Chapter 10. Assessment and Lessons Learned

Introduction

10.1 This final chapter:

1. Gives our overall assessment against each of the evaluation criteria stipulated in the Terms of Reference.
2. Identifies a number of lessons learned.
3. Reviews implications for possible adjustments to the 2004 strategy.
4. Provides a summary of recommendations.

10.2 Offering a verdict on nearly two decades of one donor’s programme again raises the challenges of methodology mentioned in the first chapter. There is no straightforward way to add up assessments for different sectors/components of the programme. Contexts vary enormously (including in the degree of difficulty of what is attempted, and the ease of measuring what is achieved). Moreover, there is no simple counterfactual, and we have not had the scope to compare this country programme in detail with Danida’s programmes in other countries or with other donors’ programmes in Uganda.⁶³

10.3 The standard OECD DAC criteria cited in the ToR are the main reference points for our overall assessment. However, we also need to form a judgement on the consistency and coherence of the programme elements. Does the Danish programme add up to more than the sum of its parts? Does it enhance, or conceivably detract from, the other things that the government and donors are doing?

Assessment

Assessment Criteria / Impact

From ToR: Danish assistance will be evaluated in light of Ugandan national policies, and in the context of activities of other donors. Achievements will, to the extent necessary, be assessed against contemporary goals and standards. Assessment of impact will generally be restricted to the overall level of achievements of the combined support from donors to Uganda and the country’s own efforts. cf. DAC definition64: Impact: Positive and negative, primary and secondary long-term effects produced by a development intervention, directly or indirectly, intended or unintended.

10.4 Our assessment of Uganda’s general performance and the overall contribution of aid was given in Part II above. We concluded that the overall contribution of aid to Uganda’s performance was positive. Aid played a substantial role in financing the rehabilitation of Uganda’s economy, and then, from the early 1990s, the implementation of reforms and of a poverty oriented development strategy. Without the aid that it received, Uganda’s economic growth would certainly have been slower, its political stability might also have been threatened, and more people would have been left in poverty.

10.5 We focus here on the Danish contribution. Given our positive verdict on combined donor assistance to Uganda, a negative verdict would imply that Danish aid was, on balance, significantly less effective than aid as a whole. Danish aid did include a number of unsuccessful interventions, and, as we discuss below, there are ways (with hindsight) in which it could have been made more efficient and effective. But we have formed a strong impression that Danish aid has generally been of high quality, and that Denmark’s contribution to the progress that Uganda has made is rightly valued.

10.6 Against each of the evaluation criteria that follow, we explain our overall judgement, and then draw attention to Danish programme elements that seem to be problematic or to have some special merit. This identification and analysis of exceptional elements contributes directly to the identification, in the next main section, of lessons learned.

10.7 Only a very unambitious programme could be without blemish; the criticisms we offer should be seen in that light. It is notable that the Danish aid programme has not sought easy options: it has chosen some difficult sectors and components (e.g. agriculture, governance); it has sought to work in the more deprived and difficult environments of Uganda; and it has sought to innovate. It should also be noted that some of the things we criticise were standard practice in their day, and that much of the evidence for our critique comes from Danida’s own evaluations and reviews.

Relevance

From ToR: Was relevant in relation to country poverty needs as expressed in national poverty reduction strategies and sector policies and strategies? Adhere to overall Danish and Ugandan policy goals concerning poverty Adhere to cross-cutting issues (gender, environment, popular participation) and themes (children and HIV/AIDS), as stipulated in the partners’ policies cf. DAC definition: Relevance: The extent to which the objectives of a development intervention are consistent with beneficiaries’ requirements, country needs, global priorities and partners’ and donors’ policies.

10.8 The ToR frame relevance in terms of poverty reduction objectives; the DAC defini-tion relates more generally to “requirements, needs and policies”. We have seen that the focus of national objectives changed over time. As put in one of the Thematic Papers:

From the inception of Danish Bilateral Cooperation with Uganda in 1987 to date, GOU has pursued two overarching development policy frameworks. Between 1987 and 1996 the major thrust was reconstruction of infrastructure and achieving minimum economic recovery. Then from 1997 onwards the national development objective and strategy has been poverty eradication. Denmark’s support has also been premised along these lines. (TP2, 6.2)

10.9 For Denmark as well as for Uganda, poverty-orientation was not at first such an explicit objective as it later became. Nevertheless, from the outset the Danish aid programme was clearly relevant both to national recovery objectives and more specifically to poverty reduction (see the discussion in 9.2–9.10 above). However, poverty strategies were not then well-defined. After the mid-1990s, as both Denmark’s and Uganda’s poverty objectives and poverty strategies became more refined, Danish aid adapted to become more relevant and more aligned with Uganda’s poverty reduction strategy.

