2DFID’s strategy
DFID’s delivery model
6.As with any government department, the Department for International Development (DFID) has to make choices about how to spend its budget. DFID has a broad set of goals outlined in the UK Aid Strategy (and underpinned by the Sustainable Development Goals (SDGs)) that collectively contribute towards its overall objective of reducing global poverty.8 However, DFID is not a direct implementer of the development programmes that it funds to achieve these goals. As the Department stated in written evidence “as a government department DFID is structured as a commissioning organisation.”9 DFID thus has to work through partners to achieve its objectives.
7.With regards to DFID partnerships, Dr James Morton, an experienced development practitioner, wrote in evidence:
“The ideal would be for DFID to work directly with the governments and civil society of the countries it is trying to help. However, capacity constraints and fiduciary concerns mean the Department must normally depend on an external agency of some kind to manage its programmes.”10
Poverty is driven by a complex set of challenges. Limited specialist skills to tackle these challenges coupled with the risk of corruption in many developing country governments means that DFID’s delivery model involves operating in partnerships with a variety of external actors.
8.There are three main channels through which the department delivers its bilateral programmes:
- Multilateral agencies—International organisations formed between a number of countries to work on issues that relate to members’ common priorities. The UK works alongside other donors within these organisations and they include UN agencies—such as the World Food Programme (WFP)—and the World Bank Group, which offers financial and technical assistance to developing countries around the world.
- Civil Society Organisations (CSOs)—The term CSO is broad and includes non-governmental organisations (NGOs), charities, trusts, foundations and advocacy groups. They typically aim to address social or political issues such as human rights, the environment or health. Some well-known examples that DFID partners with are Oxfam, CARE and Save the Children.
- Contractors—For the purpose of our investigation, ‘contractors’ refers to private sector organisations that play a role in advising on and/or implementing DFID programmes. DFID uses many agents and intermediaries (who themselves use agents and intermediaries), thus the World Bank, WFP and Oxfam may all be considered DFID contractors. However, for the purpose of this report it is private sector organisations that the term refers to.
It should also be noted that there is considerable diversity within this classification of ‘contractors’. For example, “Options”, one of DFID’s ‘Key Suppliers’, is a subsidiary of a charity (Marie Stopes International (MSI)). Therefore, unlike other contractors, it pays no dividends to shareholders and all surplus funds are given to MSI as Gift Aid.11
Box 1: Bilateral, multilateral and bilateral through multilateral aid delivery
|
UK official development assistance (ODA) spending (by DFID and other government departments) is split into two core streams: bilateral (all aid provided by donor countries when the recipient country, sector or project is known) and multilateral (delivered in the form of core contributions to multilateral organisations, including the European Union). These streams were valued at £7.7 billion and £4.5 billion respectively in 2015. However, these figures do not reflect the full amount channelled through multilaterals, as sometimes multilateral organisations are selected as the delivery partner for bilateral programmes. £2.2 billion of UK bilateral ODA was spent in this way in 2015. |
Figure 2: DFID spending through various channels
Note: Total DFID spending corresponds to 2015/16 but due to availability of statistics, the bilateral breakdown is drawn from figures from 2014/15, 2015 and 2015/16
Source: DFID, Statistics on International Development 2016 (Table 10), written evidence (CON0039) and Annual Report and Accounts 2015–16
9.DFID spent £1.24 billion through contractors in 2015/16, which comprises 20% of bilateral spend and 13% of DFID total spend in that year. While this represents an increase in recent years (see figure 1 on page 4), it is low relative to other donors. 29% of funds awarded by the US in 2016 were through contractors (US$4.7 billion),12 while the share of ‘official support’ from Australia’s Department of Foreign Affairs and Trade (DFAT) delivered through the private sector in 2014/15 was 24% of total spend (AUS$1.4 billion).13 Within the UK Government, DFID’s use of contractors is also relatively low. In a 2013 report, the National Audit Office (NAO) stated that central government departments spent £40 billion through third parties in 2012–13. The NAO estimates that DFID accounted for less than 2% of this (making it the eleventh largest department in contractor spending) while the Ministry of Defence accounted for around half of all government spending via contractors.14
Selecting a delivery channel
10.DFID’s Smart Rules, the operating framework for DFID’s programmes, set out the key role of the business case in determining the channel through which a given programme is to be delivered. The business case describes the need for a programme, what it will do and in what timeframe, sets out expected achievements and serves as a report to the public on how taxpayers’ funds are being used. It includes an appraisal case, which should explore the variety of options for how DFID will address the strategic need for the programme in a way that optimises value for money.15 ICAI raised concerns about the way in which business cases are used for appraisal in its 2015 report ‘DFID’s approach to delivering impact’. It wrote “They appear to set out a justification for design choices that have already been made, rather than offer a robust appraisal of options.”16 This was supported by oral evidence from Joy Hutcheon, DFID Director General, Finance and Corporate Performance, who told us:
