The above graphic from a story in the Daily Nation in Kenya shows how much financial assistance China has pumped into Africa in recent years. It’s based on data from the research initiative AidData collected through an open-source method of tracking flows of financing from China into the continent.
Last year a report from AidData on Chinese spending on aid and development in Africa came out that announced the massive amounts that China has contributed to projects on the continent. As discussed previously, that report was was widely reported on in the mainstream media. The $75 billion that China spent in Africa was framed as a sign of China’s attempts to cement partnerships and allies among African countries. The British newspaper The Guardian said of the report’s findings that it pointed to “Beijing’s escalating soft power ‘charm offensive’ to secure political and economic clout on the continent”. Especially interesting was the report’s conclusion that China had contributed to fewer mining projects than to projects in the areas of health, education and social infrastructure. Examples of these were listed as a malaria prevention center; school for visual arts and an opera house, as well as doctor and teacher exchanges. The Guardian took these examples as evidence of an increasingly strong geopolitical agenda on the part of China.
However, the report was also strongly criticized by experts on China’s role in Africa. Deborah Brautigam, professor and director of the International Development Program at Johns Hopkins University, said the figures in the AidData report were misleading because the methodology relied largely on media reports of Chinese project financing, which are not reliable.
In a recent new development, AidData has released new updates from its Tracking Chinese Development Finance in Africa dataset. Acknowledging that last year’s pilot project, which relied largely on media reports, was ‘inherently imperfect’, the new methodology cross-checks media reports with data from official sources, NGO reports, and scholarly articles. AidData believes that there are certain conditions that make the refined methodology an effective research tool. The open source nature of the data makes it possible for users to identify errors. The overreliance on media reports for which the previous report was criticized, is now complemented by other methods such as in-country fieldwork and outreach to personnel involved with specific projects. Information from media reports is cross-checked and supplemented by data from official sources, NGO reports, and scholarly articles.
This refinement in methodology means that more sources are now being used to establish the extent of Chinese financing in Africa. To better reflect the shift to a variety of ources used, AidData has now changed the name of its methodology from Media Based Data Collection to Tracking Under-Reported Financial Flows (TUFF).
The introduction to the new codebook clarifies that the original name for the methodology, Media Based Data Collection, may have been misleading to critics who didn’t review the whole codebook. Although the initial report relied heavily on media reports, according to Charles Perla, an AidData Project Manager, these reports were not their only source: “In fact, media reports are used only as a departure point, and are supplemented with case studies undertaken by scholars and non-governmental organizations, project inventories supplied through Chinese embassy websites, and grants and loan data published by recipient governments.” To make this clear, AidData has now renamed their methodology. “In the interest of providing greater clarity, we now refer to our methodology for systematically gathering open source development finance information as the Tracking Underreported Financial Flows (TUFF) methodology”. The codebook and an explanation of the methodology can be downloaded from the internet.
Working with researchers at the College of William & Mary and Brigham Young University in the USA, AidData set themselves the objective of documenting all known Chinese development finance projects in Africa over more than a decade, from 2000 to 2012. The new release includes more than 100 new projects, and more than 130 updates to existing project records. AidData seems confident that the refined methodology will withstand scrutiny. The pilot project was subjected to a test referred to as a ‘ground-truthing approach’ where researchers visited project sites in South Africa and Uganda to corroborate the data compiled by AidData. This testing involved interviews with local recipients of financing and other stakeholders. The pilot, conducted by researchers at AidData, the University of Cape Town, Zhejiang University, the College of William & Mary, and Brigham Young University in collaboration with local enumerators, did find some new information that was used to amend and correct the TUFF data. By and large, however, these researchers say that the interviews and site visits they conducted supported the open source data gathered by AidData.
The writers acknowledge that there is as yet ‘no consensus’ on how the flows of development finance from China to Africa can be tracked. They claim that this inability to follow the money in a new geopolitical environment marked by rapid shifts, flows and contraflows is a weakness of traditional academic research that cannot keep up with the ‘rapidly changing global development finance architecture’. It is however important to find a way of keeping track of these flows, so that African communities can better understand, interact and engage with the influx of funding from China so as to make best use of the opportunities it offers.
Although these researchers acknowledge that the TUFF methodology isn’t flawless, they see it as a ‘fairly robust method of independent data collection’. Where information gathered on the ground in South Africa and Uganda differed from the information gathered through AidData’s methodology, these differences were fairly minor. The places where two sets of data diverged, involved issues like dates, contact information and similar details, while the larger, more significant items of scale, scope and sector of financial assistance remained largely the same.
The revisions of AidData’s approach remind us that scientific methodologies are never perfect, nor are they complete. And while the media is playing an important role in bringing to attention the flows and contraflows between China and Africa – especially by stimulating debate and interpreting these developments – media reports alone cannot serve as reliable empirical evidence. For this reason, the new multi-source approach by AidData seems like a step in the right direction. By revising their methodology, AidData has also displayed a willingness to open their work up to questioning and scrutiny. It may be expected that in a new, rapidly changing and in many ways controversial area such as Chinese involvement in Africa, research methodologies will be looked at especially closely. AidData should be commended for undertaking this scrutiny themselves, and for displaying a willingness to engage, revise, and rework their findings. Ultimately such a collaborative effort such as their open source project – as well as the contributions of subsequent critics of this work – can contribute to a better understanding of what has become one of the most important shifts in the global political economy of development and aid in Africa.