10.10 On relevance to the main cross-cutting issues (gender, environment, HIV/AIDS), our discussion in 9.13-9.15 indicates a qualified endorsement. There was conscious attention to all these issues; on none of them is there room for complacency. The ToR criteria of “participation” and “children” have not been such explicit themes in the programme. However, it has included elements (e.g. special education, primary health care) in which children are direct beneficiaries, and the emphasis on economic and social services that benefit poor families means that children are prominent among the beneficiaries of the programme as a whole. Participation has a been a design principle of Danida interventions (e.g. the involvement of communities in roads and water programmes), and there has been support to democratic participation throughout the period, including noteworthy support to the Constitutional Commission and to decentralisation, as well as the human rights and democratisation programme.

10.11 The Danish programme was also consistently relevant in the nature of the aid offered. Uganda’s poverty and its debt status made Denmark’s reliance on grants rather than credits appropriate. Danida’s practical readiness to finance recurrent costs and not only capital investments also added to the relevance of its assistance. Denmark’s readiness to work with a range of partners – from NGOs and civil society, as well as government – was also relevant. It was natural to focus first on a reforming government that deserved support, but Denmark also recognised the role of civil society in holding government to account, and of non-government service providers (e.g. the PNFP sector in health). Over time, too, more explicit collaboration with other donors has also been both appropriate and responsive to national strategy.

10.12 Despite this generally positive assessment, not every element of the programme passes the test of relevance. Two early major projects – the grain silos and the Kampala dairy rehabilitation – were not relevant to Ugandan needs. It is a matter of judgement whether the acclaim they brought as “flagships” of Danish aid could justify their cost. At least one programme that was relevant in principle, but a failure in practice – the support to national farmers’ associations – was continued for longer than justified. The Private Sector Development programme is not particularly relevant beyond the immediate level of individual firm constraints (see Box 8.11).

10.13 The financial programme aid provided during the 1990s (especially in its balance of payments support and debt relief forms) was highly relevant. Unfortunately its relevance to policies and public expenditures impacting on poverty was not well recognised and it was not given the strategic consideration that it deserved. In discontinuing finan-cial programme aid when the MDF assistance terminated, the programme missed a relevant opportunity.

10.14 We have also noted that Danida deserves special credit for the relevance of its support to the health sector. The focus on building basic health systems, including the capacity for planning, supporting and managing district services, was appropriate, as was Danida’s persistence in focusing on basic health systems in the face of changing donor fashions.

Effectiveness

From ToR: Was effective in making contributions to the overall political, economical and human development of Uganda – both at the national and local level – and in specific sectors. cf. DAC definition: Effectiveness: The extent to which the development intervention’s objectives were achieved, or are expected to be achieved, taking into account their relative importance.

10.15 Overall we judge that the programme was effective. Areas where Danish aid was conspicuously effective included:

• debt relief;
• rural roads programmes (arguably rural roads have been the programme’s most successful contribution to agricultural development);
• support to the rehabilitation and expansion of basic health services, especially through the essential drugs systems;
• rural water and sanitation programmes;
• support to economic and institutional recovery in Rakai District;⁶⁵
• support to special education;
• in HRD, support to the justice and accountability systems of government.

10.16 Danish aid has made distinctive contributions (over and above provision of infrastructure and services) in at least the following areas:

• water and sanitation (system development, innovation – including decentralised implementation of water projects and privatisation of construction, capacity building as well as service delivery);
• gender (formative influence on gender analysis and advocacy during the 1990s);
• decentralisation (effective partnership with GOU in designing and implementing decentralisation strategy; the demonstration effect of Rakai DDP);
• labour-based roads;
• special education;
• in HRD, its advocacy and support for human rights.

Efficiency

From ToR: Was efficient, i.e. compared to different ways of delivering assistance, and efficiently aligned with national policies and harmonised with assistance of other donors. cf. DAC definition: Efficiency: A measure of how economically resources/inputs (funds, expertise, time, etc.) are converted to results.

10.17 We judge the Danish aid programme in general to have been efficient. There are a number of bases for this overall judgement: we have found few conspicuously inefficient programme components; the correspondence between commitments and disbursements is impressive; and the management arrangements for the programme (Volume 4, Thematic Paper 10) have generally been well adapted to requirements. There have been conscious efforts to make the programme more efficient by aligning it, where appropriate, with GOU systems, and by coordinating with other donors. And observer perceptions of Danida’s efficiency are generally favourable. There is a particularly positive perception from the water sector:

At institutional level, the respondents at DWD overwhelmingly submitted that Danida’s support has delivered more efficiently compared to other donors. The reason for this assessment is that Danida has completely integrated its support into the Government structures and that disbursement of its funding is more predictable and timely. (TP2, 6.5)

10.18 There are some contrasts in efficiency between different components. For example, commodity import support, as operated, was highly inefficient while debt relief was extremely efficient; the grain silos and the Kampala dairy were not an efficient use of resources; the ratio of benefits to costs of the PSD programme is unfavourable unless a high value is placed on intangible political benefits; by contrast, labour-based road works (and the focus on feeder and secondary roads) and the water and sanitation programme rate highly for efficiency.

