarlett Fondeur to my right at UNCTAD at the Partnership on Measuring ICT for development.  We also have Enrique (phonetic) as a rapporteur, and Chenai Chair as our online moderator.  I'm going to turn to Alison for her presentation. 

>> ALISON GILLWALD:  Welcome, firstly, everyone.  Thank you very much for joining us.  I think on sort of the last session on the last day.  I'm sure everybody is ready to go home.  This survey, the data that we are going to show you brings together studies done across the global south, 16 countries all in all.  There's slim progress and we are hoping to do a number of further studies. 

We have studies still in the field in Asia being wrapped up in the field in Asia and we are very pleased to be doing another two countries in Africa and one country this Latin America and Asia as well.  That will take us very close to 20 I think which is back to where we were just Africa a few years ago. 

All of these regions have been doing surveys for over a decade but this is the first time we have used the same instrument to do nationally representative surveys.  These are important because in our prepaid mobile markets there's no way to get the number of people connected who they are and what they use them for. 

In a prepaid market you cannot identify the user of that SIM card or what they're doing on the Internet as a woman.  It's very simple things like connectivity figures that we have such diverse data on from GSMA or ITU and our data from the global South is really only possible because you can identify the actual unique user of a SIM in this case or a user of the Internet by identifying the number of duplicate SIM cards. 

Even as we move to a more competitive environment and move into the Internet we are still finding very high numbers of duplicate SIM cards in many countries.  Used to be very high with mobile phone usage but now it tends to be pretty high.  This data is important in terms of that. 

And so it's also the data is really focused on the after access, moving beyond the simple connectivity issues, looking at the challenges of the day‑to‑day environments, OCTs, voice and text surfaces.  Looking at the effectiveness and inequality that effect the use of the Internet once one is connected or prevent one from being connected. 

So that's the important contribution as well.  Who isn't using the net or why aren't they using it optimally.  If we can click on the slides.  I'm not going to say very much about the methodology and there's people that would like to come back to it.  Basically the survey is nationally representative.  It has a 95% confidence measure.  Essentially we work off the national census frame.  At the household level we randomly select an individual from the household so we get both household individual data as well. 

The data is weighted once we get it according to various methodologies that have allowed us to use relatively poor samples with high levels of accuracy, and to get data from urban areas where we get data but also rural areas but weighted accordingly at the national level.  If you can move on from that slide very quickly.  We have a couple of slides. 

We don't claim for these studies to be regionally representative because we don't do enough studies in these countries but they are indicative of what is happening in the regions.  Africa we can see that we have very high mobile phone access from Kenya, South Africa.  Our lowest Mozambique, Rwanda, Tanzania all consider to be lower.  High is Latin America and low is Asia.

>> PANELIST:  In Latin America ownership is not an issue now compared to what it was ten years ago when we were doing this kind of research.  That's why maybe as Alison said Latin American countries would be more similar to South Africa in those terms but maybe we will have a look later at this ‑‑ (speaker off microphone) ‑‑

>> PANELIST:  I think there are some slides later is basically that, you know, phone access it high but the problem is they are feature phones.  India as over 1.2 billion people.  If you think about Asia, south Asia is pretty much wiped out by the India numbers.  If you want to get a sense of what is happening in south Asia look at the India numbers because it will be very close to what India is.  Next?  Next slide. 

Okay.  I mean, Internet use is abysmal, let's put it that way, you know.  Under 20%, India around 19%.  Bangladesh around 13%.  Cambodia is the only high number.  Cambodia interestingly in southeast Asia is very similar to the Myanmar numbers which are not shown which is again in the high 30s as a percentage of the population, newly entered into the market but abysmal for a country that has reformed its sector 16 years and going now.

>> PANELIST:  The Africa data, again we have this enormous diversity between the different regions, South Africa still below Latin America interestingly but then these least developed countries having really, really poor outcomes.  But basically we have Rwanda, eight%, Mozambique 10% and Tanzania 14%.  Also interesting because so much effort has been made to address the issues of connectivity, particularly with Rwanda.  World Bank, money poured into it with projects associated with it, and there's so little penetration which I think is emphasizing on the constraints and affordability issues that are still affecting it.

>> PANELIST:  Are we talking about fixed line or mobile?

>> PANELIST:  Basically Internet usage is most but across most of these countries it's 99% mobile.  The figures aren't here.

>> PANELIST:  I'm from Latin America.  I wanted to point out when we were thinking of which countries to choose to doing the surveys we thought about having one with high income levels, middle income levels and a little bit low income levels.  So all the results that we will see we will more or less reflect that ranking because Argentina is always on top then Peru and Colombia.  This is the same as we are seeing here in the case of Internet use.  Next slide.

>> PANELIST:  Perhaps if we can go to the next slide and average Internet years of use, we have some Latin America and African data on.

>> PANELIST:  As I mentioned before in Argentina we find users with more experience in terms of Internet.  This is important as we have been listening in these sessions through all these days that, for example, in terms of cybersecurity there are more threats to more recent users.  We can see these countries that has less experience in Internet use are more ‑‑ there's more danger in terms of what they can do, what websites they can access or anything like that.  So this is something we will discuss maybe later.

>> PANELIST:  Yeah, I think for South Africa as well, South Africa a few of the countries had initial take up with PC's but it was really around 2010, 2012 that we saw the massive up take of a mobile Internet or Internet because of the access to mobile and broad band that's lived to the uptake of Internet so again we have people with the poor countries with much lower, less experience and probably the same group when we look at the data more deeply that we are going to see to use the Internet is optimally people that have been online longer or using devices and software that is only really feasible to use on a PC. 

