Governments and their international partners are always looking for ways to make aid more efficient and effective. There are many dimensions to efficiency and effectiveness but an important one is to understand how important aid is to supporting government sector policies. The problem is that despite there being ample data, governments and donors do not routinely track and analyse how much aid is allocated to different sectors, and how this relates to government spending in those sectors. This happens partly because policy makers and donors use different systems for classifying their spending, which may also be denominated in several currencies, a further complication.
The central problem is that there are two different classification systems. Donors report aid to the OECD according to Development Assistance Committee – Creditor Reporting System (DAC-CRS) classification standards, while governments report their fiscal activity to the IMF according to Government Finance Statistics (GFS) standards. Hence, to be able to analyse aid and government expenditures by sector both datasets need to be comparable so that policy decisions are better informed.
A paper from Artificial Fiscal Intelligence (AFI) presents a methodology that should help resolve this classification and comparability problem. Using a global database (available here), AFI provides the necessary data to allow free and easy analysis of how important aid is to government sectoral funding. The system takes three global data sets and establishes a set of bridging tables that allows analysts to determine the size of donor assistance in a sector as a share of government spending. These data can help donors and governments allocate aid and government resources much more efficiently.
One of the aims of the methodology is to use readily available public data to enable analysts to quickly understand how important donor assistance is to a government’s sector financing position.[2] The primary sources used are the OECD DAC-CRS database for aid data and the IMF’s Government Finance Statistics (GFS) database for fiscal data. To overcome the classification differences, AFI developed bridging tables that map data from OECD-DAC-CRS to COFOG. A single currency (US$) was also established. A set of Sector Absorptive Capacity Analytics and Donor Dependency Indicators can be constructed from the dataset.
The application of this type of sectoral/functional data can help donors and governments allocate resources better. Analysing aid in relation to trends in government expenditure helps donors assess whether they are potentially over- or under-providing aid to a particular sector. And it allows governments to develop a detailed picture of how total resources are being allocated. Both donors and recipient governments would then be in a better position to discuss aid allocations. The data would also assist in defragmenting resource allocation systems.
These measures can help donors and governments establish critical information systems and analytical functionalities. There is a strong case for donors to do the heavy lifting from a financing and technical cooperation perspective when a country is establishing new systems - for example, setting up new social protection programs. The paper provides an example of this for Papua New Guinea and Sub-Saharan Africa.
Another important message is that many countries do not yet provide COFOG data to the IMF. Other countries may supply data but not at a level of granularity required under GFS. If such detailed data were provided, a more robust policy and resource allocation dialogue could occur. Moving forward, donors and government could prioritize the timely provision of COFOG data to allow indicators to be made available to all stakeholders. Technical assistance might be provided by the IFIs or donors to strengthen the construction of GFS compliant data.
To conclude, further research is warranted on the development of Sectoral Absorptive Capacity and Donor Dependency Indicators and their application. AFI will continue to update the datasets on a routine basis. Suggestions for further refinements of the dataset can be sent to info@artificialfiscalintelligence.com.
[1] Andrew Laing is a Director of Artificial Fiscal Intelligence (AFI). AFI is a startup formed to help governments develop and apply fiscal intelligence in order to get better fiscal outcomes.
[2] There is nothing to stop any government or donor applying the same methodology to non-public data drawn from their own systems.
Note: The posts on the IMF PFM Blog should not be reported as representing the views of the IMF. The views expressed are those of the authors and do not necessarily represent those of the IMF or IMF policy.
