A picture speaks a thousand words, or in this case, raises nearly a thousand questions!

A couple of weeks ago at Publish What You Fund, we were having one of our internal Aid Transparency Index technical meetings. The nature of the methodology (which we’ll be using for the 2020 Index) is such that we regularly have to double check that our indicators and approach are still relevant and implementable bearing in mind changes to the IATI Standard and the practices of publishers.

Specifically, we were meeting to discuss the different ways in which publishers ensure transparency of their activities. The decentralised nature of IATI means it is possible for each actor to publicly publish information about the part of the development chain which they are responsible for, and (if they all do it properly and if the proper tools exist) then it’s possible for an ordinary data user to see the whole flow. This is good for a number of reasons: information is in theory published in a more timely manner, lines of responsibility are clearer, reporting can be more easily standardised, and resources can be more accurately traced.

Today, some donors publish activity data on behalf of their implementers, while some donors publish their outgoing disbursements then mandate their implementers to publish the granular detail about the activity themselves. As you can imagine, having these two approaches to publishing activity data might pose a challenge to anyone seeking to determine the transparency of each donor per se. In due course, once we’ve tested our planned approach, we’ll let you know how we’re intending to incorporate both of these approaches within the Aid Transparency Index methodology. But this blog is actually about what happened after the meeting.

Following our discussion about the different practices of donors, and the upstream and downstream links that should ideally connect activities, our then Developer Andy Lulham was inspired to visualise the links between all current IATI data. The end result is here, and for those of you who prefer data to diagrams there is a table below showing the top ten providers of links.

The first thing that struck us was that neither the diagram nor the table look like we’d expected them to, nor, we suspect, like most of the IATI community would want them to. To put these numbers into context the IATI Registry contains more than 1 million activities, yet the current data suggests there are less than 5,000 links. At the same time recent analysis from Devex (here and here) illustrates that in 2018 DFID made payment to a total of 1,259 distinct suppliers – however the visualisation shows that only a handful of these suppliers are undertaking activities which are linked to DFID.

We posted the visualisation on Twitter, predicting some challenging questions, but none came. So here are some from us:

• Links from, or to DFID, Dutch MFA (minbuza_nl) and Belgian MFA (be_dgd) make up 99% of all links available. With almost 1,000 IATI publishers, where are the rest, including big donors such as USAID, KFW, and the World Bank and the big implementers such as JSI, AbtAssociates, Oxford Policy Management Ltd, Oxfam, Save the Children, IMC etc.?
• What is it that encourages some implementers (see DFID suppliers manniondaniels or tripleline_crownagents) to report their disbursements and receipt of funding so comprehensively?
• What does all of this mean for traceability and for users trying to follow ODA and other spending down through layers of donors, and international and local implementers?

There has recently been some discussion on IATI Discuss about ways in which the process for adding reciprocal links could be automated once one link is already in place but realistically this is only addressing the tip of the traceability iceberg and would result in just an extra few thousand links. We’re also glad to see that IATI is currently looking for a service provider to develop visualisation models to trace selected funding streams and develop recommendations for publishers on how to improve the traceability of funds. But frankly, considering the scale of the challenge identified here, we’re not sure that anything other than a step change in publisher behaviour resulting in a significant increase in the provision of links, will be enough if we expect users to be able to trace funds from source to frontline expenditure.