The national roads system in Uganda is experiencing a period of exceptional activity and promise following a large increase in public spending in the sector and institutional reforms designed to enhance substantially the efficiency of this investment. This report undertakes a political economy study of the changes in the sector as a contribution to thinking about the most suitable form and content of donor support to the reform process.

The study builds on previous work showing that an increase in both the volume and the efficiency of investment in national roads in Uganda would contribute substantially to raising the pace and improving the structure of economic growth in Uganda and in the East African sub-region. It can therefore be seen as an exercise in the political economy of growth policy options.

The report uses a ‘layered political economy approach’, the generic features of which are described in the Preface. Sections 1 and 2 then describe some major features of the political context of policymaking in Uganda, the sector background and the major stakeholders in the process of reform. They then set out three possible scenarios. These are:

• That the 2008 policy changes regarding national roads signify a substantial shift in presidential priorities and policymaking style, such that there is considerable scope for donors to have an impact by plugging financing gaps and providing conventional technical assistance (TA) on a large scale.
• That, although the above is not the case, the institutional changes of the past year have nonetheless altered the incentives or decision logics within the sector in important ways, and henceforth the balance of forces will be more favourable to those wishing to see real change in ways of working. This would be another scenario favourable to a conventional donor response.
• That the changes involve neither a transformation of the systemic political environment, nor a definitely transformed set of incentives for sector actors, but there is nonetheless some ‘room for manoeuvre’ arising from the dynamics of the reform process.

In the last scenario, there is scope for donor-supported action to tip evenly balanced equilibria in a pro-reform direction and contribute to pro-growth and pro-poor outcomes ‘against the odds’. However, this calls for a particular sort of donor programming, which permits the solution of a large number of small institutional and collective action problems in a very dynamic setting.

The report argues that the third scenario is in fact the one that corresponds most closely to the current situation in Uganda. This argument is grounded in previous political economy studies in Uganda by the authors and others, and it draws heavily on two and a half weeks of interviews and discussions with participants in and close observers of the roads reform process.

The first possible scenario is rejected on the grounds that the evidence of a sea change in President Museveni’s approach to the roads sector is unconvincing. Even if he has new reasons to take road building and maintenance more seriously, and this was an important factor in the initiatives of 2008, his own system of rule will tend to prevent this leading to any dramatic change in the political factors affecting what happens in the roads sector. While the funding of maintenance through the new Road Fund seems secure, it would be unwise to assume that either the Uganda National Roads Authority (UNRA) or the Ministry of Works and Transport will be immune in the future to the pressures to provide political rents that damaged the performance of the sector in the past.

The second scenario is one in which the decision logics of several key actors move in a pro-reform direction. This seems unlikely because the demand for service delivery outcomes remains weak compared with the appetite for patronage, which is the key to the cost benefit calculations made by the President and his supporters. There are doubts about how far the creation of UNRA has ushered in an organisation of a new type, characterised by more performance-oriented incentives. Meanwhile, the suppliers of construction and consulting engineer services in the private sector, who are victims of the system in some important senses, face prohibitive collective action problems as advocates of reform. Including donor money and influence in this discussion does not open up markedly more hopeful prospects, as donors suffer from well understood political economic impediments of their own.

That leaves the possibility of an agenda of exploiting the limited but not insignificant room for manoeuvre created by the reform as a process. We are concerned with dynamics in which different actors define and redefine their interests partly in response to what others have decided to do. Typically, change processes of this type present opportunities in which an intelligent and respected third party can broker meetings and agreements, and help the main players to overcome what would otherwise be situations of stalemate or logjam. In the particular case of Uganda roads, there seems scope for third party action of this general type in at least the following fields:

• Communication about performance;
• Brokering otherwise missing dialogue among key players;
• Facilitation of countervailing networks of influence;
• Lowering barriers to collective action by private actors;
• Facilitating appropriate forms of ‘infant industry’ support to local firms;
• Mobilising influence to enable otherwise blocked organisational transformations.

The central proposition advanced in the report is that donor support to the sector, particularly but not exclusively the UK Department for International Development (DFID) component, should be conceived primarily in terms of this type of third party facilitation of change in a dynamic, multi-stakeholder environment. This is where the most important gap is to be found.

Such an approach would not be unprecedented. It has been deployed effectively in the series of DFID-funded programmes originally titled Making Markets Work for the Poor. Additionally, in Latin America and elsewhere, the types of outreach and engagement activities that DFID’s relatively flexible grant funding permits has been the basis of mutually satisfying partnerships with World Bank task managers among others. This is another relevant precedent.

The report does not get into details about possible programme designs. However, it follows from the conclusions of the political economy analysis that there will be a strong case for adopting a process, not blueprint, design. The relevant opportunities cannot be mapped out in detail in advance, and there will be considerable scope for learning from experience about what kinds of facilitation work best. Similarly, the investment in staff with the relevant networking and facilitation skills should be viewed as the most powerful component of any package of support, not the funding, the hardware or the conventional TA. Staffing decisions will be crucial, and in this instance we would recommend a careful pairing of local and international personnel.