Department For International Development
01 This report is an evaluation of the Department for International Development’s (DFID) country programme in Afghanistan from January 2002 to December 2007, commissioned by the Evaluation Department of DFID (EvD) and undertaken by ITAD Limited, an independent consultancy company. The team spent two weeks in London interviewing key stakeholders in Whitehall, and then undertook fieldwork in Kabul and Lashkar Gar (Helmand Province) during April 2008, with these locations reflecting the focus of DFID’s Afghanistan programme. This report was then prepared and subject to review and comment by EvD and DFID Afghanistan.
02 The evaluation period is from 2002-07. It covers (i) the final year of humanitarian assistance led by DFID’s Conflict and Humanitarian Affairs Department (CHAD), (ii) the Transitional Country Assistance Plan (TCAP) and the (iii) 2005-06 Interim Strategy for Afghanistan which was subsequently extended to 2008.
03 Afghanistan is a uniquely complex country, socially and politically. The evaluation notes the constraints and pressure that DFID staff are under and the high levels of commitment demonstrated throughout the history of the programme. Where the evaluation identifies problems, these should in no way reflect poorly on the professionalism of those involved; rather, it is an opportunity retrospectively to learn from the strengths and weaknesses of DFID’s approach to a volatile and intense environment that offers very few easy answers.
Political and development context
04 Afghanistan is a “fragile state”, one of the poorest countries in the world, and is off-track on progress towards all the Millennium Development Goals. Since the foreign intervention to replace the Taliban in 2001, there have been some marked improvements in health and education, and 4.8 million refugees have returned to their homes (though just as many have migrated to Iran and Pakistan). By 2008 protracted conflict in many areas, global price rises in wheat (a staple food) and periodic drought meant that periods of extended emergency could be anticipated.
05 Economic growth has settled at around 10% per annum and foreign debts have been cancelled. Domestic revenues are only about 8% of Gross Domestic Product (GDP). By contrast, in 2007 illicit opium comprised 47% of GDP. The poppy sector constitutes the largest source of export earnings; its illegal nature not only implies significant lost tax revenue to government but also breeds lawlessness which undermines economic confidence and social cohesion.
06 There is visible economic activity and improvement of infrastructure in many parts of the country, yet persistent unemployment and high levels of corruption underline the need for international commitment to long-term, balanced development and good governance in Afghanistan for many years to come.
07 Following the defeat of the Taliban government, the December 2001 Bonn Agreement saw a collective determination by donors to lend maximum support to the new interim government. Within three years there were successful presidential and parliamentary elections – supported by DFID – with a high turnout. Political stability, however, remains precarious. By originating in a foreign military intervention, the state-building effort in post-Taliban Afghanistan became perceived by many Afghans as tied to the power of foreign troops and capital. This has challenged the legitimacy of the state and lent credence to the propaganda of the insurgency. Levels of insurgency increased from 2006 onwards.
08 Total annual Overseas Development Assistance (ODA) to Afghanistan more than doubled during the period 2002-2006, from $1.3 billion to just under $3 billion. This brings the total since 2002 to almost $11 billion. Only about one-third of development expenditures are controlled by the Government, with most of USA funds being off-budget.
09 Afghanistan receives a low proportion of aid per capita; due to its poor absorptive capacity, by 2006 the country had spent only 23% of what it had been granted for the development budget since 2001. Moreover, levels of aid effectiveness have drawn some criticism. Recent evidence suggests that for every $100 spent; only $20 actually reaches Afghan recipients. Between 15 – 30% of aid money is spent on security for aid agencies, and 85% of products, services and human resources used by agencies are imported and provide few jobs for Afghan workers.
10 The 2006 London Conference launched the Afghanistan Compact; this and the Interim Afghanistan National Development Strategy (I-ANDS) lays out the framework for international engagement with Afghanistan until 2011 in a number of vital areas: security, governance (including human rights and rule of law), social and economic development and cross cutting themes such as counter-narcotics, gender equity, and anti-corruption. DFID has supported the ANDS process – essentially a Poverty Reduction Strategy Paper (PRSP) for the country – through to its final presentation in 2008.