Can we have the next slide, Internet for work. 

So I think this is an interesting slide.  We have a whole set of questions on micro work, very interesting questions on micro work that provide some of the first data on national representative data on what is happening in the area of micro work particularly in Africa there has been work done by LIRNEasia on micro work and digital platforms for some time but this is the first data.  Many looked at it from the platform side, the supply side. 

The use of micro work is extremely limited for Africa and puts serious questions about the hype of micro work being able to provide a solution to structural unemployment across the country.  The figures are so low when penetrate is only 7 or 8 or 10 or 13%, then the fact that you have 4, 5, 7 percent of people doing micro work is not surprising.  Next.

>> PANELIST:  So 10% of Internet users number is actually a really large number when you look at it from another side, who do micro work.  If you actually look at my co work platforms, India is the second largest provider of labor in work outside of the U.S.  And the U.S. has Amazon, that kind of work.  But if you take the top four platforms, India is the largest provider.  And yet it's only about 10% of the total population between the ages of 15 to 65.  I think that's the more interesting thing.  And Bangladesh is trying very hard right now with international donor money to roll out training people online doing digital work, doing websites and so on.  We'll see.  Basically these are small blips at the moment.

>> PANELIST:  Again, in Latin America this area of using the Internet for work is not that much, has not developed that much.  Basically there's the Internet is used for, I don't know, taxis or maybe some kind of translations but there's not so much of experience in this area.  I think that even there were some issues in terms of how to regulate these activities.  Maybe Danielle can comment later about these applications.  I think there were restrictions in Argentina for this.

>> PANELIST:  Do you want me to do it now or later?

>> PANELIST:  Sure.

>> PANELIST:  Okay.  Well, first, there were some restrictions in the previous administration.  Actually this changed in the last two years.  This year we passed a law for entrepreneurs to create in their own company in the first 24 hours, technology is getting bigger. 

The companies for Argentina, for us it's a great challenge looking at the four companies grow up in Argentina although all the challenges are blocking what is given them.  There's a lot of work outsourced from Argentina plus we are in the same time zone as New York so maybe a huge company is outsourcing Argentina from India. 

If you think about Latin America, 60% of the homes in Argentina has access to the broad band.  Mobile penetration of 98%.  We are 44 million people but only 27 million people use Internet.  We think how can we fill that gap, something more besides infrastructure is happening.

>> PANELIST:  Next slide.

>> PANELIST:  I'll just say in Asia you might as well look at social media numbers or look at Internet numbers.  They're basically exactly the same, like plus, minus, 1%.  Social media is the killer in all the countries, these are just three of them.  In Cambodia it might be higher than the Internet number.  Social media includes Facebook, Facebook messenger, anything where you can basically broadcast including Twitter, what's app as long as you're using it for broadcasters.

>> PANELIST:  In Latin America the figures are there.  We have the highest levels of social media use in Colombia and Argentina.  And then Paraguay is the country with the lowest percentage of social media use.  What we found something interesting was doing the field work and it was that when we were asking people if they use social media like Facebook, whatever, they said yes.  When we asked them if they used the Internet they said no.  They were not aware of the fact that social media works through the Internet.  That's why we had to make distinctions about this to see what people understand.

>> PANELIST:  Can I just ask a question for clarification.  Is that the percentage of the population that uses social media or the percentage of people that have Internet access?

>> PANELIST:  No, it's just the percentage of the population.  This is why we didn't put a filter.

>> PANELIST:  Just quick also on social media use was there any question around frequency that was factored in here?

>> PANELIST:  Yes.  I don't think we are showing it.  I'm showing it for Asia but there were lots of questions about frequency.

>> PANELIST:  Yeah.  And so it's a massive data set and there are lots of questions around frequency of different things that you do on the Internet.  So there's certainly a lot of granularity we have got there.  Just to say again South Africa 42% of people actually using social media.  Much higher than Nigeria or Kenya that followed just below 30%.  Even South Africa 42 is way below your lowest levels in Latin America. 

If we overlay these with prices which we haven't really done here, Asia's prices are extremely low.  A lot of these African countries, for example, these four countries here and in fact all the middle section all have very good prices, very low prices.

So and Latin America has very high prices.  But it's really reflected here.  Many of these countries are extremely poor.  Arguably may never be able to afford to use the Internet at the amount of Bandwidth that we need as long as we continue to use purely commercial sort of GSM models.

>> PANELIST:  Next slide.  I'm not going to get into detail because you can't see it from your slide but when we post it on the website it's just got the sample sizes for the various countries, the populations, the percentage to the population size, the break downs for urban and rural, and female and male.  The next slides look at a little bit more detail.  I think the next one is mobile penetration by gender in Africa. 

Just to look at where we are seeing these biggest disparities, there's obviously a reasonable concerning gap in Ghana between men and women but the really interesting very, very gap is in these poorer countries.  We see a very big gap in Rwanda, 60% of men, as opposed to 37% of women.  In Mozambique 52% of men and 32% of women.  This supports our earlier rounds of research that shows there's not a big gender divide between men and women of a similar income.  In fact the drivers of inequality are education and income, and women are concentrated at the bottom of the pyramid not receiving education and income. 

So that's where our policy attention needs to go.  By penetration you've got a bigger gap there with men and women in South Africa.  These bigger gaps sorry, smaller gaps then you saw in the Internet over all because again to support the idea that it's income driven is men and women of a similar income who can afford a smart phone are the people accessing this first half of the people with cell phones. 

If we look at Internet use by gender you see that really big gap in Rwanda.  Very interesting, close in Kenya, sorry for that, that's men and women.  This index down here, if you're looking at the male it's the red bar and the female is the black bar.

