A (3:09)

So if Nigeria was off by that much, it seems plausible that maybe lots of other African countries are know how big their GDP is. They don't know what the birth rate is like or the death rate or diseases or lots of other things that planners might want to know.

B (3:25)

Exactly. And I mean Ghana also had a rebasing that made a 60% difference to its GDP estimates. And now lots of countries in Africa are looking again at this and saying we need to do the same thing.

A (3:37)

So a lot of times at CGD we focus on what well intentioned outsiders could do differently. The donor governments, the big philanthropies, international institutions. In this case this is a kind of a knotty problem. There's a lot of it is a domestic component. It's like if the data is not getting collected, well, whose fault is that? It's the government's fault, right?

B (4:01)

Well, I think there's kind of a collective guilt going on here. Certainly we continue to focus on those international organizations and their role in the problem. The donors had something to do with this. First we found that most statistical activities were funded by the donors themselves. So if you have your non salary costs covered 90% by an external funder and you see that they're really only doing household surveys and they're failing to produce accurate data on things like growth, things like poverty, things like labor markets, well that tells us something about the kinds of incentives that the donor is creating in that statistical agency. On the other hand, we also know that governments themselves in Africa are not putting very much money in the. They're certainly not compensating for the volatility of donor funding in this area, for example, and it just can't be described as a priority. So it's definitely a shared agenda. And that's why we partnered with the African Population Health and Research center to do this working group with the idea that they would be based, they're based in Kenya, they would have an audience of policymakers in Africa and then we would have our traditional audience of the international organizations, the World bank, the un.

A (5:14)

So the report is a joint product. It includes the perspectives both of the African governments themselves and recommendations for them and recommendations for the donors. I want to get into those recommendations in the second half, but first I want to understand better since we at CGD tend to focus on the role of the outsiders. What are these perverse incentives? Is it that the outsiders just want household surveys and they don't want to know births and deaths? Wouldn't they like to know sort of the basic vital statistics as well?

B (5:47)

I think they certainly would like to see more vital statistics data, but it just hasn't been a priority because household surveys are so. They're not easy to do, but they are relatively easy to do. You don't depend on the bureaucracy of the Ministry of Health or Department of Civil Registry to generate that statistic.

A (6:05)

Sort of freestanding one off thing.

B (6:07)

Exactly.

A (6:07)

You can come into a place with no capacity, bring in your enumerators, run the the survey, get your answers.

B (6:12)

Exactly. You can even just go through the private sector. So that's why the household survey is so appealing. And household surveys are incredibly important because they allow us to say something about the accuracy of the administrative data that is being produced. So it's very great, it's a great thing that the household surveys have happened. But that said, household surveys are infrequent. They're generally at a national level of aggregation and they're not useful if you're a policymaker that has to make decisions on how much should I allocate to districts for education or how should I change my composition of spending in health so that I address the major burden of disease. So that's really why we're emphasizing this administrative data part and the accuracy of that data. Another issue that's come up with the donors is our love of pay for performance. And as you know, at CGD we love results based aid and cash on delivery. But the issue is without a very good, independent, accurate data system, if you incentivize the production of certain kinds of data, well, you get certain kinds of.

A (7:13)

Data, enrollments are up. Good news.

B (7:16)

Exactly. I mean within governments there's budget incentives for misreporting, but there's also between the donor and the recipient, let's call them misaligned incentives. So we like results based aid. It should generate virtuous incentives, but only if we have an accurate source of data that is not going to be affected by the financial incentives that are flowing.

A (7:42)

Which country has the best statistical system in Africa?

B (7:46)

Well, that's a tough question. Certainly Rwanda and South Africa have been mentioned repeatedly during our working group as being particularly strong. For example, stats. South Africa has a very interesting way of working. They've basically placed staff from stat South Africa in the line ministries to help them build better, more accurate Data systems, administrative data systems. But you know, South Africa, it's almost an upper middle income country I think now, so it's probably a little bit different, but for sure they've done well. I would also say that the Francophone African countries in general seem to be a bit better than the Anglophone African countries because of a tradition, I guess, of an independent national statistical organization. So. So they have that tradition and maybe that has something to do with their ability to produce better data and to invest more in data as well.