10.19 A number of factors had a positive influence on efficiency over time:

• The relaxation of (explicit and implicit) tying of Danish aid, which has made it more efficient in implementation, as well as reducing the risk of inappropriate project/design choices because of conflicting interests.
• The decentralisation of aid management by MFA, which has helped the programme to be responsive at local level.
• The integration of development and diplomatic aspects of MFA work.
• Increased harmonisation and alignment with other donors and government in various sectors.

10.20 Thematic Paper 7 notes that: much donor funding for NGOs has been ad hoc, short-term and ’projectised’, an approach to programme management that combines high transaction costs with uncertain impact. This has prompted donors to look for alternative funding modalities. Danida’s pragmatic response, the Human Rights and Good Governance Liaison Office (HUGGO – see Box 9.4) has made for greater efficiency in the management of governance interventions, both by its concentration of capacity and by its link to programmatic approaches for managing NGO support.

Sustainability

From ToR: Was sustainable in relation to its choice of modalities, i.e. whether it was developed and undertaken in co-operation with relevant national and local authorities and in different sectors? And was sustainable in the sense that it strengthened the organisational capacity of relevant national and local authorities or organisations. cf. the DAC definition: Sustainability: The continuation of benefits from a development intervention after major development assistance has been completed. The probability of continued long-term benefits. The resilience to risk of the net benefit flows over time.

10.21 We note that there is potential ambiguity in the sustainability criteria offered. The DAC definition has a number of distinct components, while the ToR criteria leave open the possibility that an intervention could strengthen national capacity without necessarily attracting the national resources to assure continuity. It is clear that the programme increasingly tended to use more sustainable modalities (reflecting the moves beyond Danida-specific projects and sector programmes to fuller engagement with government-led SWAps). At the same time it is also clear (a) that Uganda’s PRS will require significant donor support, including support for recurrent costs, for many years ahead; and (b) that Danida (and like-minded donors) accord higher priority to certain institutions than the Government does; their adequate funding is therefore likely to continue to require continued donor support (either through dedicated funding, or as a result of donor influence through budget dialogue and agreed conditions). This applies particularly to institutions for accountability and justice. Taken together, these, factors suggest that Danida (and other donors) should pay more explicit attention to the DAC criterion of “resilience to risk” in future plans. In particular, will GOU be willing as well as able to maintain benefit flows over time? What are the risks to continuation of donor support where continued support is appropriate?

10.22 Danish procedures for taking account of sustainability in project design are strong, but they cannot easily overcome institutional weaknesses, which determine, for example, whether the necessary resources for operation and maintenance are actually mobilised and effectively used. This point is illustrated, for infrastructure, in Box 10.1.

Box 10.1: Sustainability in Infrastructure Sectors This evaluation’s Thematic Paper on infrastructure draws attention to persistent issues in sustainability, despite Danida’s careful attention to the issue: Danida-funded investments are known for being elaborate in addressing the issue of sustainability. This is obvious not only from the comprehensive planning that appears in the project documents but also in terms of concrete action taken to ensure sustainability. In the water sector, this has been pursued through charging water-user fees as well as putting in place appropriate institutions to assure sustainability such as water users’ committees and the training and equipping of pump mechanics to carry out preventive maintenance and small repairs. On the project for renovation of Entebbe International Airport, Danida funded a two-year contract for ICAO maintenance experts to train Civil Aviation Authority staff to maintain the airport facilities. Despite these efforts, sustainability remains a major problem in infrastructure projects. In the Water and Sanitation Sector, the problems of operation and maintenance have been a consistent challenge. This has been true not only at the facility level but also in areas where local governments and central government have responsibility. It has also been the case with facilities like urban water where management of facilities has been contracted to private operators. The perpetual problem of sustainability is partly the reason why reforms were launched, in order to ensure that users take greater responsibility for financing of the services provided. The urban water sector reform strategy was more forthright – “the sector cannot sustain itself, service its debts, raise capital for new infrastructure, and fund renewal or growth, despite adequate tariffs”. The problem was also acknowledged during the first joint GOU–donor review and as a result the mission emphasised the need to focus on operation and maintenance in subsequent interventions. The challenge of sustainability is also an issue in the Road Sector. GOU had made a commitment in 1996 to provide additional resources amounting to at least USD 4 million every year for maintenance; this figure was reduced to USD 2 million in 2002, but even this has not been fulfilled. Source: Volume 4, Thematic Paper 2.

Coherence

Under this heading we consider: the consistency of Danida programme elements with each other (do they complement each other in a positive way?); coherence between interventions supported by Danida and the wider activities of the government and other aid partners. We note that coherence is not just a matter of a static “balance” between elements of the programme: it concerns also the dynamics of building relationships and applying influence.

