The Role of International Cooperation in the Decentralisation Process in Low- and Middle-income Countries

The Role of International Cooperation in the Decentralisation Process in Low- and Middle-income Countries


Decentralisation has become a major priority for international development cooperation in recent years. Yet its impact has often fallen well below expectations. A more critical examination of such cooperation is necessary in order to avoid the pitfalls of the past and identify best practice for the future.


Decentralisation is an increasingly common feature of international cooperation today. This reflects a prevalent view among donor agencies that the strengthening of sub-national levels of government is a prerequisite for improved governance and sustainable development. However decentralisation is no panacea and the impact of national decentralisation programmes have often been disappointing. Two forms of international cooperation in support of decentralisation are identified: one that supports the central government unit spearheading the decentralisation process and another that directly supports individual municipalities, usually in the form of decentralised cooperation. Three structural problems are identified that reduce the effectiveness of international cooperation for decentralisation: the persistence of clientelism and local corruption, the tendency for larger urban municipalities to ‘capture’ the benefits of international cooperation, and the lack of institutional memory caused by the high rotation of municipal officials, which reduces the impact of donor-funded capacity-building projects. Finally, some initiatives are mentioned that seek to overcome these problems: support for national municipal associations, the dissemination of ‘good practice’ in municipal development and ‘demand-led’ programmes in support of municipal projects.


Decentralisation is a major feature of the agenda for governance reform today in most low- and middle-income countries (LMICs). Although this is not the case in the Middle East and Central Asia, almost all countries in Eastern Europe, Africa, South and South-East Asia, and Latin America are experiencing some process of strengthening local government systems through the transfer of political, administrative and financial powers from central government. There are several reasons for this growing worldwide emphasis on decentralisation. Many policy-makers believe that it can help to raise citizen participation in government and that it can help to make government itself more accountable to citizens, thereby reducing the level of corruption and improving the delivery of basic services. More recently, others have stressed the enormous potential role of local government in helping to attain the Millennium Development Goals (MDGs) for poverty reduction by 2015.

Contrary to conventional wisdom, the vast majority of decentralisation programmes are ‘driven from above’ rather than ‘demanded from below’. At the same time, there is widespread recognition of the great difficulties that many governments face in their national processes of decentralisation. Most LMICs have a long history of centralised decision-making and powerful stakeholders inside and outside central government are often opposed to the transfer of powers to lower tiers of government. Furthermore, because of this history of centralisation, the human resource capacity of local government systems is usually much weaker than in central government. This means that there is a real risk that local management of the new responsibilities transferred from central ministries will be poor.

Multilateral and bilateral donor organisations have been active supporters of the decentralisation process in LMICs, which they see as a crucial ingredient of the ‘good governance’ agenda. But in some countries, particularly in Latin America, donor support for decentralisation is still regarded with some suspicion. Critics argue that the transfer of responsibilities to local government is merely a tool of neo-liberal policies imposed by foreign donors, designed to weaken the overall role of the State in the development process in favour of the private sector.

International cooperation in the decentralisation process of LMICs takes two distinct forms. First, there is the support by multilateral and bilateral donor agencies to central governments for on-going national processes of decentralisation, which has concentrated on the following activities:

• Support for drafting new legislation affecting local government.
• Support for new central government policy-making bodies charged with leading the decentralisation process.
• Support for reform of intra-government fiscal relations.
• Support for capacity-building programmes for local government.

Amongst bilateral donors, GTZ, the German official development agency, has achieved a well-deserved international reputation for its long-term commitment –usually a minimum of ten years– in support of various national decentralisation programmes, thereby generating a genuine sense of ‘partnership’ in development. By contrast, DFID, the UK official development agency, has displayed a sceptical and more ‘stop-go’ attitude towards support for decentralisation, reflecting the more ‘centralist’ tradition in British public administration.

The second form of international cooperation in the decentralisation process of LMICs is the support by multilateral, bilateral and non-government donor agencies that is given directly to local government institutions, which has concentrated on the following activities:

• Promoting good urban governance and municipal management.
• Protecting the urban environment (water and sanitation, environmental planning and pollution reduction, and the cultural heritage).
• Promoting equality, social inclusion and urban regeneration (through health and welfare, education and housing).
• Local economic development (transport, enterprise development and co-operatives).

The professional skills and organisational capacity of municipalities in OECD countries is often comparable with that of central government and is usually of a much higher standard than in LMICs. This fact underpins the alleged ‘comparative advantage’ of so-called ‘decentralised cooperation’, namely the direct collaboration between municipal organisations in OECD countries and their counterparts in LMICs, often through ‘town twinning’ arrangements and ‘people-to-people’ solidarity linking. The Federation of Canadian Municipalities (FCM) is one of the national municipal associations with the longest experience of decentralised cooperation, beginning in 1987. The FCM has involved over 100 Canadian municipalities internationally, including 40 partnerships linking Canadian municipalities with local governments in over 20 countries in Africa, South-East Asia and Latin America. Among multilateral donor agencies, the EU has emerged as a major player in this field, funding the creation of thematic cooperation networks that promote ‘knowledge transfer’ between small groups of municipalities in Europe and their counterparts in Asia (Asia Urbs, and Latin America (Urb-AL,, respectively. The AsiaUrbs programme has supported 65 projects, involving a total of 211 EU and Asian local governments. In addition, 252 non-governmental or community based organisations, universities and private sector bodies are participating in the project partnerships.