“We would like to see a strengthening of the appraisal case. We think sometimes teams rush a bit to the option. [ … ] It is a robust process, but there are improvements we can make to it.”17
11.With respect to appraising delivery options, Dr James Morton also felt that: “It is not clear that DFID has a clear framework to guide this central strategic choice.”18 This reaffirms past ICAI findings. In its 2013 report on DFID’s use of contractors, ICAI stated that DFID “lacks strategic guidance [ … ] on how to use contractors to best effect, including the circumstances in which contractors should be used and which contractor skills are most valuable.”19 While DFID responded referring to a Departmental review of strategic choices in 2013, evidence suggests that concerns remain. As ICAI reported, DFID does not have “an overall process for assessing how the use of contractors compares with other channels for delivering aid.”20
DFID staffing
12.One of the stated advantages of DFID’s use of contractors is that it allows the Department to engage in a wide range of contexts and types of intervention that it would not have the capacity to deliver through in-house staff. In written evidence, ASI cited a report that estimates that if DFID were directly to implement its programmes, it would have to recruit an additional 25,000 to 30,000 personnel.21 According to Crown Agents, outsourcing services to contractors “allows DFID to have flexibility and speed to its response, access to wide reaching technical experts, drive more targeted VFM [value for money] and support services tailored directly to the local context and DFID strategic objectives than compared to other programme delivery channels.”22
13.However, there are also concerns that DFID’s approach to using contractors to deliver aid programmes is a necessity owing to the growing aid budget rather than a strategic choice. As Bond stated in written evidence:
“The decisions about when and where DFID uses contractors are internally driven, and not much is known about the decision-making process. It is widely recognised that because the increased aid budgets—which have, overall, enabled DFID to increase its impact in tackling global poverty and inequality—have not been matched by increased staffing levels with DFID, a greater use of contractors has become inevitable. It has also resulted in more ‘light touch’ management, with little time for field visits, etc.”23
14.The decision processes that determine DFID’s use of contractors are a long-standing concern. In a 2011 report, our predecessor Committee considered DFID staffing and use of contractors and concluded “the use of external suppliers should be determined by assessing the best way to achieve outcomes, not by external constraints on administrative costs or staffing levels.”24 In our recent report ‘UK aid: allocation of resources’ we note that “DFID’s administrative capacity appears to have fallen below what is required to spend its increasing budget effectively.”25
15.Figure 3 below shows the growth in the aid budget in proportion to the growth in DFID staff numbers since 1998, the year after the Department was founded. The pressures on the Department underlying this trend were clearly stated in written evidence from Crown Agents “DFID is expected to achieve more, despite having less staff.”26
Figure 3: Index of growth in DFID staff and Net ODA Budget, 1998=100
Source: DFID, International Development Statistics 2016 and UK Civil Service Statistics
16.In oral evidence, Joy Hutcheon reiterated DFID’s role as a commissioning organisation in the context of staffing, saying “we are not equipped, in terms of staff members, to be a delivery organisation; we are a commissioning organisation [ … ]”.27 However, even in taking such a role, there is a requisite level of resources and technical expertise to effectively oversee delivery. As Palladium wrote in evidence “for partnership to be successful, DFID staff must retain both the capacity and the ability to oversee the delivery of these programmes and to work in partnership with their supply base across a range of technical and operational issues.”28 International development company and contractor DAI added to this, stating:
“DAI would welcome greater investment by DFID in its technical depth. It is important for DFID’s implementing partners to have a technically capable partner in the field. A “hollowed out” DFID would be a less effective department: less able to achieve its development mission, less able to design, award, and monitor development programming.”
17.We recognise that in DFID’s role as a commissioning organisation, there are numerous advantages associated with contracting outside expertise. However, despite the value that contractors can bring, there are certain functions that in-house staff must serve and which cannot be outsourced. With both the size and number of contracts growing, there are increasing pressures on staff who, even with the assistance of commercial advisers, may not be equipped with the necessary skills and expertise to manage the technical, operational and commercial aspects of such large and complex programmes. Replacing development expertise with commercial expertise or overburdening staff is likely to have implications for the effectiveness and impact of programming. Evidence suggests that the decision-making process at the core of programme design—which channel to use—needs to be strengthened.
18.We urge the Department to ensure that the choice of implementing a given programme through contractors is always driven by the strategic value of that delivery channel. DFID Smart Rules should include more detailed guidance on the appraisal portion of the business case, and business cases should clearly justify why the particular channel was chosen, with reference to the relative benefits over other types of delivery. DFID should ensure that the emphasis on commercial expertise associated with the increasing use of contractors does not make generalists of DFID staff at the expense of technical knowledge and depth.