10.23 Overall, the Danish aid programme to Uganda has shown strong coherence.⁶⁶ It has maintained a balance between the different “legs” (social, productive, institutional) that were established at the outset. The breadth of the programme has been used to build support from both Danish and Ugandan stakeholders. The airport and the grain silos are the most spectacular examples of constituency-building,⁶⁷ but it is clear that Denmark’s track record of social and infrastructure support has also given it credit to be used in pursuing its institutional and governance objectives.

10.24 Clear elements of complementarity in the programme include the balance between work with government agencies and with NGOs and civil society on governance issues; between experience gained in Rakai and support to the decentralisation programme of central government. Given Danida’s exceptional involvement and expertise in the development of Uganda’s mechanisms for decentralisation, it is disappointing that some of its other activities (e.g. inputs to health planning) were not more attuned to the institutional requirements of decentralisation. However, reconciling decentralised and sectoral approaches is always a difficult challenge.

10.25 Denmark has also sought coherence and complementarity by staying out of sectors and programmes considered adequately supported, and by participating in joint government–donor coordination mechanisms. In this vein, it stayed out of basic education, it focused on basic health systems rather than add to the imbalance of HIV/AIDS-specific funding, it chose geographical locations for water and roads programmes that dovetailed with what other donors were doing, and for the most part, it kept clear of large-scale infrastructure. It played an influential role in enhancing donor coordination in water and sanitation, in agriculture and in various aspects of governance.

10.26 Increasing attention to harmonisation and alignment under government leadership changes the way in which the internal coherence of an individual donor’s aid programme should be viewed. The more that donor interventions are consciously fitted into an overall national strategy and national priorities, the less relevant it is to expect an individual programme to be internally “balanced”, e.g. between sectors or geographically: what matters is whether there is an appropriate balance in the aggregate efforts of donors and government. It also means that, for individual components of the Danida programme, coordination with what other actors are doing in the same field is likely to be more important than coordination with what Danida is doing in other fields. Danida’s long-established practice of trying to focus on a limited number of sectors and sub-sectors should make it well attuned to the requirements of achieving an effective division of labour with other partners (a requirement that is reinforced by its intention to join the Uganda Joint Assistance Strategy, discussed later in this chapter).

10.27 At the same time, all agencies have to take account of their financiers (in Danida’s case, ultimately, Danish taxpayers) as well as their Ugandan beneficiaries. From this point of view, the balance of the programme continues to be a relevant concern (e.g. in continuing to contribute to economic growth, to social services and to institutional development, in demonstrating Danish contributions towards reducing fiduciary risks faced by its aid, and so forth). There is also a legitimate concern to avoid a focus so narrow that it might undermine Danida’s country-level understanding of Uganda’s situation and requirements.

Lessons Learned

Introduction

10.28 The context for Danish aid to Uganda has changed dramatically in the period under review. Uganda itself presents a very different picture in 2006 than it did in 1986. And the concepts that inform an aid programme have themselves evolved significantly. This is manifested in assumptions about appropriate economic policies for different contexts, about the need for explicit poverty reduction strategies, about the direct and indirect roles of the state in development, about good governance, about the importance of country ownership for capacity development and sustainability, and about many other aspects of what constitutes effective aid and good practice in aid delivery. Ugandan experience has itself made an exceptional contribution to these developments in thinking. Against this background, the present section identifies a number of lessons that may be drawn more specifically from the experiences of Danish aid to Uganda.

The value of concentration and focus

10.29 Denmark has almost certainly achieved more by operating as one of the larger bilaterals in Uganda, than if the same resources had been spread more thinly over more countries. As well as enabling it to operate at a significant scale in its chosen sectors, this approach has increased its access and legitimacy in voicing concerns to government, as well as vis-à-vis other donors. It has been able to build up experience and comparative advantage in a number of sectors (and sub-sectors within them) including health, HRD and decentralisation.

The value of consistency and staying engaged

10.30 In spite of difficult periods, the Danish programme has shown great stamina in a number of important areas, including support to the development of institutions for essential drugs management; support to justice and accountability institutions; and support to decentralisation. It has been less fickle than some agencies (e.g. in resolute support for basic health systems when more headlines were available for support to vertical HIV/AIDS programmes). Agriculture is another sector where Danida has shown great persistence (but its staying power may not yet be fully tested – see later comments on future directions for this sector).

The value of consistent strategy, flexible implementation

10.31 The Danish programme has shown strategic consistency without an over-rigid advance programming of expenditures. In some sectors, there was not always as much flexibility in implementation as might have been desirable (especially in the context of support for a broad government strategy). There may be quibbles over the pace and pattern of moves towards more reflective and collaborative aid modalities in different sectors, but it makes sense to be pragmatic when there is such a variety of sector contexts.