If we go to social media, we can again see in countries where we are getting higher penetration where there's similar income between men and woman we are getting quite similar uses of social media where we have bigger inequities, we have differences. 

Interestingly, Nigeria has a big gap between men and women on social media which we need to look at more closely because they have this very dynamic content industry and large numbers of women in the workforce but they still have a very predominantly rural population so that might be explained by that.  Just for work, are you using the Internet for work‑related by gender?  We see predominantly men are using it for work.  We also see predominantly rural people.  And in Nigeria we have more women. 

And again by location we do see more urbanized countries, less of a divide between urban and rural.  In Tanzania a gap between the two.  Some of these countries are still struggling to get the kind of coverage that is available in many other countries on the continent. 

This is rural, for Internet.  The education I'm going to leave because it requires a bit of more of a discussion there. 

I think this is the rural one, it's actually duplicated.  I'm going to show you one quick point about the micro work in Africa because we are not going to have much time for discussion on this but there's pay whole lied, whole data set on this.  We see these very, very low levels of micro work in Ghana, for example, only two% of women whereas you've got of the people only two% of women whereas 55% of that is male.  Again Kenya very low, three, Mozambique a little bit higher.  Nigeria around 10%.  And Rwanda only three percent, nearly four percent of people doing micro work of that very small number of people on the Internet and 77% of them are men. 

So in Rwanda across all of these measures we have really high discrepancies between men and women.  I'm going to leave it there.  We are definitely out of time.  Social media we discussed.  Some limitations to use.  We need to look at this in different countries and the slide is far too busy.  Ghana is at the bottom in the red.  Ghana had large numbers of people who said they don't have limitations.  Kenya, 16% said they don't have limitations, 19% in Nigeria, 21% in South Africa.  If we look at the cost of data we actually see although we will identify Ghana as having quite good prices it's still a barrier for large numbers of people. 

The barrier is still high for large numbers of people.  For Rwanda one of the big things is a lack of content in my language.  It's a very rural population and people struggling to find the Internet meaningful or being able to use the Internet.

Interestingly because they seem so offline is it has one of the highest levels of concern about using the Internet around malware and certain things like that.  So whether there's awareness that's keeping them off, it's something that requires further investigation.  Still large numbers find it very difficult to use.  Rwanda a lot of people are unable to use it.  I'm going to leave it there.  We require more discussion on that.

>> PANELIST:  Next slide, please.  Thank you.  So we know the countries, the data is not being shown because it's so fresh the last bits came in last Friday so it will be integrated as was last year's Myanmar's surveys so I'll talk about the Bangladesh, India and Cambodia.  The four countries we studied is 23% of the world's population and sadly 36% of the world's poor.  Solving India's problems would really help in this whole region.  Next slide, please.  Sample slices are 5,000 for India, 20,000 each for the other countries.  Next slide.

So when it comes to mobile phone ownership we are doing quite well in the mid ‘90s even in rural areas.  People have had mobile phones.  Liberalization started back in 1996ish so it had a long track record.

At the household level so when asked do you have a working Internet connection of any type, you can see the red is the total number, it's quite low at household level meaning you can share somebody else's, Cambodia five percent, Bangladesh 11%, there's a big urban rural gap that you see.  The dark gray bar being urban and the other one in rural.  Next.

So what is this Internet connection at household level?  All countries pretty much there's hardly any fixed ADSL.  It's mostly mobile.  The exception is Cambodia which interestingly has other technologies other than mobile coming into the household.

Next.  Access usage and ownership shows the pattern you would expect which is access is high as we said for mobile because do you share phones as well.  That's the lightest color.  Usage for voice and data, this is really voice, the middle bar, is low in access but you do use it quite well.  Ownership is lower of a device that you can call your own.  Next.

So this is the total mobile phone ownership.  I think we have seen these numbers before.  There's an urban rural divide big in Cambodia and India, close to 30%.  Next.  When it comes to gender again we will see a gender gap it's about 34% in Bangladesh for mobile phone.  36% in India.  That's half as likely to have a phone as a man.  Type of mobile owned, it's mostly the basic phones. 

And then you've got the feature phones which are the Internet enabled forms but not touch screen, fancy smart phones.  Lower penetration of smart phones except again Cambodia has high penetration of smart phones.  India sees this generational shift where most people had the very basic key pad phones then they skipped the other slightly better feature phones and are jumping straight on to the affordable smart phones many manufactured in India for very affordable prices.

Next.  Owning your own device is one of the biggest barriers because the Internet experience especially using something like where you need to sign in like Facebook is a personal experience.  People are not willing to take someone else's phone to log in to Facebook.  Sharing behavior is low when it comes to Internet.  And no need for a phone is one of the reasons that they give for not owning their own phone.  Some can't afford it.  Also in these graphs they say I don't need it and that's a harder problem to solve where people don't see the need.  Next? 

Internet access let's skip this.  We talked about low Internet access.  Let's look at urban rural divides.  Again we are reaching the 40% kind of divide between urban areas and rural areas.  In India it's close to 50%.  Next. 

Male/female, again for Internet access.  Before we showed for phone ownership.  60% in India, extremely high.  34% even in Bangladesh 52%.  Next.

And this is one of the big problems.  People actually don't know what the Internet is.  When we ask them do you know what the Internet is and we define it sort of rather broadly to include anything from browser use to using maps to use Facebook to what's app, the broad number of people who say yes is shown here.  In India about 35% of the people know what the Internet is, which means all the rest, 65%, actually do not know.  Similarly in Cambodia and then again Bangladesh about 37% of the people don't know about it.  Next, awareness of the Internet by men versus women there's a gap in awareness, 40% in India, 30% in the other two countries.  Next. 