19.Questions were raised about DFID’s use of contractors by our predecessor Committee in an inquiry into Recovery and Development in Sierra Leone and Liberia. In its 2014 report, the Committee recommended that DFID “investigates the potential to set up an arm’s-length wholly or partially-owned consultancy that can either challenge these other providers and help to drive down costs or provide a greater pool of expertise.”29 Evidence to that inquiry suggested that some contractors have “limited experience and understanding of social programming”.30 The idea of an arms-length body was also raised by Contractor B in evidence to this inquiry.31 In its response to the 2014 report, DFID ‘partially agreed’ with this recommendation, and added that it has “established a commercial capability reform programme, building commercial capability within DFID.”32 However, there was no clear indication in the Government response that an arms-length body was being investigated. Joy Hutcheon said that:
“In a world where we were able to have a discussion with [the] Treasury about the implications of that for our running costs, that would be a very interesting proposition. I know you have discussed that with the Permanent Secretary on a number of occasions but that, at this stage, is not the world we are in.”33
20.Evidence to this inquiry and past IDC inquiries has suggested that an arms-length body to assume some project management responsibilities could be beneficial in both strengthening the expert oversight in DFID projects and deepening competition in the market. It is our view that such a body could contribute towards a suitable balance between in-house and outsourced expertise. It would also afford DFID greater direct scrutiny of contract-based delivery, a notable advantage given recent questions surrounding contractor conduct. We repeat earlier calls for DFID to investigate the idea of an arms-length body to assume programme management responsibilities. This should be done as part of its Supplier Review and the outcome reported to us. If DFID disagrees with this or believes it is not feasible, it should set out why in its response to this Report.
Contractors and fragile states
21.In its Aid Strategy, the Government committed to allocating 50% of all DFID’s spending to fragile states and regions, up from a previous target of 30%.34 In certain fragile and conflict-affected states (FCAS), the majority of DFID programming is delivered through contractors. A 2014 report from the Organisation for Economic Cooperation and Development (OECD) found that 75% of DFID’s bilateral programme in Nigeria (DFID’s third largest in 2016/17), which is designated by the Department as a fragile state, was spent through contractors.35
22.In response to questions on the impact of the fragile states agenda on DFID’s use of contractors, Nick Ford, Head of DFID’s Procurement and Commercial Department, said:
“It is not necessarily right to say that, because we work in FCAS, we are going to do more through contractors. It would depend on the nature of the programme.”36
However, DFID’s written evidence stated that growth in the number and total value of supplier contracts awarded “is driven by the shift towards fragile and conflict affected states where financial aid and delivery route options are limited [ … ]”. Comparative advantages of contractors in FCAS were highlighted in written evidence and include flexibility and adaptability in rapidly changing environments, access to relatively dangerous locations and the ability to maintain a low profile to keep security costs down (relative to multilaterals).37 It is thus feasible that, even on a case-by-case basis, increased spending in FCAS may be associated with an increased dependence on contractors. While contractors may represent a useful tool for DFID in delivering on its FCAS commitment, there are some key considerations that go with this.
23.Firstly, the Stabilisation Unit, a cross-government unit supporting UK government efforts to tackle instability overseas, identified certain challenges associated with partnering with for-profit organisations in FCAS. In its guidance note on outsourcing conflict, justice, security and stabilisation work, it wrote:
“[ … ] profit motives of implementing partners can at times be in tension with other objectives, and tensions between private companies and other state and non-state actors can affect overall impact, coordination the quality of partnerships. Unless HMG commits sufficient time to programme management, delivery can get divorced from HMG’s political strategy.”38
While robust programme management is particularly important in FCAS, it is in such environments that this is most challenging to deliver. As Joy Hutcheon noted “there are security and duty of care implications, and the access for monitoring is restricted.”39
24.Secondly, the Government’s November 2015 Aid Strategy also committed to an expansion of payment by results (PbR). This Committee has previously expressed reservations about whether there is sufficient evidence to support increased use of PbR,40 though there are particular issues with its use for contracts in FCAS. In ICAI’s 2014 report on delivering impact, it referred to PbR as a way of shifting risk from DFID to the implementer. However, in FCAS “many of the risks are beyond the control either of DFID or the supplier. In such cases, shifting the risk to the supplier may simply drive up costs for the aid programme as a whole.”41
25.In its 2015 Aid Strategy, the Government committed to allocating 50% of all DFID’s spending to fragile states and regions, up from a previous target of 30%. While DFID decides on the delivery channel for its programmes on a case by case basis, any perceived comparative advantage held by contractors is likely to result in their increased use. We share the Stabilisation Unit’s concerns about possible tensions between profit motives and programme objectives in fragile states. We also question whether the shifting of risk to contractors through Payment by Results (PbR) is likely to lead to substantially higher risk premiums which are then reflected in the cost of contracts. We note that with fewer contractors operating in such difficult environments this may have implications for market competition and value for money.
