Share

This is a joint post with Yuna Sakuma

Funding for health in any country doesn’t always go where it’s needed most.  But this is particularly true where donors and national governments have to decide how to distribute scarce resources within a country and coordinate their efforts.  Often, governments don’t know where donors are spending their health funds, and vice versa, which can result in a misalignment between funding and need.  

The AidData Research Consortium (ARC), a research collaborative based at William & Mary, aims to tackle this issue using geocoded development finance data.  The Consortium will compile sub-national spending data from both public and aid sources, make data available to the public and researchers, and facilitate a consortium of researchers –including CGD researchers- working on policy-relevant issues.   

The potential benefits of this work for global health are many.  As the Consortium rightly states: “Spatially referenced aid information makes it possible to know ‘who is doing what and where,’ which is essential to ensure that resource allocation, targeting coordination, and channel-of-delivery decisions are based on evidence. Geocoded aid data can help us better understand the factors that contribute to poor health service delivery, leverage complementary programs, and target resources to address malaria and HIV/AIDS.”[1]

To illustrate the potential of such analyses, AidData collaborated with Malawi’s Ministry of Finance to geocode aid locations from thirty donor agencies. Figure 1 shows the location for USAID health sector projects in orange, and the proportion of the population in each district with inadequate access to health services in shades of grey. Although the metric for access to health services is not specific to a type of service, this example illustrates that the location of USAID investments does not necessarily correspond to where there is the most need for services.

Of course, there are challenges with this approach, some of which are immediately evident in the Malawi example.  For instance, without comprehensive donor data –particularly from the big health funders like PEPFAR – a meaningful analysis of how all health funding is allocated within a country is not possible.  Similarly, without data on government spending on health or on the costs associated with providing health services, it may be incorrect to conclude that USAID is misallocating its resources in Malawi. 

A more fundamental question as this effort gets underway is, how can AidData use its subnational, geo-referenced niche and commitment to open data to add value to already-institutionalized efforts by the WHO, IHME, OECD, IATI and others? And how can all these groups band together to demand greater transparency in spending from the least transparent, largest donors?

AidData’s work is a step in the right direction for development and aid effectiveness, particularly in the health sector where coordination efforts can be stymied by lack of subnational data on spending and impact. We at CGD hope to build on this data to explore our interest in how other factors – beyond lack of information –may cause inequalities in the allocation of health resources within countries.  Some research suggests that inequalities are a result of perverse incentives built into budget systems (as here), electoral politics (here) and ethnic affiliations (here) which can drive the allocation of public funds away from where it’s needed most. This year, we are launching an effort to analyze and identify better practices in fiscal transfers for health – from the national to the sub-national level – and will tackle these and related issues (stay tuned!)

Note: The AidData Center for Development Policy is a partnership of the College of William & Mary, Development Gateway, Brigham Young University, University of Texas-Austin, and Esri.

Source: [1] AidData “Analyzing Spatial Relationships: Global Health Aid” (2013)

Figure 1. Inadequate access to health services in Malawi and USAID health sector projects