International cooperation has not been a major factor explaining the success of national decentralisation programmes. This is in sharp contrast to international cooperation for improved macroeconomic management, where success has often been achieved through externally-imposed conditionality from the IMF and World Bank. Decentralisation programmes require strong domestic ownership in the form of political support, leadership and commitment by central government for their ultimate success. By contrast, decentralisation programmes that were primarily donor-driven have invariably proved to be unsustainable.

A number of major issues can be identified that have implications for international cooperation in support of decentralisation, with associated lessons that can be learnt in order to improve the positive impact of donor involvement. First, decentralisation is no panacea. The persistence of local structures of clientelism and nepotism in many countries has meant that decentralisation programmes have often not produced any noticeable improvement in the delivery of basic services. In some cases, levels of municipal corruption have actually increased, with negative effects on service delivery. For this reason, there is no a priori case for assuming that international cooperation should necessarily give a greater role to sub-national government in achieving the MDGs. In those countries where local power structures of inequality are deeply entrenched and local civil society organisation is weak, the transfer of new service delivery responsibilities in the priority sectors of basic education and primary health may actually slow down the achievement of national MDGs.

Secondly, the distributional impact of international cooperation for decentralisation that goes directly to sub-national bodies is extremely unequal. This is just as true for programmes financed by multilateral and bilateral donor agencies as for ‘decentralised cooperation’. The vast majority of municipalities in LMICs are small, rural and impoverished. By contrast, most international cooperation for decentralisation has focused on large, urban municipalities that are both richer (in terms of per capita income) and that have local administrative systems that are better endowed in terms of human resources. A major consideration here is the more ‘globalised’ cultural environment of large urban municipalities, which are far better able than their small, rural counterparts to know about and access opportunities for international cooperation. This bias takes on an added relevance in view of the emerging debate concerning the role of local government as a vehicle for attaining the MDGs. This will require a major shift towards channelling support for municipalities in rural areas, where the basic human development indicators that underpin the MDGs are much lower than in urban areas.

Thirdly, the ‘value for money’ of donor-funded capacity-building programmes for local government has generally been very low because of the far higher rotation of senior staff in local government systems in LMICs than in OECD countries. This problem is especially grave in many Latin American countries where senior local government personnel occupy ‘confidence posts’ and leave office when the municipal political leadership changes via the electoral process. One possible way of increasing the ‘cost-benefit’ ratio of such municipal training programmes would be to introduce limitations on the number of confidence posts. This is already the case in Chile, where confidence posts are strictly limited to three per municipality.

Three recent positive initiatives by donor agencies that address some of the above problems in international cooperation for decentralisation merit special mention. First, in several countries bilateral agencies are giving support to the establishment and strengthening of national municipal associations as an integral part of their support for decentralisation programmes. The move reflects a tacit recognition that a strong and representative national body capable of lobbying and negotiating with central government on behalf of all municipalities is an essential factor in ensuring the sustainability of the decentralisation process. By representing the common interests of municipalities that may be controlled by different political parties, such national municipal associations also contribute to the wider process of democratisation through the implanting of a consensual political culture.

A second positive donor initiative involves support for the creation of databanks for the identification and dissemination of ‘good practice’ in municipal development. Several such databanks now exist within countries undergoing decentralisation programmes (eg, the Banco de Experiencias Locales operated by the Universidad de Quilmes in Argentina,, at a regional level (eg, the Knowledge Fair on Local Governance in Latin America operated by UNDP, and at a global level (eg, the Thematic Centre on Local Government Initiatives for Sustainable Development, www.iclei.or/habitat-centre). A third initiative involves the introduction of a demand-led approach to donor support for municipal strengthening. Under this approach, first pioneered by the World Bank in Colombia and followed by DFID in the Enabling State Program in Nepal (, municipalities and local community organisations are invited to propose projects for grant-aided funding. By rejecting the traditional supply-led approach under which funding priorities were imposed from outside, the initiative seeks to ensure that the allocation of funds in support of decentralisation more truly reflects local-level priorities. In order to counter the advantages referred to above that larger, urban, municipalities have in accessing external funding, these donor-funded programmes make a determined effort to advertise their existence to small, rural, municipalities using local radio stations.


Decentralisation is no ‘quick fix’. Even where a municipality has been granted new financial resources and enhanced responsibilities, the delivery of basic services to its citizens will not necessarily improve. Decentralisation will always be a necessary but not a sufficient condition for improved governance. Foremost among these sufficient conditions is the existence of strong mechanisms of accountability –both ‘downwards’ to local citizen organisations and ‘upwards’ to national monitoring bodies of central government–. Through its support for these accountability mechanisms, international cooperation can play an important role in ensuring that those sufficient conditions are in place.

Finally, as in other areas of development cooperation, donor coordination needs to be enhanced in support of national decentralisation programmes. This is especially necessary in the case of decentralised cooperation, where national ‘programmes’ exist in name only. In practice, they often consist of a multiplicity of pilot projects that have arisen on a purely spontaneous basis, and with minimal communication between each other. In order to enhance the overall benefit of this form of cooperation, donors should establish forums for the co-ordination and dissemination of information and establish systems for basket funding where appropriate.

Andrew Nickson
International Development Department
University of Birmingham
United Kingdom