The UK and DFID in Afghanistan
11 The UK is one of 36 nations involved in Afghanistan since 2001. Its efforts are coordinated by the Foreign and Commonwealth Office (FCO), DFID and the Ministry of Defence (MOD). Aside from a large UK military expenditure, DFID has the biggest civilian expenditure. On conflict prevention, resolution and dialogue all three departments have, until 2008, accessed a common fund, the Global Conflict Prevention Pool (GCPP). The FCO has an additional portfolio through the Global Opportunity Fund (GOF). The MOD and FCO support security sector reform, and the Counter-Narcotics strategy is led by the Afghan Drugs Inter-Departmental Unit (ADIDU).
12 The UK is the third largest development aid donor to Afghanistan, maintaining about 10% of total ODA since 2003, with just over £100m per year since 2004/05. The majority of funding has been through central government, notably the Afghan Reconstruction Trust Fund (ARTF) which by early 2008 accounted for 47% of DFID’s Afghanistan budget.
13 DFID’s assistance since setting up the office in Afghanistan can be traced through four phases:
· 2001-2003 Essentially a limited humanitarian effort – funds through UN, International Organisation for Migration (IOM), Red Cross and NGOs – with longer term commitments foreseen. At the Tokyo donor conference in 2002 DFID pledged £200m over five years (02/03 – 07/08), part of a wider $4.5billion pledge made by the international community.
· 2003-2005 The Transitional Country Assistance Programme (TCAP) designed as a relatively short-term plan with a long-term strategy, to cover the duration of the Transitional Administration in Afghanistan. It was informed by a Government/international agency report “Securing Afghanistan’s future: Accomplishments and the Strategic Path Forward”, prepared for the International Conference in Berlin, March 2004. Based on a World Bank assessment that Afghanistan needed $28 billion over seven years, the UK increased the pledge made at Tokyo to at least £500m over the same five-year period.
· 2005-2006 Following national elections, the Interim Strategy for Afghanistan 2005/06, a three-pillar programme – state building, economic management and aid effectiveness, and livelihoods. This built on Government ownership and capacity, with upwards of 70% of DFID funds being channelled through the national budget and national programmes. The ANDS (essentially equivalent to a poverty reduction strategy, though also covering security) was under preparation, and understood by DFID to be the ideal mechanism to provide a necessary strategic framework for the long term.
· 2006-2008 The fourth phase of DFID’s development strategy was marked by its commitment to the 10-year Development Partnership Arrangement (DPA), under which it announced its commitment to spending £500m over three years. This phase was characterised by the continuation of the 2005-06 Interim Strategy for Afghanistan coupled with emphasis on supporting the UK’s increasing focus on Helmand province.
14 By 2007, there were 58 projects in the DFID Afghanistan portfolio with an aggregate value of about £520m. The programme was dominated by the World Bank managed Afghanistan Reconstruction Trust Fund (ARTF) – classified as a DFID ‘project’ – with DFID’s aggregate contribution towards it totalling £317 million funded through the central Treasury Account in Afghanistan. Without the ARTF, other ‘live’ programmes had a total value of £201 million, 11 of which were valued at over £5 million.
Relevance of DFID’s programme
15 The TCAP – and, indeed, the subsequent 2005/06 Interim Strategy – was predicated upon several assumptions: first, that the formal political transition process would result in a stable political settlement; second, that conferring legitimacy on the state means building it from the centre first – with a particular emphasis on economic management – then extending to provincial and local levels; and third, that formal institutions (judicial, legislative, banking, private sector) are the pillars of growth.
16 DFID placed a strong emphasis from the outset on management of the economy. The aim from the start was to create a strong public finance system in order to implement the National Development Framework (NDF) and enable the Government of Afghanistan to lead the co-ordination of development activities. The quality of technical assistance (TA) has been high, but there are drawbacks in terms of scope and sustainable results. For example, the ‘standard package’ assistance to the Ministry of Finance has failed to convert ARTF procedures into a regularised budget formulation process at the speed originally envisaged (the closure on the ARTF has been extended twice, from 2006 to 2010 and recently to 2020).
17 DFID’s assistance to revenue-raising through taxation has been effective and efficient, but DFID did not articulate the strategic challenge of how to foster a ‘social contract’, a tangible demonstration of representative democracy. If not balanced with clear evidence of benefits elsewhere in the system, taxation alone risks reinforcing public opinions of a predatory state.