A (8:41)

Have you come across any examples of where a president championed this and said, I mean, I would think that the main person demanding this would be the president because she or he would want to know what's going on and that if the chief executive is not demanding better data, then it's hard to imagine how an initiative by the National Statistical Office in and of itself is going to make a difference.

B (9:05)

I agree. And we're not only talking about national statistical organizations, we're talking about the national statistical system. We haven't seen presidents take leadership in this area, but I think with these very high profile GDP rebasing efforts in Nigeria and in Ghana, people have gotten the picture of how important data can be in terms of your access to financing, in terms of investment. These data matter for development. So I hope to see in this next stage presidents stepping up to the plate and advocating for better data in their own countries.

A (9:38)

Well, and I guess with this report in hand, a president that wanted to do that would have a blueprint. He or she could say, this is really important. I want to make this happen. And then the people who were charged with following through on that would know what that was. They could look into the report and find the answers. We're going to take a quick break and when we come back, I want to hear about the recommendations for, for this hypothetical president champion that we're gonna wish into existence. This is the Global Prosperity Wonkast from the center for Global Development. I'm speaking with Amanda Glassman about a forthcoming report, delivering on the data Revolution in Sub Saharan Africa. We'll be back in a bit. Welcome back to the Global Prosperity Wonkast. My guest is Amanda Glassman. We're talking about delivering on the data revolution in Sub Saharan Africa. Amanda, there's a lot of excitement about the data revolution with the post2015 planning the whole idea of big data. People in the rich countries have heard a lot about unwanted surveillance. There's a sort of a feeling that we're awash in data. Isn't Africa's problem with data just going to be solved by this big data revolution tsunami that's going to come and lift all boats.

B (10:56)

Well, certainly both big data and new technologies are really exciting and offer some really interesting opportunities to collect new data. You can imagine, for example, in the area of environment, the use, for example, our colleague David Wheeler. Our former colleague David Wheeler started to use the satellite imaging to count the number of trees in a given country. So there's a lot of places where new data, big data, non traditional sources can play a role in revolutionizing the information that we have for policy. On the other hand, if we don't have a national statistical system in place that allows us to just produce the basics, well, we'll be missing the opportunity to harness these new technologies and opportunities to use big data.

A (11:42)

These basics. Births and deaths, growth and poverty, tax and trade, sickness, schooling and safety, land and environment. How are you going to get those building blocks in place?

B (11:54)

So the working group came up with one main recommendation which is the idea of data compacts that could be led by a president or minister of finance together with interested funders from the outside world. So the idea would be to say we're going to prioritize some aspect of the building blocks that we have not achieved in our country and you would agree on progress in accuracy, timeliness and openness of that data. And you know, you could basically have a phased set of actions. It would allow for a big high level political commitment. It would require greater resource mobilization from both donors and governments and then some independent tracking of how progress goes on the building blocks. That's our big idea.

A (12:42)

So the compact would let both sides pre commit to doing something to solve the problem, put in place the incentives for it to be addressed. What sort of changes in donor behavior would be required then? What would you want the donors to be doing ideally in a compact country?

B (13:03)

Well, first what we'd like them to do is increase the amount of investment in data we'd like them. They can continue to do all their household surveys and their project specific surveys.

A (13:13)

And evaluation their RCTs.

B (13:15)

Right. We're not going to, you know, that's not going to change. But what we would like to say is we'll do all those other things, but we will at least assure that the building blocks are adequately invested over a period of time, contingent on progress in, to repeat again, accuracy, timeliness and availability.

A (13:34)

So this could be a kind of a pay for performance contract.