First time Internet use is mobile which is why I kept saying getting a good functioning mobile smart phone into the hands of people is essential.  Desk tops are used in India and I mean we should break this down but I suspect this will be people who work outside at a job that requires a desk top, not people who have it necessarily at home.  Next. 

We can skip this.  Okay.  When using the Internet most of the time is spent on social media.  As I said, social media is the killer app when we asked them this is shown in the data, social media, and then entertainment.  And chatting, chatting is the what's up and text. 

We have a whole module on social media and behavior, who are your friends, how do you accept friends, did you unfriend anybody, why, et cetera.  Just to give one thing what is the information that you share on social media is something we asked.  Can others see this information like gender, like age.  A marital status, religion.  And this becomes important in what kind of Internet experience people have when they do get online. 

And we see this in Cambodia, they continuously share less to the public on social media about gender, religion and age.  Whereas in south Asia they are much more willing, certainly more than Cambodia.  Next.  Just on the phone SMS is frequently used, calls, chatting, this what is we have seen across Asia.  Next.

And then there's a whole set of questions around capacity.  Are they able to use the phone, are they able to change the settings and so on.  Just to highlight one is that most when they need help they get it from friends or family.  So it's not that they go to the Internet and find things or ask others.  Friends and family are the first line of technical support when you can't jiggle with your phone.  Next. 

Oh then we have a specific question about apps for people who could actually install apps.  As a percentage of smart phone owners can you install apps and what kind of apps.  We see social media and messaging and entertainment apps and the most downloadable, installed and used.  Mobile banking is another module, I wanted to highlight that in all the three regions these modules in the questionnaire. 

Mobile banking is heavily used in Bangladesh and you know that's the dark sort of blackish 33% all the way on the left‑hand side.  But again low penetration.  And much needed because this is a hugely unbanked population.  So a long way to go.  Next.

And then we asked do you know about all these platforms everyone goes on about?  Do you know about residential surf like Airbnb, buying and selling like Amazon and so on?  This is the awareness.  And India has not too bad.  About 50% of the people know about transport, gig economy work.  Actual usage, again India has a thriving E‑Commerce market.  You see that use pretty high in India, very low in the other countries.  Reason for not using the Internet is lack of understanding what the Internet is and knowing how ‑‑

>> PANELIST:  Can I just throw in one question because this will be very interesting for me.  Is there any awareness of cross‑border big economy activity?  Because this is a topic that has come up quite often at IGF.  Do we have any data?

>> PANELIST:  We can't differentiate between cross‑border in country because the question is when it's localized includes cross‑border and in country apps so Airbnb and something that's let's say particularly used in Myanmar so you can't break that up.

>> PANELIST:  Thank you.

>> PANELIST:  I'll stop there.  We addressed problems, lack of understanding.  Then we asked the people who are using now would you use it more and why don't you use it more.  Huge variation by country.  Interestingly in Cambodia this light gray area is speed.  So we did a whole lot of testing.  We sent the mobile phone to the field with phones which actually tested so that's about 24,000 data points for India which we analyzed.  We will get a sense of whether it's perception or it's shown in the data for the speed test.  And Africa as well did that so that will be another whole set.  Anyway, thank you.

>> PANELIST:  So now I will show some data for Latin America.  I'll just skip this.  As I said, mobile phone ownership is not that much of an issue.  In terms of smart phone penetration in our countries, especially in the country where I come from, Peru, that is quite high and the same thing in Paraguay.  In the rest of the countries it's much more equal. 

Can you click, please?  No, no, it's okay.  Yeah.  Here I wanted to show what we have called mobile usage, traditional versus non‑traditional.  We have in the question there more or less nine traditional mobile uses which are making or receiving calls, sending or receiving SMS, it could be pictures. 

We have non‑traditional mobile users, they use mobile applications.  What we can see here especially in rural areas of Colombia, Paraguay, the difference between these more traditional users and non‑traditional mobile users are much higher.  Also in terms of traditional users it is like we can see the levels of these users are more equal but there's a lot of variations between non‑traditional and mobile users.  And we were also curious about what was the main reason for not owning a smart phone. 

And some things have changed when we compared these results to the previous research that we did some years ago. 

And we find that, for example, in Paraguay and Guatemala the main reason is the respondents do not need one.  For example, in my country affordability is still an issue because 30% of the respondents said that this was still a problem for them.  And in the rest of the countries the issue of not needing one and the affordability are almost the same.

As we have already talked about I just wanted to focus in this government‑related activities online.  So I've heard in the different panels that we have been through these days that there are many E‑government strategies, plans, digital agenda.  When I see these results I started to question what is going on as only in Argentina these activities are close to 25% while in Colombia, Guatemala and Peru they represent less than 20% so maybe we will have to look deeper into what is happening.  Again we will explain what has been doing in Argentina to address this problem.  And again I wanted to show here what was the main reason for not using the Internet. 

And we have different results.  For example, in Peru there is no interest for 50% of the respondents they're not interested in the Internet.  And while in Argentina the main reason is that again there's no interest. 

So we have to listen to them.  Maybe you can tell us already what you've been working on, all the trainings that you are doing, your organizing throughout your country.

>> PANELIST:  First of all about government services in Argentina yes it's true beside the national level Argentina we have local governments.  When we took the office two years ago at least a thousand.  So we are working with local governments to bring services.  On than side when we go forward it's a local government making all things digital.  We are also creating our digital gap between access to the Internet if we are getting digital services online we are excluding them.  So what we did we create a national digital inclusion plan.  In Argentina 30% of our citizens are in poverty. 