18 DFID’s state-building strategy has had a strong focus on technical assistance and capacity development of formal state institutions, particularly in the executive branch of government. Drawing on its long-standing experience in public administration reform (PAR), DFID has helped achieve initial results in the areas of civil service reform. Until recently, relatively little attention was given to developing a political economy and conflict analysis to underpin the strategic choices made. DFID has given little attention to accountability issues and the demand side of governance, including the monitoring and advocacy role of civil society and other accountability mechanisms.
19 Prior to late 2007, the conflict agenda was pursued primarily through Global Conflict Prevention Pool (GCPP) which had a global budget of approximately £245 million between 2004/05 and 2006/07. Some 46% of this was allocated to Iraq and Afghanistan, and expenditure on these two countries has dwarfed all other GCPP outlays. In Afghanistan, the share of Security Sector Reform (SSR) expenditure within overall investments have decreased from about £12m in 2004 to almost zero in 2008. This is regrettable in a country that is in the midst of building a new security sector and is facing multiple challenges relating to SSR. The security sector is the one element of the Government’s ANDS for which a sectoral strategy is still lacking.
20 In the country as a whole, the HMG division of labour, and in particular the FCO lead on rule of law and justice issues, led to a marginalisation of DFID’s role on governance, an area in which it has a comparative advantage and experience from other fragile states. Five years on, DFID is now giving renewed attention to the rule of law and justice sector as well as to sub-national governance – an important evolution of its approach to state-building since these are major impediments to effective state building in Afghanistan.
21 The objectives of DFID’s livelihoods programme are twofold. Firstly, it is to maximise opportunities for rural Afghans to make a legal living; and secondly, through the stability that derives from increased income and employment, to increase confidence in the state. This means tackling key constraints such as access to markets and credit, labour opportunities, and training. From 2006, DFID moved from area-based poverty focused programming through non-governmental organisations (NGOs) to increased funding of central government programmes, and a consequent change in approach to pro-poor interventions. There was now an emphasis on funding TA and capacity building within key ministries such as the Ministry of Rural Rehabilitation and Development (MRRD) and the Ministry of Agriculture, Irrigation and Livestock (MAIL) through, for example, the Support for Strategic Planning for Sustainable Rural Livelihoods (SSPSRL) project. Support here was strategically appropriate, given the importance of the agricultural sector in Afghanistan. Likewise, the National Solidarity Programme (NSP), National Emergency Employment Programme (NEEP) and Microfinance Investment Support Facility for Afghanistan (MISFA) have direct links to poverty alleviation.
22 .But more upstream programming has consequences, not least the fact that the GoA’s capacity at national level is limited. Aside from the issue of geographic focus, DFID had a well balanced rural sector programme in support of the GoA that included capacity building of ministries plus a good mix of development, infrastructure, poverty reduction and CN related projects. However, cutting the Badakhshan programme prevented further learning from an interesting initiative.
23 DFID has a well-established reputation with respect to donor harmonisation and in upholding Paris Declaration principles with Government and partners. The early engagement through the ARTF was directly relevant to needs expressed by the Interim Government at that time, and DFID’s flexibility and responsiveness were appreciated by Government and other donors alike. DFID has conformed well to some of the central tenets of its fragile states policy including long-term engagement, support to donor co-ordination and the use of innovative aid instruments, although less well in areas such as prioritising governance reforms that will address the causes of fragility.
24 The balance of risks within DFID’s Afghanistan portfolio – as assessed in project memoranda and evaluated regularly in output to purpose reviews – has moved from low-medium risk in 2004/5 to medium-high risk in 2006/7. This risk inflation has been driven by two factors: firstly, the security situation in Afghanistan; and secondly, the difficulty associated with delivering a well-performing programme given current government capacity constraints in Afghanistan. Programme risk has also been heightened by s well assis has been weakons (of state building(GOF)the political imperative to engage in ‘difficult’ areas. By mid-2007, eight of DFID Afghanistan’s 31 live programmes were high risk, with a total value of £110 million, representing 25% of the live portfolio.
