Let’s face it folks, Africa is flying data blind, but a big revolution could be coming

We simply stumble from one day to the next. And we guess. A lot.

A CELEBRATED columnist launches a stinging criticism of the government, his argument anchored by official health and education data. City authorities earmark a few billions for “Big Brother” type of security surveillance equipment. 

Elsewhere, ministry technocrats disburse even more to schools, based on enrollment information gleaned from “the ground”. And a big fancy international consultancy group conjures a major report carrying futuristic projections of rosy African growth. 

What are the odds the data relied on by all these groups of people are broken, and in many instances just guesswork? Overwhelming.

Last month, Kenyan police unearthed some 75 Chinese nationals holed up in adjacent upmarket homes in the main city of Nairobi, triggering a diplomatic headache over which of the two countries would try them.

Responders had gone to investigate a fire at a home in the city’s pricey Runda estate in which one person died, when they stumbled upon an intimidating array of telecommunication equipment. Nairobi insists the Chinese ring, thought to be in the country illegally, was targeting its communication and financial systems, while Beijing claims its citizens back home were the prey and wants them extradited. 

As the tug of war over deportation continues, the furore has quickly been painted up as Kenya’s first major international cybercrime case, highlighting the new digital—and data— threat posed to African cities.

Risk of flying blind

Last week, the Economist Intelligence Unit released its well-watched Safe Cities Index, in which Johannesburg was the only African city that met the criterion of being in the focus group of 50 cities. This was not because it was the safest, but due to the relatively poor quality of data available in the region to weigh up other regional cities, the EIU said.

This deficit is felt even in the most basic of data, stoking concern that Africans simply have no way of seizing up the myriad risks posed to them everyday, from whether they will be the victim of mugging or terrorism to whether the building they are passing under will collapse, or the nearby clinic will be able to handle the newest disease threat. Essentially, most Africans are lurching from one day to the next, surviving on a mixture of hope, religion and blissful ignorance.

This also means that it is difficult to challenge the persistent global perception that the region is crime-ridden and graft-infested, while any progress simply gets lost through the cracks, such as say, the declining crime in Lagos, gleaned from crowd-sourced statistics, or a light urban rail system in Addis Ababa. 

In the next 15 years, Africa is expected to cease to be predominantly rural, at current explosive rates of urbanisation. But far from providing an opportunity to pare down poverty, that holy grail of governments, the region is instead looking at battling the resulting asphyxiating pressure on essential services, if they can be found at all, exacerbated by its other millstones such as pollution and crime. 

It also means scarce resources are spent on the wrong priorities, while taxpayers have no currency to use in demanding accountability.

Without data to support policies aimed at tackling this, many authorities would be forgiven for looking at the next decade with fear and uncertainty. 

A vicious cycle

Safety is in many instances linked to wealth—more developed countries have higher indices of wellbeing. And if Africa can’t guarantee security, it loses out on investment, entrepreneurs and tourists, further shrinking its ability to provide basic public goods, in addition to its ability to measure progress on key indicators, such as the Millennium Development Goals.

It is a vicious cycle, and one that has been around for a while, only that the post-independent continent did not pay much heed as it shifted through the development cycles—optimism, tempering, despondency and on to the current Africa Rising cheer.

A former World Bank director recalls being apprehensive when she was tasked to lead the flagship 1990 World Development Report on Poverty. “…all we had at our disposal was information on levels of poverty and their changes over time in barely a dozen countries,” Lyn Squire wrote.

“What was so desperately needed, and what I would have requested, was vastly more knowledge about the poor and how their wellbeing was, or was not, improving.”

The lack of solid African data is a frequent complaint. Every year since 2007, the Mo Ibrahim Foundation has released its Index of African Governance, named for its Sudanese-British billionaire founder. Its stated aim is to help nuance the public discourse over government performance, especially in a continent where style often trumps substance.

Despite having a host of rich indicators, the organisation’s familiar refrain has been an acute lack of national data, leading it to lean on international sources by default, and adopting a simple disaggregation of regional data.

Therein lies another challenge: the overwhelming data generated for the continent is by donors looking to allocate resources. The World Bank for example has since developed its flagship development dataset—PovcalNet, which draws from both its own and national sources. But it was only last year that data was made widely available.

But in addition to using data from its own surveys, it also relies on country level numbers, which have in recent months been shown to be significantly biased. 

Over-reporting and under-reporting 

Centre for Global Development Researchers for example in a recent research found that African governments over-reported—intentionally or not— school enrolment data; only coming to light when many authorities abolished user fees and started disbursing cash per pupil from the centre.

The same survey showed that when the Global Alliance for Vaccines and Immunisation (GAVI) in 2000 offered some countries in the region a fixed payment for every extra child immunised using official data, a 5% overestimate of coverage rates across 41 African countries was revealed.   

Even consumer price indices, calculated by seemingly better prepared offices are not safe: research shows that correcting for inaccuracies suggests the rates of growth and poverty reduction in Africa are slower than put forward in official data.

This inevitably plays out into the national stage, with several African countries having carried out a rebase of economies that had been pegged on extremely dated information. (Read: Why African countries should re-benchmark GDP statistics)

With only a dozen African countries having autonomous national data offices, the rest of the polities are dependent on government and donor funding. This leaves them wide open to machinations by political and interest groups—the story is told of one serving East African leader who routinely padded figures ahead of major meetings with lenders.

Donor funded projects are also significant cash-cows for staff and statistics offices, in addition to being tailored for sponsors’ needs, leaving them with little incentive to build national capacity.

Another formidable hurdle is that data collected is not accessible, with those collecting said to be either hesitant to share data or lacking the ability to do so under international best practices, to go along with a host of other problems. The situation is then one of even accurate data going unreported.

With this type of obstacles it is easy to be gloomy about the data picture in Africa. And with the MDGs expiring, how can anyone prove the progress countries claim to have made is not based on creative and pliable accounting?

Sources of new hope

There is new hope that the situation can be turned round. This seemingly inherent weakness of national data is the renewed focus of the high-level Data for African Development Working Group, backed by among others the UN, African Development Bank, African Union, Comesa and the World Bank.

Since the High-Level Panel on the Post-2015 Development Agenda’s report dropped the term “data revolution” there has been a global fashionable buzz around it as agencies seek to redefine how development data are collected and disseminated. The panel is drawing up a new set of rules to replace the MDGs, which expire this year.

Among proposals have been calls for a global data compact that would align both national and donor funding for better statistics, which are seen as crucial building blocks for global development.

Africa would stand to gain most from such a major change, with the Working Group calling for more and different funding to improve data collection, including weaning agencies off donor money.

Others are building institutional capacity to generate accurate and reliable data, part of which includes ensuring autonomy and cutting out crafty politicians, while building quality control mechanisms.

Leadership and data literacy programmes will also be vital, the UN says.

Finally, the adoption of open data practices—why have better data that you can’t use and which you won’t allow anyone else to access?

The tragedy is that the African Union back in 2009 pushed out a statistics charter, that called for pretty much the same thing the world is now focusing on, but the bloc has struggled to give it the necessary visibility.

Part of the reason is that the price tag is not cheap, but if Africa is to truthfully slash poverty rates, the growing consensus is that the focus must be on better, not necessarily more, data—as is the current trend is with numerous censuses and surveys. 

This is one instance where big is not always better.

Related Content


blog comments powered by Disqus