You can't learn to swim until you get into the water: policy formulation and implementation on a practical level

Can rural livelihoods in Mozambique be improved by agricultural reform? How do donors and NGOs influence policy and practice to better address poverty and gender issues? This paper reviews the first year of DFID's involvement in PROAGRI (Mozambique Agricultural Sector Public Expenditure Programme), a traditional 5-year agricultural sector investment programme which aims to reform the Mozambique Ministry of Agriculture and Rural Development (MADER). The paper briefly discusses the challenges for PROAGRI implementation. Finally, the implications of these issues for the way DFID and its advisors work are discussed. This paper is of particular relevance to NR advisors and policy makers.

Prior to PROAGRI, donors funded 90 per cent of all public expenditure in agriculture and natural resources through more than 50 separate projects. Now DFID is one of 18 donors signed up to PROAGRI, which aims to co-ordinate activities in the agricultural sector. However, in the short term there have been few returns to government and donor effort. The following challenges have limited the impact:

• conceptual questions: in particular lack of consensus about the scope, functions and target clients of the Ministry, and differing or conflicting needs of different client groups
• sequencing of PROAGRI in relation to wider reforms: ideally, wider civil service staffing reform, reform of government financial management, legal and judicial systems should have preceded PROAGRI implementation. These are just taking off now
• structuring: planning and funding are managed centrally within MADER, with few mechanisms for incorporating views of local government, farmers or other stakeholders
• power struggles: while often not evident in paper policies and plans, these become patent during implementation
• capacity limitations: the programme is ambitious and has moved quickly to implementation, straining both GoM and donor capacity and leaving many key areas (e.g. monitoring and evaluation) as yet unresolved.

• who PROAGRI should consult, on what issues and what are the expected benefits to them
• what are appropriate mechanisms for stakeholder consultation over different issues
• decentralisation of certain decisions to local government may offer more hope of long-term local empowerment than simple deconcentration to local MADER offices. Which decisions should be taken at which level
• how to support farmers and other stakeholders in organising themselves and in preparing them to engage in the policy dialogue.

The following implications are of relevance to those working on Livelihood issues:

• livelihoods advisers do and should work on policy issues. The livelihoods area has important policies with major impacts on poverty and which justify attention
• careful and objective prioritisation of policy issues is needed. In large and complex programmes like PROAGRI, it is easy to dissipate energies on many small issues without making a significant impact
• there are many different ways to influence policy including policy research and dissemination, dialogue with policy makers, and other donors, and support to advocacy groups and the media. One method which has not proved very successful is donor conditionality
• we may need to re-examine the skills mix and location of advisers. The majority of collaborative policy work is time consuming, sensitive and carried out in country. This has implications for the physical location of advisers and the skills they need (e.g. technical, negotiating and language skills).