25 Where risk analysis has been weak, this has mainly been due to a lack of robust DFID methodology on differentiation of risk types (macro, fiduciary, capacity, conflict); this impedes accurate assessment. Moreover, there has sometimes been a tendency to express existing weaknesses within the Afghanistan government as programme threatening risks within programme design – and later to express this again as a reason why the programme failed to meet expectations. This avoids the needs to identify specific mitigation strategies that should be built into the programme to ensure the risk is dealt with.
26 The UK’s 10-year Development Partnership Arrangement (DPA) with Afghanistan, signed in 2006, committed DFID to furthering its on-budget support (through the ANDS) with the intention of moving towards Poverty Reduction Budget Support (PRBS) on condition that the appropriate administrative, technical and financial systems could be established to support this. Given the fragmented and uncoordinated nature of international aid to Afghanistan, the declaration was both timely and appropriate. However, DFID and the Afghan Government do not have a system in place for mutual review and assessment of the commitments (initially £330m over 3 years), although this was specified in the agreement. With 80% of DFID’s funding being channelled to the Afghan Government, this would seem an important element of risk management, as well as being required by the UK’s conditionality policy (2005).
The UK and DFID in Helmand
27 Militarily, the international community moved from peacekeeping to counter-insurgency in 2006; it was at this time that the decision for the UK military to deploy to Helmand was taken. This shift in the political focus of Her Majesty’s Government (HMG) towards countering the growing insurgency made it difficult for DFID to focus concertedly on reconstruction and development issues in a conflict environment – issues that require longer time-frames, specialised expertise and sophisticated forms of interaction with target beneficiaries.
28 DFID is committed to a “whole of government” approach, one of the DAC Principles. Maintaining the right balance between this commitment and putting into practice a strategy and portfolio has proved to be a significant challenge. DFID was continually under pressure to demonstrate that it was contributing to the Helmand effort both through its contributions to tri-departmental mechanisms, and through its bi-lateral programme. To some extent this skewed its priorities and choices, and undermined the coherence of DFID’s overall strategy.
29 HMG as a whole has not had a shared vision over the link between security and development and the means to achieve these goals. Pursuing multiple objectives in Helmand is problematic because approaches towards counter-insurgency, stabilisation, counter-narcotics, peace and development have not necessarily been mutually reinforcing. The problem is complicated by the presence of two distinct foreign military forces operating in the country, one (the Coalition Forces (CF) under Operation Enduring Freedom) on a war-footing but occasionally involving itself in humanitarian, reconstruction, political reform, information gathering, psychological operations and special operations, and the other (NATO/ISAF) under UN mandate. Both forces are in uniform and are, irrespective of their functions/affiliations/mandates, indistinguishable to the public, with the image portrayed by one, inevitably affecting the acceptance of the other.
30 The role of DFID within the Provincial Reconstruction Team (PRT) in Helmand raises issues over ‘humanitarian space’, and priorities within the civil-military nexus. The mission of PRTs is “to assist the Government of Afghanistan to extend its authority, in order to facilitate the development of a stable and secure environment in the identified area of operations, and through military presence, enable SSR [security sector reform] and the reconstruction efforts.” The evaluation is concerned that there was no “do no harm” analysis to consider whether PRTs, counter-insurgency and counter-narcotics activities had the potential to undermine rather than reinforce longer-term state-building and state legitimacy. For example, recent research notes that “although arguably necessary in some highly insecure areas, by diverting resources which otherwise could have been devoted to civilian development activities, PRTs have in many cases undermined the emergence of effective institutions of national and local government, and other civil development processes.”
31 DFID’s perception of Quick Impact Projects (QIPs) in Helmand as being “rapidly implemented projects that serve as down payments on promises of political and economic progress…” was challenged recently by an independent evaluation commissioned by the Stabilisation Unit. The report speaks of less than ideal relationships between the PRT and Government in Helmand. In the eyes of local people little distinction was made between UK forces and those of the Taliban, even if tactics may differ; the question was simply a matter of who would provide lasting security. Within the work carried out by the PRT (the QIPs in particular), there was dissatisfaction over procurement, construction quality, lack of monitoring and over the role of interpreters. On the positive side, support for police training was well received and that DFID support through line ministries (particularly MRRD) was seen as being effective.