We are working with people basically teaching them how to use the Internet, not also literacy but how to do services online, how to look for a job, how to learn, how to get e‑mails.  It's extremely true what Aileen said this connection between what is Internet and the access to Facebook or what's app, if we ask the Internet the obvious answer is no but they answer about Facebook is yes.  I know it's generational thing.  Most of us, nowadays no one texts to the Internet. 

Using the Internet as a productive thing, we are working on that.  We are working on parents on the challenges between surfing the net.  So we set up a plan like a million people each year getting them online.  We believe there's about 10 million in Argentina that don't use Internet and we split the connective. 

One thing is infrastructure that we are working and, and then using the Internet to get jobs and development.

>> PANELIST:  So it's good to hear that there is some work from the government in getting to know the reality of the users of the citizens.  And many things can be done if we look at these, for example, this kind of evidence, to know what is the main reason that prevents people from using the Internet, mobile phones or whatever.  Good news is we have all these questions for the three regions. 

Can we continue?  Then again we have social media data.  And once again there's a huge difference between urban and rural areas in Peru and also in Paraguay. 

The rest we wanted to show also some gaps between males and females.  But I would just leave it there maybe because the results are almost the same as it's again in Peru, Paraguay and in Guatemala where we find more differences.  And in the rest of the countries we don't have much problem.  So thank you. 

>> MATTHEW SMITH:  Thank you for that.  This is a nice small groom so actually instead of doing something formal and opening it up to the floor I think it's good to raise your hand, click the red button and ask your question.  If there are any questions specifically on findings and methodology. 

>> AUDIENCE:  Thank you for this very informative presentation.  First of all, as an organization I would like to see statistics for the entire continent.  I know there's work under way to include other countries.  I have three questions actually.  One on the methodology.  Maybe I missed that part.  Was it done through telephone interview?  I wasn't sure.  If you can shed more light on that. 

And on the findings was there an effort to cross‑check the findings with the existing data by other institutions and then an explanation why the results are different from existing findings by other institutions? 

My last question, do you have a plan to have this as a one‑stop shop for all ICT related parameters so that we just go to your survey and then you find everything to do with ICT and we don't have to alert other institutions to get the data from.  Thank you.

>> PANELIST:  Yes, regarding the methodology, all the numerators were in the field.  And before we did many trainings in each of the countries.  And we used also in the majority of countries tablets so that we could register information there.  In Latin America specifically we couldn't do it this Guatemala and Paraguay because of the technical limitations but in the rest of the countries we traveled and did the trainings.  More or less in our case the survey lasted one hour approximately.  In some cases less, even half an hour.

>> PANELIST:  So it's not a telephone interview, it's a face‑to‑face visit to the house randomly selected and then go back to the house to the randomly selected individual.

>> PANELIST:  So there is a very rigorous methodology to ensure that you get countries that the households and individuals are sampled and you have to go back, et cetera.  What I wanted to speak about a little bit more was the complimentary aspect of the data in terms of the data sets and data collection.  And possibly as I said before actually this is the only way you can get this data and it doesn't actually exist anywhere else.  Just to speak about the complementary nature of it.  If we are to bring together a few critical questions into the national census we have at least a set of critical questions, 5 or 10 on ICTs, these could be used as verification points that we can then do much deeper in-depth surveys in between so obviously we are only doing these relatively small samples nationally represented.  Annual surveys we can get five or so questions in.  We can get these in between larger surveys.  Also into the translation into languages and stuff make it a bit longer.

But what is important for us is to try to get some coordination and that's why we work closely with the ITU and national census offices where we can and to try to use the data in a complimentary way but also for verification of the data out there.  As we said the supply side data is unable to give you certain data on these prepaid mobile markets in particular.  So this really is the only way you can get this data.  And you can challenge or you can check the supply side data.  This is something that GSMA over the last few years has acknowledged itself. 

We are getting these percentages that have been quoted by our governments that are still quoting.  Even the UN statistics are stalled using those figures.  It's only through the qualification of the demand side data that you can get these figures.

>> PANELIST:  I just want to add one very quick thing.  We think in our agency census is way too infrequent, it's every ten years.  The model and Internet market is like fast‑moving consumer goods.  Every two years is labor surveys, and income and expenditure surveys.  That means you put some questions into those.

>> MATTHEW SMITH:  Great.  Andrea, one question?

>> ANDREA:  It's more discussive input.  I can reserve my responses if you want to take more questions from the participants. 

>> MATTHEW SMITH:  I think it's fine if it follows what they were talking about.

>> AUDIENCE:  I wanted to reiterate what has been said about the value of this data.  So as Matt introduced me I'm from the association for progressive communication.  And we work in both ICT development, we are a global network of Civil Society organizations around the use of ICTs and the Internet for social justice and development so we do ICT development work and we do human rights work.  For us this data is just indispensable.  I'll use one example, Rwanda.  I'm not sure if you noticed how Rwanda scored.  It scored low on Internet usage. 

We know it's a very largely rural population that scored even lower on social media usage.  Rwanda is considered one of the big successes when it comes to ICT for development.  Donors are very attracted to investing in Rwanda and Rwanda government is very successful at convincing donors to invest.  And there's general acknowledge that Rwanda doesn't have a very open media and freedom of information and expression environment that is usually skewed on the basis that Rwanda is doing so well in terms of economic and social development.  This is a very interesting example. 

If you look at the lines for affordable Internet index, South Africa scores one step down from Rwanda on affordability but South Africa scores very well relatively. 

