Last year, AidData and the Center for Global Development launched a massive online database of China's development financing in Africa. At the time, many concerns were raised about its methodology and the headline figure reported extensively in the media.
As I explained at the time, 'understanding and calculating Chinese foreign aid is a challenging task. It requires a deep understanding of how the Chinese aid system works, how to classify different forms of financing, as well as the politics surrounding the reporting of data.'
The team at AidData has just released an important update to their database, based on a revised methodology. They've dubbed this 'TUFF' (Tracking Underreported Financial Flows) and it is a useful document for anyone working on the development financing activities of donors who don't report to the OECD DAC (or any other database). The database itself is a treasure-trove of project-level information on Chinese activities in Africa.
Things I like:
- Excellent search function (country, sector, loan type, organisation).
- Powerful visualisation and infographics make it easy to navigate.
- Transparent (links to resource documents, can export raw data).
- Provides snapshots of individual projects.
- Constantly being updated (including through user suggestions).
Things that remain troubling:
- Still heavily reliant on media sources. In my random selection of Eximbank-funded projects, for example, the majority of resources were media. Very few link to budget documents or Chinese reports. These are often challenging to access but important in the verification process.
- Lots of projects listed in the database have status 'unknown'. It is likely that many of these will never eventuate, as 'pledges' and even 'letters of intent' and 'loan agreements' are not set in stone. This gives the impression that there are many more Chinese projects in Africa than there actually are.
So, in all, the updated database is a useful tool for researchers, analysts, and commentators interested in 'China in Africa'. Kudos to AidData for the improvements and transparency. Let's hope people use it with the nuance intended.