On capacity development

10.32 The programme’s experience shows the value of working with government systems and drawing on local human resources, as well as the value of basic capacitation (providing the facilities to enable organisations to function). Danish-funded consultants were less successful when they attempted to export a Danish model (farmers’ associations), and more successful when they worked with government in listening mode (decentralisation).

The dilemmas of visibility and the “resource base”

10.33 Most of Danida’s concerns for visibility are quite understandable but they raise problems. In the early years the direct tying of aid led to inefficiencies (CIS) and probably had an unhelpful influence on some of the less successful major investments. It is good that Denmark has generally retreated from the tying of its aid; hopefully this is an improvement that will not be reversed.⁶⁸ On the other hand, there is a quite legitimate concern to maintain the support of the Danish public for aid. This has two direct aspects. One challenge, in the light of the increasing, and justified, trend towards collaboration in aid delivery, is how to maintain domestic support for aid if it is less obviously visible as Danish aid. This can be addressed by considering the balance of the Danish portfolio between merged and separate interventions, by seeking better reporting on all forms of aid, and by seeking to raise public appreciation of what constitutes effective aid (see discussion below of the UJAS framework). A second challenge is to allay fears about corruption and the diversion of Danish aid funds. Here Danida’s approach has been admirably strict, but also rather one-dimensional, and has hampered other objectives, including the building up of domestic institutions and participation in more collaborative aid modalities. This links to our next observation.

Need to avoid rigid paradigms on aid modalities

10.34 Partly because of its concerns about accountability, Danida appears to have adopted a rather restricted paradigm concerning aid modalities. It has been very reluctant to move into disbursement via government systems where it perceives significant fiduciary risk. A number of other donors have been more ready to accept a degree of risk, and seek to mitigate it, because of the weight they attach to the strengthening of initially weak government systems. It would have been helpful for Danida to seek more of a balance between different modalities rather than setting such a high entry threshold for new modalities. This would provide more opportunities to gain experience with modest levels of risk. In this connection, budget support should be viewed more as a complement and less as an alternative to other modalities. That Danida did not continue with any form of general budget support after its support to debt relief was a missed opportunity to learn from experience.

On ownership and influence

10.35 The entire Ugandan experience with aid, not just Denmark’s Uganda programme, has important lessons about ownership and about the extent, and the limits, of donor influence. Ultimately, donors’ ability to influence political governance issues is limited. So is their ability to dictate policies. But where there is government ownership of strategies and an overlap of government and donor interests, there is great scope for collaboration in the design, implementation and monitoring of effective programmes and policies.

10.36 At the level of individual interventions, in the interests of ownership and capacity development, Danida needs to resist the temptation to interfere in the day-to-day decision-making of the activities it supports. (The concern for strict accountability easily leads to an involvement in the details of resource allocation and management. This point was particularly commented on by a number of the NGOs that Danida supports.)

Lesson learning

10.37 It is encouraging that the Danida programmes in Uganda generally appear to have a good record of seeking to learn and apply the lessons of experience. For example, the 2004 country strategy includes an extensive discussion of lessons learned (see Appendix E, 22 for extracts). It concludes:

In terms of challenges presented, insufficient focus has been targeted at fostering sustainability of the development interventions, at integrating cross-cutting issues (gender equality, environment, human rights and democratisation) and at fostering greater interplay between the various sector programme initiatives. (MFA Danida, 2004h)

Future Directions

Context

10.38 Uganda faces major challenges: in economics, to achieve sustained growth that does not perpetuate aid dependence; in politics, to advance democratic governance and national coherence. These two come together inasmuch as the biggest opportunity for poverty reduction and a boost to economic growth would be peace and a peace dividend in the north.

10.39 Donors need to recognise that the high quality of their relationship with the Ugandan government over the past decade may have been exceptional. There is no longer such a strong degree of common interest between donor objectives and the basic interests of the government. It remains important to stay engaged, to work with government and with civil society, and to apply the lessons of experience on aid effectiveness.

Danida strategy in the context of UJAS

From ToR: Finally, the evaluation should review the progress against the specific indicators of the Danish 2004 country strategy – and make recommendations to possible adjustments of the strategy.

10.40 The specific indicators of the 2004 country strategy relate to modes of aid delivery (they are reproduced at 25 of Appendix E). All of them imply moving in directions that the lessons of this evaluation would support. In many cases, their achievement requires combined efforts by Denmark, other donors and the Ugandan government. At Appendix G, we reproduce RDE’s 2006 status report on progress against these indicators. The report demonstrates encouraging progress by GOU and donors together towards greater alignment of aid with the national poverty reduction strategy, and the development of more integrated monitoring systems. A first Annual PEAP Implementation Review (APIR) is being held in 2006, and is linked to an exercise to achieve a better division of labour among Uganda’s donors (this is spearheaded by the participants in the Uganda Joint Assistance Strategy (UJAS) – see below).