Of course there's a huge income level difference and a huge rural urban distribution difference.  And it's these differences that we need to be able to see, to understand, to make sure that our advocacy efforts plus our investment in dealing with certain gaps like the gender gap in access actually makes sense. 

So I just think it's really, really indispensable.  And that's coming from those angles but I think there are other area as well such as if you are working in ICTs in education and you want your projects to have impact beyond the polarized pilot, you really need this type data in order to design your initiative, design your effort so that it makes sense that it reaches the right people and also then to be able to learn.

And that's my final point, we are part ‑‑ along with research in ICT, there's an effort to measure Internet access on the point of openness, accessibility and the multistakeholder process.  And we can come up with indicators framework but if there's no data take actually addresses some of the questions that we are asking around content creation, for example, around how people are using the Internet, we are not really going to be able to get data that allows us to learn around what is happening to the Internet.

So really I want to commend you for this and I hope it continuing.  I think it's good to get the questions into the census but I think Helani's point it's not frequent enough.  I hope you can cover more countries, that's what I would like.  And what the plans are, if you can tell us what the plans are for that. 

>> MATTHEW SMITH:  Respond?  Okay.  So we have one more question and then I'm going to turn to another discussant. 

>> AUDIENCE:  Sorry, I was late to this session.  Just on this question about how to capture the data and through censuses labeled, it's a point that complements that also with the enterprises because they provide data that people cannot.  This whole area is of growing importance.

The other thing I think we need to also strengthen the not just the bi‑lateral support here of two countries but we need to bring countries together also in the whole statistical challenge, and we have now got the member states to agree that we should set up a working group on measuring E‑Commerce and the digital economy for once to bring statistic measurements of the member states to dialogue. 

They want to use the data but they're not the best how to collect them and analyze them.  ITU has its groups but they cover relevant to ITU and we have to call on you to play an active role in this area and we will to have call on donors as well to help make sure that they can come here and participate. 

>> MATTHEW SMITH:  Thank you for that.  So let's turn to ‑‑ oh we have another ‑‑ okay.  Please, go ahead. 

>> AUDIENCE:  Hi.  Eshan (phonetic) from the Internet governance project.  My question is there's this large group in the surveys that say that they're not interested in using the Internet.  Is that an awareness issue?  And do you see mobile service providers or ISPs or mobile service providers in these countries targeting those groups to bring them online?  Is there a market incentive for them to do that?  Or are they trying to better services for the urban population?  One would image there's a huge untapped market for them and incentives would align them to reach out for these markets.  If 60% of the country says they don't see the Internet as useful, why are these private businesses not trying to get those markets?

>> PANELIST:  Yeah, I mean this is the next stage.  Up sell, sell more services to the urban people that you've been selling to, and then get the new consumers.  Clearly, this is in the strategy of mobile operators.

Just to clarify that I don't think what the Internet is among people like in the whole population, 15 to 65.  If you probably go down lower you'll have higher numbers.  And I think the financial priority is there.  If you look at marketing you see that, this very Facebook‑centric advertising price sensitive stuff.  Another reason people are not getting online is I'm afraid of what content my kids, et cetera.  In Sri Lanka selling a dongle preloaded with content blocking software already preloaded into the first dongle they get.  So this is something we very much see in the strategies because you have to do this, right? 

Just to respond to, I think you're right, the price level data is really important, formal and informal.  It's not only important to understanding what the level of digitalization is and E‑Commerce but have to understand the challenges of digitalization.  If you look at a job comprised of many tasks, there's a subset of these tasks that are going to be automated and lost which is going to be a huge impact even when employment and the nature of the firm will change.  So understanding this impact at firm level I think it's really important.

>> PANELIST:  Thank you for these questions.  And I wanted to introduce one to the point about the Internet firstly from a methodological point of view.  We had this question of people said they don't know what the Internet is but they're actually using it.  That was one of the challenges.  From a policy point of view if you have people online it doesn't really matter, as long as they're using it what they call it.  So an answer is now like a default.  If people said they don't use the Internet and then they use Facebook on the tablet it's more taken care of. 

It's more interesting of the people saying I had no interest or it can't do anything for me.  This is a classical question in survey work how do you ask people about things they don't know?  There are techniques but they are very time consuming so you show people various things and they say oh I would be interested but that we don't do because it's so difficult.

Just to say I think generally we do see across all regions, especially the regions why we have these very low levels of Internet take up and these high levels of unawareness of interest.  We do have extensive strategies by operators’ particular products and things that we mentioned here but I think if you look at the GSMA connected women's study, it's overtly indicated that getting women online they're going to address problems of saturation in their markets because these are the people who are largely the people who are unaware or can't afford the services, et cetera.

But the other thing which is a much bigger policy challenge is even when you've addressed your nice branding issues you're still dealing with fundamental issues, the fundamental challenge for us is a human development challenge.  People are not able to use these services optimality because of the socioeconomic inequalities that exist there.  I think you see that.  The Rwanda data that was referred to, for example.  You can make all these issues but unless you address the fundamental issues of inequality there's not going to be any digital equality. 

>> MATTHEW SMITH:  Okay.  We have about 15 minutes left and I want to make sure we get to our last two discussants so I'm going to turn to Scarlett.

>> SCARLETT FONDEUR:  Thank you, very much.  So I'm going to try and be very quick.  I'm here to present not only UNCTAD but the partnership on measuring ICT for development which some of you might be aware of.  We are going to present progress on ICT statistics in the next UN statistical commission in 2018. 

And in particular we are going to talk about how the global indicators framework for sustainable development goals has 232 indicators but only seven ICT indicators are included, and this is despite the fact that ICTs are recognized as a development enabler. 