B (13:37)

Yes, yes, that's our favorite recommendation and it's coming back in this iteration because we think it's really the prerequisite to all other kinds of pay for performance. Performance too.

A (13:47)

How would you expect a compact like this to come into being? If a president, hypothetical president, decided to do this, what would chier he do? They'd start talking to the donors, they'd start talking to the ministers. And how would it happen?

B (14:06)

Well, so I think the first step would be for the country government to come up with their, you know, what it is that they'd like to achieve over the next couple years. It could be embodied in a national statistics development strategy, but it should be something that could realistically be achieved in let's say a two to three year time frame. And then I would shop it around to a coalition of the willing, let's say of donors. I don't know that there are some people out there that are advocating for a global statistical agency. Maybe that happens eventually. But for now we have the usual suspects, right? We have the World bank, we have the Gates foundation, we have usaid. These are all donors, DFID that have been historically investing in statistics, mainly on the household survey side. But you can imagine those donors getting together with country governments around a realistic plan that would focus on the building blocks and agreeing to this compact together.

A (15:01)

So I guess part of what your report does would increase the receptivity to such a proposal among those donors.

B (15:07)

That's right.

A (15:08)

They'll see the report, they'll be aware of it. And if there was an initiative from a leader in Africa to say, yeah, I want to do this, they would say we get it right.

B (15:16)

We hope so.

A (15:16)

We understand.

B (15:17)

Yes.

A (15:20)

Got any short list of countries where this might happen? Your collaborators on this are based in Kenya. Does that be a good starting point?

B (15:28)

Well, certainly that's one of the countries that we would investigate further, but I could certainly imagine a Nigeria that's already taken the brave step of rebasing their GDP. After 20 years of sticking to the old method, they've already taken the plunge. Maybe they would like to undertake a country compact for better data. You can imagine Liberia being a very good one. These are all countries that rely a lot on donors for the funding so that maybe restructuring the relationship with donors could really make a difference.

A (16:01)

Final takeaway for our listeners. What have I forgot to ask you about?

B (16:04)

Well, I think one thing that's really important is to think about this recommendation in the context of the post2015 process. There have been a number of meetings at the un. Maybe a number is even a.

A (16:18)

A large number. Sure.

B (16:19)

Large number, yes, a large number of large meetings, a lot of meetings, of lists of indicators of things we would like to measure. But how can we have these discussions in the absence of understanding really what kind of data is and what quality that data is? I think if we want these goals to serve as a motivation for investment and for accountability and development results, then we have to make sure that the data underlying them is very good, very reliable and accessible to all those people, whether they're in a country working for a civil society organization, whether they're a local government or whether they're from a donor based ngo. These are the people who use the data to hold governments accountable. So if we don't focus on this as part of the post2015 process, if this isn't what we're talking about when, when we say data revolution, I think we're missing the boat.

A (17:07)

Would you expect to see some kind of commitment or call in the post2015 development goals framework around this question of data quality and data availability?

B (17:18)

Yes, I would.

A (17:20)

Is that going to happen?

B (17:20)

Yay. Lawrence. Well, we'll do our very best to let it be known that this would be a good idea. Certainly we're engaged in the process and talking to all the different people who are involved.

A (17:31)

Thanks very much.

B (17:32)

Thanks a lot.

A (17:32)

This has been the Global Prosperity Wonkcast from the center for Global Development. I've been speaking with Amanda Glassman about a forthcoming report delivering on the data revolution in Sub Saharan Africa. It's a joint product of the center for Global Development and the African Population and Health Research center, aphrc, based in Nairobi, Kenya. You can read more about the report, the report itself, and a brief and everything that you might want to know about improving statistical data capacity in Africa on our website. Thanks very much. You can find the Wonkast online on itunes and on Stitcher. Just search for Wonkast or CGD and sign up to hear a new interview every week. Until next time, I'm Lawrence McDonald. Thanks for listening. It.