32 Much effort has been put into working relationships between the MOD, FCO and DFID in Kabul and Helmand and these were generally reported as positive and constructive. However, the evaluation notes a marked divergence of opinion among Whitehall departments over the scale and appropriateness of DFID’s initial engagement in Helmand. DFID staff point to the degree to which the Cabinet Office drove the agenda – and to a large extent determined the various departmental roles within that agenda – by creating sometimes unrealistic time-bound targets to be met by each department. This was the case for both the counter-insurgency (COIN) and counter-narcotics (CN) strategies.
33 Conversely, the FCO criticised DFID for failing to ensure sufficiently senior staff were present at Whitehall meetings. The evaluation found this to be the case, particularly during the crucial meetings in 2005-06 over the deployment and scope of work expected of UK departments in Helmand. The evaluation is unable to comment on whether a more regular senior DFID representation would have influenced what was essentially a pre-ordained decision over the deployment.
34 Support through the cross-Whitehall GCPP (and subsequently the Stabilisation Aid Fund) provided a funding mechanism that could be used for security spending, including non-ODA spend. In this respect, it protected DFID’s bi-lateral programme budget from the wider security demands of HMG while opening possibilities for innovative, pro-development, approaches to stabilisation. From 2007, pressure on the DFID bilateral programme to divert funds to Helmand started to reduce due to the increased role proposed for SU.
35 The evaluation also supports the argument that delivering assistance in ways that will have a sustainable impact on the lives of Afghans requires developmental approaches which PRTs are ill-equipped to cater for. The experience of DFID’s development advisors deployed in PRTs prior to Helmand suggests that PRTs should remain e ssentially military vehicles. Only when independent government and/or NGO capacity and access is assured should DFID become involved as a bilateral partner through conventional channels. Meanwhile, PRTs should be staffed and funded solely through the inter-departmental Stabilisation Unit and SAF.
Effectiveness and Efficiency of DFID’s support
36 Staffing of the DFID office in Kabul was limited by the ‘light footprint’ approach agreed by DFID’s top management. This effectively imposed a ‘cap’ on international staff levels (6) prior to 2006 which meant that the ratio of staff between Kabul and London was 6:9. This was deemed wholly unrealistic by Heads of Office at that time, given DFID’s delivery objectives. Since then staff ratios have improved, but staff pressure to deliver on HMG priorities has been immense. Security constraints have limited engagement with local and provincial government restricting data gathering and personal knowledge of projects.
37 DFID has ensured that thorough independent reviews are carried out for all the large projects in its portfolio, particularly where there have been indications of poor performance. Impact assessment has been difficult, partly due to the weaknesses in project-level results frameworks, but also due to the inherent difficulties of measuring impact in an insecure environment. For example, DFID activities towards developing poverty data sets and strengthening the national statistics function have to date rendered few results. .Added to this is the familiar difficulty of measuring impact of capacity-building and institutional reform.
38 DFID’s performance management is based on the scores given to projects at purpose and output levels. From 2001-mid 2006, 74% of DFID Afghanistan projects over £1 million were scored ‘completely’ or ‘largely’ successful. This compares well with the DFID average of 61.8% across all fragile states in the same time period. Given the dominance of a handful of projects (notably the ARTF), however, these figures should be treated with caution. If one deducts the ARTF, in 2006/07 48% of the total size/value of the ‘live’ (i.e. ongoing) projects were unlikely to achieve set targets.
39 Not surprisingly, smaller projects performed better than the larger more complex Government-run National Solidarity Programme (NSP), National Emergency Employment Programme (NEEP) and the National Rural Access Programme (NRAP), but even these have shown solid improvement over the past 12-24 months.
40 Projects under economic management have performed relatively well, notably DFID’s timely and effective inputs into the ARTF as the central mechanism for support to the Government. The ARTF investment window includes activities in three areas: infrastructure, public sector capacity building and community development. Under the last of these – the sector that has received a majority of funding – are national rural development/service delivery projects such as the National Solidarity Programme (NSP) and Microfinance Investment Support Facility for Afghanistan (MISFA), as well as the provincial Helmand Agriculture and Rural Development Programme (HARDP). These projects would not have been able to get off the ground without DFID’s support. Combining core project funding with the use of TA has enabled DFID to maintain flexibility while also providing information to support and influence multilateral agencies.