10.41 At the same time, progress towards channelling more of Denmark’s aid to Uganda through sector and (especially) general budget support, has been slower than the 2004 targets implied. Against a target that 80% of Sector Programme Support should be in the form of sector budget support by 2006, RDE reports an actual figure of 15%. Commencement of general budget support has been deferred (citing governance concerns) and the target for 2008 has been reduced from 25% of the country frame to 7%.

10.42 Since the evaluation commenced, Denmark has decided to join the other partners in the Uganda Joint Assistance Strategy without delay. This makes sense. The UJAS framework (see Box 10.2) provides common indicators but allows separate decisions. The variety of approaches amongst the existing UJAS partners is indicated by their very different commitments to budget support, also illustrated in Box 10.2. Thus joining UJAS does not require Denmark to submerge its identity or lose its autonomy as an aid partner; but this way of working will create challenges about how to present and to justify Danish assistance to the Danish public.

10.43 UJAS also provides a framework in which Danida can strengthen progress towards the 2004 country strategy targets, including those that have been lagging. In the light of the evaluation’s findings, we comment below on:

• continuing areas of focus (comparative advantage) for Danida;
• issues for Danida to pursue within the partnership;
• implications for reporting, monitoring and accountability.

10.44 The “division of labour” exercise summarised in Box 10.2 implies an even further narrowing of Danish focus within Uganda. Based on our evaluation findings, we would recommend:

• Denmark’s involvement in the health sector should continue, based on the depth of experience and comparative advantage built up over many years.
• Denmark should take a continuing interest in decentralisation, and seek through UJAS to encourage moves towards greater coordination and a “decentralisation SWAp”.
• HUGGO represents valuable planning and management capacity for the HRD sector, with potential for shared benefits among UJAS partners.
• Denmark should maintain its role in the agriculture sector (see specific suggestions below).

10.45 Danida clearly recognises that, if poverty reduction is to be achieved, the share of support to agriculture and efforts to improve its effectiveness must reflect its importance to rural livelihoods. Increasing the effectiveness of aid to agriculture will be crucial. Recent shifts in modes of aid delivery present both opportunities and risks; in the case of the latter, a reduction of learning and innovation that has been a characteristic of past Danida support to agriculture. Policies and interventions supporting sustainable pro-poor agricultural growth must become more central to poverty reduction. The concerns which were raised in Thematic Paper 1 about the design and delivery of support services to farmers suggest that the agricultural extension system for the poorer areas of the country with little ability to pay should be re-examined. While budget support has a number of advantages, the particular characteristics of agriculture, which is essentially a ’private’ sector, suggest that it will be important for Danida to continue to balance different aid instruments for Ugandan agriculture. In this regard, a revival of the approach adopted by Danida under the Household Agricultural Support Programme should be examined. In a widely praised ASPS1 initiative, Danida successfully devised and tested ways and means of reaching poorer households and districts. HASP pioneered extension approaches which were later successfully scaled up under the PMA by NAADS. It is to be expected that the PMA will continue to need strategic inputs of this nature to blaze the trail in hitherto deprived areas of the country.

Box 10.2: The UJAS Approach UJAS partners will continue to support policy analysis and engage in dialogue to help the government prioritize expenditures and reform policies to more rapidly achieve PEAP targets and the MDGs. Such initiatives may be particularly relevant for conflict resolution and for strengthening of political governance. But in all instances the UJAS partners will avoid parallel interventions and will work within appropriate Ugandan frameworks.   A flexible approach to assistance is required. UJAS partners therefore will adapt their level and nature of support in response to emerging national developments and needs. Adjustments will be made in consultation with the government and with other partners. The annual UJAS review process (based on the annual PEAP implementation review process discussed in chapter 6) will provide the opportunity for identifying and agreeing any changes.   The UJAS partners understand “working better together” to mean increasingly using common arrangements to deliver aid and to achieve a more effective division of labor among themselves in supporting specific sectors or programs. To this end, partners will strive to increasingly harmonize programming and policy dialogue, and to rationalize engagement in sectors, choice of aid instruments, and advisory capacity. The UJAS partners will build on progress made in harmonization. They will seek to increase their selectivity in the use of aid modalities, and development partners who are able to provide general and sector budget support will make this their priority.   All development partners are increasingly aware that there is still substantial scope for harmonizing approaches to the provision of budget support. Aligning budget support timetables and the framework for decisions is an important objective for the early stages of the implementation of the UJAS.   UJAS partners are also committed to achieve greater harmonization at the sector level. Currently, progress is most advanced in the health and water sectors, and a clear model has emerged for best practice on the basis of this experience. This involves aligning development partner support (including through projects) with sector priorities agreed with the government, providing assistance predictably, complementing support provided by others (including through the choice of aid modality), and harmonizing development partner assessments with a government-led review. UJAS partners’ aim is to apply this model as relevant in all sectors. A key area where further coordination is needed is support for capacity building.   UJAS partners will increasingly rationalize the program and policy areas in which they are engaged. It is hoped that the government will take the lead in this process of increasing selectivity, although it will also require development partners to critically self assess their comparative advantages. An assessment of comparative advantages will take place in the period leading up to the first UJAS annual review in 2006. Efforts will be made to involve non-UJAS partners in this process. At the time of the review it is intended that UJAS partners will agree on the areas in which each will remain engaged and those in which they will delegate responsibility to others.

