Corruption & DBS:

Norad's comments to the U4 Issue Paper: "Direct budget support and corruption", and CMI Brief: "Should Corrupt Countries receive Budget Support?"

Direct budget support and corruption
(Ivar Kolstad, CMI, July 2005)
Navigate the ISSUE Web Pages or

Download the Issue Paper (pdf 360 KB)

Should Corrupt Countries receive Budget Support?
CMI Brief: November 2005, Volume 4, No 4 (pdf 170 KB)

This comment is based on the U4 Issue Paper: "Direct budget support and corruption" (Ivar Kolstad U4 Issue 1:2005, June 2005). We have also reviewed another paper which is based on the U4 Special Issue Paper, but with a different title: "" (Ivar Kolstad, CMI Brief, November 2005, Volume 4, No 4, November 2005)

The paper gives useful contributions to the discussion regarding specification of what kind of analysis should be undertaken before a final decision on budget support is made. Questions such as whether the support changes the balance of power in an unfavourable manner, and whether the support promotes rent-seeking activities, are important and should be considered to be brought clearer into assessments of budget support. This is, by the way, equally relevant for other aid modalities and for country strategies in general.

As the paper focuses on corruption, it provides mainly arguments covering the "downside" issues in assessing budget support. Even though the paper states that budget support is suitable for partner countries with good institutions, it may easily give too negative associations regarding the effectiveness of the use of budget support in partner countries. The overall objective of budget support is mainly to support a partner country's effort to reduce poverty through providing funds for the country's poverty reduction strategy through the government budget. A decision of providing budget support is based on a comprehensive analysis of whether the expected development results outweigh the identified risks. Corruption is a part of this analysis through the assessment of governance as well as the public financial management system, which is mandatory in an appraisal of possible new Norwegian budget support arrangements.

One section of the paper deals with how corruption can weaken some of the rationale for providing budget support (including increasing the partner countries' ownership to their own development processes, and increased accountability). It is argued that in terms of corruption, ownership is problematic since:

Firstly, budget support increases the discretion of governmental officials allocating funds. The increased availability of finances could potentially increase corruption in countries were control and sanctions are weak, according to the paper. However, in our view one should be mindful of drawing too hasty conclusions. Budget support intends to provide the partner country's government with a greater amount of untied financial resources, which are to be used according to the government's own plans and budgets. These finances shall help the government free resources for e.g. poverty reduction purposes.

Secondly, budget support promotes ownership by the central government which according to the paper, can be more or less democratically elected. To the extent that budget support allows a government to monopolize decisions regarding allocation of public funds, this can lead to increased corruption.

These concerns are relevant, and good arguments for not providing budget support to countries with poor governance. However, in countries with legitimate governments striving to improve governance and make better use of available resources, budget support will most likely have the opposite effect, - and improve national ownership, improve transparency and reduce corruption. In countries where Norway is providing budget support, both the partner governments and the donors monitor the implementation of the budget support agreements, and focus especially on how to increase accountability and transparency in the budget processes through reforms and capacity building within public financial management. In addition, a broader public debate on use of government funds (domestic tax and development aid) is being stimulated. The institutional development issues captured through the budget support co-operation are much more far reaching than what would normally be the focus of a development project. Some improvements made in the general public financial management system of a partner country may be of greater importance in terms of hindering corruption than implementing a well-functioning project management system covering a few donor-driven projects only.

The paper argues that in countries where other (national) institutions are largely unable to hold a government accountable, budget support will itself have little effect on accountability. This is to some extent correct. Weak institutions and the need to improve accountability are main challenges in most developing countries. However, it is important not to have too static an approach when analysing the strength of institutions in our partner countries. If an assessment of a country concludes that there is a credible reform agenda and that the power balance within a country is favourable for implementing reforms and strengthening institutions, one should expect that the institutions may develop over time in favour of more accountability. If an assessment concludes that the power balance in a country is not in favour of reforms, one should not enter into budget support co-operation. In such countries, it is reasonable to question whether funds for development purposes should be channelled through the government at all, and whether the country should be a partner country for Norway's development cooperation.

The paper presents an interesting discussion regarding the sequencing of public financial management reform and budget support in order to avoid increased corruption. In the practical work there are many challenges related to this. The discussion regarding what should be "good enough" standards for public financial management is important and should been given continued attention.

The paper further underlines that distribution of corruption matters, and in some cases there may be relatively "clean" areas in an otherwise corrupt set of institutions. There is thus a possibility that islands of integrity may develop in a generally corrupt institutional setting. The paper furthermore suggests that where islands of integrity exist, sector budget support could be channelled towards them. In practice, identifying these "islands" on a sector level is a very difficult methodological process, and the relevance of such an analysis may shift frequently, given e.g. shifts in management and performance of core functions. In addition, it should be noted that Ministries of Finance negotiating budget support with strong international donors, may often be exactly those "islands of integrity" pushing for a reform agenda against the more "corrupt-prone" ministries.

The paper points out that some determinants of corruption is captured by the Public Expenditure and Financial Accountability (PEFA) framework, but that the PEFA framework does not incorporate corruption explicitly. We see the PEFA framework as a useful joint tool in analysing the public financial management in a country, which is important in mapping the framework for i.a. public service delivery. We share the assessment that since the PEFA framework focuses mainly on the public financial management, it needs to be supplemented with other analyses to cover corruption issues more broadly. We agree that assessments such as whether other agents have the ability to sanction government officials if misuse of funds occurs, should be considered included in governance analyses of budget support.

The aim of the paper is to compare the effectiveness of budget support versus that of other aid modalities, in particular project support, in partner countries characterized by corruption. The paper states that to avoid the appropriation of a large part of the resources by corrupt officials, donors might prefer project support in a country even if aid is somewhat fungible and project aid more costly. We find the discussion regarding this issue too limited. There are also experiences indicating that sector and project funds are more easily misused than budget support because they involve parallel flows of funding.

Finally, we are continuously trying to improve on aid effectiveness, and input from CMI and others are very welcome in that regard. We also hope to learn more about budget support as an aid modality through the comprehensive OECD/DAC General Budget Support Evaluation which will be finalised later this year.

Norad, 27 January 2006