This goes back to the fact that Alison just mentioned that we have to be able to use this information to gauge the different levels of development both social and economic.  And it's important that all areas where ICTs will play a role are measured and monitored beyond what is this in monitoring framework. 

The partnership established in June 2017 a task group on ICT for SDGs which will aim to propose a thematic list of indicators that countries could use the measure ICT availability and news in sectors that are relevant to the SDGs and not covered in the current SDG indicators framework and these might include indicators on skills, E‑Commerce, financial inclusion. 

And we will also aim at improving the availability of desegregated indicators and desegregation would mean by gender, by income level, by geographical measurement.  And finally this task group is opened to all members of the partnership which are mostly international organizations but also other interested agencies and stakeholders such as IDRC and LIRNEasia. 

So I take off my hat from the partnership for development and put on my UNCTAD hat.  UNCTAD is the UN conference on trade and development.  Our focus is really on the economic aspect of things.  And the information economy report of 2017 this one focuses on digitalization which was mentioned now by Helani. 

And the second chapter of this report focuses in particular on the issue of measurement.  And the rationale behind the report is that we are at a turning point in the global economy that has been enabled by technology.  We have emerging technologies such as artificial intelligence, the Internet of Things, which all of us are aware of because the IGF is for the ICT community but not necessarily the policy people are aware of these implications.

And while the statements on the transformative and disruptive potential of ICTs might appear to be self‑evident the IR points out that we ignore more than we actually know about the evolving digital economy and fundamental indicators that are still not widely available for developing countries are the extent that enterprises have affordable access and whether they make productive use of them. 

And we also know that while ICT up take is improving there are wide gaps in the ability of businesses and individuals to make effective use of them, and in least developed countries women and micro small and medium size enterprises are likely to benefit from the digital economy.

We also know that while ICTs can help businesses become more efficient and better connected and research has shown its potential impact and increase in productivity, and there is relatively good data on the use of technology by enterprises, very few low-income countries measure the use of ICT by enterprises.  Where data is available, we always see a lower proportion of small enterprises ‑‑ tasks online, et cetera. 

So what do we need to know?  We need to know more about not only the way that enterprise uses technology, we also need to know about the impact of technology in the labor market.  We have ILO global statistics on employment by industry but it's limited even for large economies.  We also need to measure the availability of ICT skills and of non‑cognitive skills. 

The enrollment in education, lifelong training, and in addition this kind of data will help countries the supply and demand for digital skills by helping forecast their digital skills need.  We need ways to measure new jobs and occupations, the skill set require and how the workforce is made up.  In order to ensure that everybody can benefit from the digital economy is essential to produce policy relevant data and statistics on the multi aspects of the digital divide.  We cannot measure how ICT used by the public sector or health sector or education has indirect impacts on society or social well‑being. 

So the lack of data means the implications of the evolving digital economies for these countries generally are not well researched.  This is where we find ground level research from you guys really, really enriching and illuminating.

So finally with regards to the work done at the policy level, our level, in term of measurement of the digital economy I would just like to let you know that after we published this, we had an intergovernmental expert group that decided to form a working group to measure the E‑Commerce and digital economy

We hope to be able to announce this in the upcoming months but we already take the opportunity to invite all of you to express an interest if you would like to contribute to the work of this working group. 

If I may, I would like to invite all of you to the E‑Commerce week next year, because it's a gathering that will allow you to look specifically at digital economy issues and E customers and will bring stakeholders from both ICT community and the policy side so thank you. 

>> MATTHEW SMITH:  Great.  Thank you, Scarlett.

>> (Speaker of microphone)

>> MATTHEW SMITH:  Sure, please.

>> PANELIST:  Actually have a factual point.  We have only focused on our household and individual survey data but alongside this data are informal sector surveys.  Although we are contributing now by the micro work on the household and individual, in fact there is an informal survey which gives us insight into vast parts of the economy in most of the areas we work in so small contribution in the formal digital economy but this very informal survey that provides us some of the informal digital economy activity. 

>> MATTHEW SMITH:  Okay.  In some of the few minutes we have left I would like to turn to Adeel (phonetic) to give his thoughts about the role of this data for policy for evidence‑based policy formation.  I'm curious from your perspective from the African community commission.

>> ADEEL:  Let's get practical.  We are in the process of implementing a program which is policy and regulation initiative for Africa.  This is going to be digital platform that has ‑‑ it will be one‑stop shop for all national policies.  It will also be all the at the national level, you have it there, and also tools for all the national policies as well as the implementation of policies and at the national level. 

So what is important for me to attend this session is to see how we can leverage this because this is the common thing for this tool to work we need to have data and statistics on this platform so that's why we are interested in attending this session just to make sure if you can leverage your expertise so that we have this platform not only in the research‑based policy making, else on the implementation of policy and see how are we going the right direction or going the wrong direction and so forth so this is something that is of value for us.  Thank you. 

>> MATTHEW SMITH:  Great, thank you.  And let's open it up.

>> AUDIENCE:  Thank you for the floor.  I want to bring up some questions and see how they link up to the overall discussion.  I'm taken by this narrative that is predominant in the global South about mobile first and to some extent about mobile only. 

I wonder at what point we start reflecting on the limitations of that approach and even to the perceptions of relevance of the Internet because if those are the approaches that we are seeing being especially reflected in policy much less in research because research will follow what is there is how much meaningful work can anybody do on their mobile phone, when is the last time you want to report on your mobile phone so even when you reflect on E‑Commerce what we are looking at here is the consumers and not the creators and undermining digital Africa, Asia, Latin America, whatever you want to call it, at which point do you start reflecting on the limitations of mobile first and how we start challenging it because there's short changes down the line.