Budget support portfolios within UJAS The table (below) is based on UJAS Table 6, and indicates the variety of approaches to aid modalities among UJAS partners. The figures shown (in USD million) are in addition to support through non-government partners. Base Case   Average annual financing to MTEF of which is budget support budget support % Netherlands 38 32 84% Sweden 28 23 82% DFID 113 90 80% Norway 20 12 60% World Bank 294 150 51% AfDB 70 20 29% Germany 32 5 16% Subtotal 595 332 56% Other partners 305 80 26% Total 900 412 46% Source: World Bank, 2005a.

10.46 Danida’s support of the PMA framework is justified, but government’s waning confidence in the PMA must be a source of concern. Further, PMA structures have yet fully to address issues raised by the recent evaluation (Oxford Policy Management, 2005). This recommended 29 actions to address weaknesses in PMA implementation. Of these 9 were identified as priority recommendations of which 5 could benefit from attention from Danida under ASPS2 as cross-cutting issues (Box 10.3). Actions taken by the ASPS may already be tackling some of these (e.g. tailoring its approach to areas that are highly diverse in terms of resources, livelihoods and people; addressing marketing and environmental issues more generally). However, there are other areas in which the PMA may need more help (e.g. with the resolution of apparent design flaws of NAADS, and with pervasive, but neglected, land problems). The PMA evaluation noted that as part of its M&E, there is need for more systematic impact assessment. As explained in Box 9.2, Danida could be well placed to assist with resolving the issues relating to PMA impact assessment by building on its pioneering External District Gendered Poverty Monitoring.

Box 10.3: PMA Evaluation Recommendations Relevant to ASPS2 PMA pillars Differentiated strategies for farmer categories should be developed for each pillar and the NSCG. This should make sure that the particular concerns of women farmers are addressed. As part of this process, NAADS should be encouraged to pay more attention to staple food crops for poorer farmers.   There is a need to address marketing and land issues more generally, not just within the marketing and natural resource pillars, but across all relevant pillars, and NAADS in particular. Environment issues are not being effectively addressed as a cross-cutting issue, particularly by NAADS. This should be addressed as part of a broader review of enterprise provision under NAADS. Policy, regulatory and institutional reform There is a need to inject a sense of urgency into the whole of the land reform agenda, particularly to address women’s poor access to land. Monitoring and evaluation As part of its M&E, there is a need for more systematic impact assessment. This should be linked to the implementation of the PMA M&E system, which should include key impact indicators and periodic case studies, to address more qualitative aspects of PMA implementation and impact. Source: Oxford Policy Management, 2005.

10.47 It would be unfortunate if the “division of labour” exercise became a superficial one. It needs to go beyond a simplistic “pruning” of the number of donors in each sector, and consider explicitly the balance between aid modalities that is appropriate for each sector (Lister et al., 2006). Danida also has an opportunity to influence the analytical work programme of the UJAS partners and should take a special interest in issues of decentralisation, accountability, and capacity building as well as issues relating specifically to the sectors in which it remains directly involved. It also needs to ensure that RDE has adequate staff with economics skills to enable Denmark to participate credibly in discussions of budget support. Working through non-UJAS (or sub-UJAS) groupings of donors will continue to be relevant in pursuing governance concerns that the International Finance Institutions consider to be beyond their remit. And UJAS should be encouraged to develop further its approaches to risk analysis and risk mitigation.⁶⁹

10.48 Progress on harmonisation and alignment also has implications for future evaluations. Joint assistance strategies imply joint monitoring and evaluation, and it makes little sense to evaluate aid separately from the national strategy that it supports. We understand that the Office of the Prime Minister (OPM) envisages a joint evaluation of the PEAP and UJAS in 2008/09. Danida should support this. To ensure quality and maximise the credibility of the evaluation with both Ugandan and donor country stakeholders, it should be conducted by external independent consultants in line with DAC quality standards.

Summary of Recommendations

10.49 The main recommendations of this report are summarised below. All are relevant to Danida. Where they also require action by GOU and/or other donors, this is highlighted.

On Danida’s overall strategy and approach

10.50 Experience in Uganda confirms the value of several principles of Danida’s overall strategy and approach, which should be continued. These include:

• Concentration and focus (see 10.29).
• Consistency and staying engaged (10.30).
• Flexibility within the country programme – avoiding rigid advance programming of expenditure (10.31).
• Capacity development by working with and through local systems (10.32) and recognising the importance of national ownership (10.35). See also 9.35 which highlights a number of factors to be taken into account in Danida’s future capacity development efforts.
• Avoiding the tying of aid (10.33).