41 Within the state-building portfolio the consequences of under-staffing have been apparent. Only 25% of projects achieved high score rates (scores of 1 or 2 in output to purpose reviews) from 2002-06. This was skewed by the Afghanistan Stabilisation Project (ASP) and Strengthening Counter Narcotics in Afghanistan Project (SCNIAP), the two largest but also worst performing programmes. As part of the ASP, the performance of the Provincial Stabilisation Fund was poor, with only a few projects completed and no progress at all in the area of administrative reform. DFID took appropriate steps to end its support to both ASP and SCNIAP.
42 Different choices within given resources could have been made. For example, greater priority could have been given earlier to security and justice work, given that the rule of law sector was consistently highlighted as one of the most critical areas for addressing state fragility, and where progress was sorely lacking. This should arguably have been given priority over support to central government institutions, or to counter narcotics institution-building through SCNIAP, where progress depends on wider reform in the justice sector.
43 Some cross-cutting issues have received relatively little attention in the DFID portfolio. Despite some impressive outputs in the above projects – and notwithstanding the strong gender focus in DFID’s support for elections – the overall integration of gender, human rights and social exclusion issues has been weak throughout the portfolio. There is no mention of gender issues or women’s rights in the TCAP or Interim Strategy, despite the Government’s own stated commitments in this area., DFID has, however, made efforts to integrate cross cutting issues into the ANDS by placing a staff member in the ANDS secretariat since 2006 with this specific brief, though there is little evidence of tangible results.
44 Efforts to institutionalise counter-narcotics as a cross cutting issue within government have also had little success. The Ministry of Counter Narcotics has no executive power (or money), only advisory responsibilities. It has suffered from the absence of a Minister for a period of a year and then the appointment of weak ministers. DFID has demonstrated strong analysis in the sector, but progress on the Counter -Narcotics Trust Fund (CNTF), for example, has had little government buy-in.
Impact and Sustainability
45 DFID is keenly aware of the difficulties of assessing and demonstrating impact in the Afghan context. The lack of good national or provincial data and security constraints on access to beneficiaries (for both DFID staff and partners) impedes the measurement of progress or decision making. DFID’s practice of putting its aid funds through common systems adds to the usual problems of attribution in development aid. Evidence of the results of specific inputs in a multi-funded project is therefore often related more to aid effectiveness than to wider developmental impacts. Moreover, in all programmes, security constraints prevent staff monitoring either outputs or impact in any consistent manner.
46 Progress has been made in building up a more professional, merit-based civil service with DFID support. For example, tangible results can be seen from DFID’s support to co-ordination, strengthening the Ministry of Finance and helping create an effective public administration while ensuring all assistance is recorded on budget. Early financial and capacity support to ARTF has been exemplary, demonstrating good practice in relation to the Paris Declaration and principles of aid effectiveness in fragile states.
47 The impact of the ARTF on the government’s legitimacy and ability to deliver, particularly outside Kabul, remains open to question. Capacity in the provinces to deliver services has been largely neglected and budget execution has been highly variable. Whereas the Ministry for Reconstruction and Rural Development (MRRD) spent 71% of its development budget in 2005-2006, the Ministry of Interior spent only 32%, education 24%, and agriculture 22%.
48 DFID nevertheless has an extremely good track record on aid effectiveness issues in Afghanistan. Pledges have translated to disbursements very quickly, and DFID has provided procurement capacity to facilitate rapid and transparent utilisation of donor resources. Rapid, high quality, early TA to the telecoms sector contributed to high levels of investment and mobile phone usage.
49 Corruption remains a fundamental challenge. DFID has contributed incremental improvements through, for example, the Tax Administration Reform Project, but the problem is more profound. Justice institutions remain the least developed among formal oversight organisations.
50 The state-building portfolio may have focussed too much on building technical capacity, primarily in Kabul, while downplaying issues of political legitimacy, especially at the local level. DFID’s crucial support to the elections in 2004-05 contributed to the political transition process that was necessary to establish the core institutions of the state. However, sustainable impact is impaired by the inability of the Government to establish national unity linked to political settlement; this cannot be addressed by the kind of technical and financial support provided by DFID to date.
51 Public administration reform has been bolstered through the DFID-supported Second Public Administration Programme (SEPAP) project, and DFID has added significant technical value throughout. However, progress towards the objective of improving Government capacity at central and local levels with strengthened links between them was limited. Improving Government leadership on PAR with in