And the point on the policy and regulation for digital Africa, I think that's a great initiative.  Speaking as an and can here and citizen, it will be fundamental to speak to the people who are users of these platforms from innovations even representatives of young groups that are often not represented to be involved in those discussions and not lock them out.  Thank you.

>> MATTHEW SMITH:  We have one more statement and then we have to wrap up as we are running out of time.

>> AUDIENCE:  Thank you.  Just to welcome this initiative I would just encourage you to participate in the next intergovernmental group of experts on E‑Commerce and digital economy and make exactly this point so all policy makers here see this is a priority for government of Africa.  If we can support you in that context, we'll be very happy to.

I think in many cases when we talk about what kind of efforts we can do to help digital development in developing countries the statistical aspects are forgotten.  It's not sexy.  But once you have the data everyone is really happy about it but it takes an effort to get those done.  Maybe not everyone but at least ‑‑ (overlapping voices) ‑‑ the final thing I want to say, after Scarlett's intervention about the indicators, indicators are fantastic but we will look at those areas where we have adult availability and we need to have more than just connectivity data even though it's fundamental otherwise we will continue to focus on just improving the ICT supply side and forgetting about demand side because it's harder to get the data on the demand side. 

>> MATTHEW SMITH:  Okay.  The discussion wants to continue so I'm going to let that happen.  The patient gentleman in the front, please.

>> AUDIENCE:  Thank you.  I come from India.  So number one I fully agree that mobile first might be a good strategy but mobile first and foremost might be a challenge.  So that's one area.  And people should be seen not only as consumers but as pre-sumers something mentioned back in early 70s in the future people will be producing.  Second thing, in terms you mentioned in terms of UNCTAD impact a small and medium enterprises.  In these cases the government policy even if it's for a different reason can actually incentivize in a totally different manner. 

The only way to file tax returns is online.  So people have to come online in some manner.  It's a first thing but it's there.  Third thing you mentioned about women having this challenge getting access.  That's right.  But at the same time they are also concerns where women are finding additional opportunities because of ICTs.  They can do those things for which otherwise they had social challenges in terms of stepping out and trying to sell those things outside or collaborating.  The last thing is about measurement. 

On the measurement I think we need to differentiate between four different concepts very clearly.  One, is there's a number of subscriptions which is just the number of connections and legal contracts. 

If I have three mobile phones and one fix line connection I have four subscriptions but I'm still a single subscriber, there's a difference between subscriber and subscription.  Third, users, if you have a single connection in a school there might be hundreds of children using that.  In the mobile industry people often use this average of a new subscription. 

We need to look at the people who are even not users but still Internet touches them.  For example, in India when my house help goes to her village and wants to travel by train instead of taking a day off she asks us to make the booking we make it online and she travels.  She knows there's a way of doing those things.  She doesn't know how to use it herself.  But they nevertheless get the benefit of it.

>> MATTHEW SMITH:  I know we have a few people.  Can I encourage you to be ‑‑

>> AUDIENCE:  I'll be short.  It's an opportunity for such like this Argentina has assumed the G‑20.  Inside the G‑20 is something called the digital economic tasks for.  In 2018 there's going to be two tracks to focus for the digital economy.  So data research and how developing countries and strategy about digital inclusion, it's an opportunity for you.  Thank you. 

>> MATTHEW SMITH:  Okay.  Thank you.  This has been a small but passionate group.  I think we are all excited that we have the data now.  And it will be coming out over the next few months and be made openly available as well.  I think I'll let Allison, would you like to say words to wrap up? 

>> ALISON GILLWALD:  Just a few words.  I'll try to be quick because I know we are over.  What we have shown you here is our very first high-level finding, it's descriptive data.  I want to reiterate the point about what descriptive data with mask and that's what we go into now in this next phase when we start modeling the states and get a meaningful understanding, particularly in relation to gender and a lot of other things the descriptive data masks what we are seeing. 

So I think I wanted to make that point quickly.  And also just to say all of us across the global South network have started doing this work.  Our mandate is to influence policy and regulation.  We would love for governments to go out and look at the statistics and we could simply get the data and analyze it. 

But we know now after doing this for a decade on the Canadian tax payers back that unless we actually mobilize and organize amongst ourselves this work is not going to go on so this is really an appeal for further collaboration.  This is not the end of our work.  Scarlett is very familiar with the work we have done around policy and regulatory influence.  They're really significant impacts we have particularly in the area of pricing, we have lots of supply data. 

And perhaps broadly from a very high level at that stage and I know there's different nuances in the different regions to say the purpose of this research is to identify the area of research needed for policy makers going forward. 

I think what is a very sort of profound point that we have reached with the very low Internet penetration rates but Internet penetration rates are nevertheless only possible because of mobile broad band.  We reached our broad band in South Africa and North Africa.  It's absolutely critical. 

And I think what is clear from our research and to respond to the point that was made because I think it's an important one is mobile on its own cannot be everything.  And intact we know from our research that it's not.  Large numbers of people are buying their tiny micro package of data, going into public Wi‑Fi where it's available and meaningfully using them, and getting on that way.  So we have to begin to look at the use of the Internet outside of this individualized thing and we did discuss some of the measurement issues there before you came in to address that.  I think it raises serious policy questions for us about how we are going to address digital inequality in the future and identify research areas as we go forward.

>> MATTHEW SMITH:  Thanks.  I think we do have to wrap up now.  As a Canadian taxpayer I'm pleased to have be able to support this important influential work and thank you everyone for attending this session and for the patience of going over time.


(Session concluded at 12:18 p.m. CET)