10.51 In line with these principles, Danida should take care not to micro-manage the agencies and the programmes that it supports (10.36).

On aid modalities, accountability and collaboration with GOU and other donors

10.52 Danida (at least in Uganda) has operated with a rather narrow paradigm concerning aid modalities. It has allowed its concern for strict accountability of aid funds to frustrate other important principles, including the use of national systems and more collaboration with other donors. It should adopt a less rigid approach in which: 

• More weight is attached to the value of using, and strengthening, national systems of implementation, disbursement and accounting.
• A lower “entry threshold” for budget support is applied, while budget support and other modalities are seen as complementary.
• There is more emphasis on learning from experience with new modalities. (10.34)

10.53 At the same time, Danida should use its participation with other donors and government to push for strengthening of national accountability systems, and also (see next recommendation) to help raise general standards of monitoring and reporting. (10.34)

On maintaining domestic support for aid

10.54 Danida is right to emphasise the importance of maintaining domestic support for the provision of generous levels of aid. The best way to maintain support is to ensure and demonstrate the effectiveness of aid. However, making Danish aid separately visible is less appropriate as more collaborative (and effective) aid modalities are adopted. A more nuanced strategy to maximise the effectiveness of a country programme and to maintain support for it will include:

• Considering the balance between joint and separate interventions.
• Seeking to improve monitoring and reporting on the effectiveness of aid (integrated with monitoring of the national programmes that it supports).
• Seeking to enhance (Danish) public understanding of what constitutes effective aid.
• Seeking to strengthen broad accountability for resources, not just specific accountability for Danish aid funds. (10.33)

On “division of labour” and future priorities for the Danida programme in Uganda

10.55 The “division of labour” exercise should go beyond a go beyond a simplistic “pruning” of the number of donors in each sector, and consider explicitly the balance between aid modalities that is appropriate for each sector. (See 10.47 – this and the other recommendations under this heading require the joint action of GOU and donors.)

10.56 As regards the focus of Danish aid (10.44-10.47, and Box 10.3):

• Denmark’s involvement in the health sector should continue, based on the depth of experience and comparative advantage built up over many years.
• Denmark should take a continuing interest in decentralisation, and seek through UJAS to encourage moves towards greater coordination and a “decentralisation SWAp”.
• HUGGO represents valuable planning and management capacity for the HRD sector, with potential for shared benefits among UJAS partners.
• Denmark should maintain its role in the agriculture sector. There is particular scope for Danish aid to reinforce the PMA in some of the specific areas that the PMA evaluation highlighted as needing attention. These areas include the promotion of extension approaches that are appropriate for reaching poorer households and districts; attention to marketing and land reform issues, with special attention to women’s access to land; more attention to environment as a cross-cutting issue; and more systematic impact assessment (in which there is scope to build on the gendered district poverty profiles supported under the ASPS – see Box 9.2).
• Danida should seek to influence the work programme of the UJAS partners, and should take a special interest in issues of decentralisation, accountability and capacity building (as well as those sectors in which it remains directly involved).
• Danida needs to ensure that the RDE has adequate staff with economics skills to enable Denmark to participate credibly in general budget support discussions.
• Working through non-UJAS (and sub-UJAS) groupings will continue to be relevant in pursuing governance concerns that the IFIs consider to be beyond their remit.

10.57 Both through UJAS and in its internal strategy, Danida should seek to develop more systematic analysis of risks and a more explicit risk mitigation strategy. (10.47, 10.21)

On future evaluation

10.58 Danida should support an external independent evaluation of the PEAP and UJAS in 2008/09. (10.58).

63) However, the Perception Study is helpful, in that many of the judgments offered involve an implicit comparison of Danida’s performance against other donors’.

64) Source (for all the DAC definitions): OECD DAC 2002, Glossary of Key Terms in Evaluation and Results Based Management.

65) Though here in particular there are some question-marks about sustainability and about the cost-effectiveness and replicability of the level of support that was provided.

66) Although there are issues in the coherence between Denmark’s aid policies and some of the other policies it supports. The EU agricultural subsidies from which Danish farmers benefit are at cross-purposes with Danida’s efforts to support agricultural development.

67) It is ironic, given the negative reactions of some Danish NGOs, that the airport was the more relevant and effective intervention in itself.

68) The short-term allocation of development funds essentially motivated by Danish domestic interests should be avoided wherever possible. This type of activity is often not sustainable, does not fit with Danish principles of partnership, and can be disruptive to carefully laid out country strategy. (OECD Development Assistance Committee, 2003a).

69) See World Bank, 2005a Table 4 on p29.