i. This evaluation is not primarily focused on accountability, but on learning. It seeks to assist DFID to improve its response to the Syria crisis and how it responds to future challenges. The following summarises the findings and provides key recommendations.
ii. DFID’s strategic approach to the humanitarian crisis in Syria was consistent with the UK’s international obligations and own policy priorities. It appears to have achieved its overall purpose and DFID has demonstrated considerable flexibility in practice (albeit this is not reflected often in its formal project documentation). DFID has worked well across Whitehall while sometimes having to fight to assert its humanitarian principles. DFID’s approach was not fully aligned with the UK Government’s Humanitarian Policy 2011, however (see Section Three for more detail). We note that DFID initially assumed that the crisis would follow a particular trajectory (towards regime change in Syria) and did not fully incorporate alternative outcomes into its planning.
iii. Grant allocation has broadly aligned with DFID’s geographical, sectoral and cross-cutting priorities and allocations explicitly address the needs of some particularly vulnerable groups.
Needs assessments, political priorities and opportunity for action all have had an influence on funding. DFID has now, appropriately, moved towards a model of longer term funding. DFID is increasingly building resilience into its programming but, appropriately, relief operations still dominate. Grant management has, however, emphasized the reported effectiveness of delivery more than the actual and verified effectiveness, economy and efficiency.
iv. The SCU has improved how it seeks to define and monitor results. To the extent possible, the results of funded programmes aggregate into the overarching logical framework, which itself could be significantly improved. A Theory of Change (ToC) exists but appears to have no obvious use in design or monitoring and evaluation, or as an early warning tool. Programmes often actively and innovatively engage with beneficiary communities, but this is not something that DFID does directly. Beneficiary involvement in reporting is improving. DFID’s choice to hold partners to account for specifically DFID-attributable results only works, however, if DFID subsequently monitors progress in a manner that gives some assurance of the quality of that reporting. Although DFID staff conduct field monitoring on a limited basis, to date it has had no sure mechanism to verify its partner’s results.
v. DFID has been able to deliver in spite of considerable staffing constraints. DFID has, however, been fragmented and the SCU has not sufficiently acted as a single unit. The current model, using a mix of contracted and permanent staff, as well as relying on staff with insufficient experience in DFID, needs improvement. Similarly, the operation and organizational structure in MENAD (with the current split of responsibilities between the two deputy directors) should be reviewed. DFID is not fully acting, or presenting itself, as ‘one DFID’, or as a single team.
vi. Staffing and programme management are too focused on London. DFID is not making decisions as close to the point of delivery as possible. Communication between the centre and periphery has not been good enough, with a consistent message of a dominance of top-down messaging rather than two-way communication. Programme staff do not spend enough time in country or meeting representatives of agencies that are responsible for grant implementation.
The physical and operational distance between technical and programme staff is too large.
vii. DFID did not have a pre-existing model of how to operate or resource a response to a challenge like the Syria crisis. As a result it made some poor initial decisions. DFID did not plan for the worst case or initially resource its management of the response sufficiently well. It is highly possible that the response to the crisis in the region provides an indication of what much of DFID’s work will look like in the future.