26.DFID should ensure the use of contractors in fragile states is carefully managed and that output targets are designed to align profit incentives and development objectives whilst still taking account of drivers of conflict. DFID should be particularly cautious about shifting risks onto contractors in fragile states through PbR. It should also ensure that all overheads and profits are benchmarked against other donors and recipient countries to ascertain whether DFID is achieving value for money.
Learning through contractors
27.In ICAI’s 2013 review, one of the major issues raised was concerned with how DFID learns from programmes implemented by contractors—an area which ICAI gave an Amber-Red assessment.42 ICAI found:
“Interviews with DFID staff show a low level of learning being disseminated from the contractors to the local DFID programme teams and, in turn, to headquarters. [ … ] There is no process which draws all the learning together and no process for setting strategic guidelines about the future circumstances in which using contractors would be advantageous.”43
28.DFID accepted the ICAI recommendation in this area, and said that it would take steps including asking contractors to “propose practical ways to bring their insights and learning from programmes back into DFID and to share these across programmes.”44 In oral evidence, Nick Ford referred to a number of practical measures implemented to improve learning, including: DFID’s annual supplier conference; specific and targeted forums throughout the year; DFID’s supplier relationship management programme; and early market engagement programmes used to scope the market and better inform programme design.45 However, evidence to this inquiry suggests that further improvements still need to be made.
29.Joy Hutcheon and Nick Ford made it clear in oral evidence that DFID believes that suppliers are forthcoming in sharing their learning, with the shared incentive of the “greater good” meaning that there is “a general ethos in the sector that people want to share learning”.46 However, evidence from Bond challenges this view:
“”[ … ] private contractor legal departments are attempting to impose increasingly heavy intellectual property clauses on sub-contractors working on programmes with them. These try to establish ownership on approaches and materials coming out of programmes and to require sub-contractors, who are often responsible for developing these, to seek permission for their use. Bond members feel this is inappropriate for publicly funded programmes and is detrimental to learning and sharing about what works and does not work.”47
It is certainly true that many staff operating in difficult circumstances on the ground are motivated by the development cause and will therefore share all their learning. However, recent reports on the conduct of certain contractors have raised questions over whether this can be considered a reliable assumption, particularly given the commercial advantage that learning gives to an organisation over its rivals.
30.Palladium wrote that: “We see the main opportunity for improvement being for DFID to specify and prescribe learning objectives for all contracted programmes, and to formalise the need for collaboration and sharing across the portfolio.”48 In line with Palladium’s written evidence, it does not appear that DFID includes concrete expectations on sharing learning in its delivery contracts with suppliers. An example is one contract with Maxwell Stamp PLC for a social protection programme in Bangladesh which makes no reference to “learning” or “knowledge-sharing”.49 This is a poorly performing project that received a ‘B’ rating in its most recent annual review.50 While many of the project performance issues are not the fault of the contractor (terror incidents made international staff recruitment and retention more difficult), it is concerning that there is no contractual obligation for the supplier to give feedback to DFID on what it has learned.51
31.Bond raised concerns that DFID annual reviews—a vital tool for learning and improving future programming—are “increasingly becoming box-ticking exercises rather than real opportunities for DFID’s own learning.” Bond again links this back to staffing constraints as DFID staff may lack the “necessary time or indeed skills to engage with the realities of these quite complex programmes.”52 A related concern was highlighted by an experienced contractor in confidential evidence:
“The biggest single structural weakness in the international development consultancy market is the apparent conflict of interest in the role of many of the firms that DFID employs to undertake reviews of its projects. [ … ] This can and does 1) lead to misleading results claims, and 2) undermine effective learning from experience.”53
The OECD Development Assistance Committee (DAC) guidelines state that evaluations should be “independent from the development intervention, including its policy, operations and management functions, as well as intended beneficiaries.” DFID stated that the independence of a review “will vary depending on the particular circumstances of the programme.”54 There is no clear guidance in DFID’s Smart Rules as to when evaluations should be conducted independently.
32.While DFID is a key purchaser of development contractor services, it is not the only donor in the international market. DFID contributed 13% of all ODA flows from DAC donors (through all channels) in 2015, and there is undoubtedly much to learn from the other 87%. Other major donors use very different contracting models, such as USAID’s “cost-plus” contracts where there is a fixed management fee at the inception of the contract, and the remuneration framework used by Australia’s DFAT in determining consultant fees. Nick Ford told us of some work he had recently done with UNOPS in Jordan which he said was “really insightful, and it made me think about how we could use our commercial advisers more in country to do similar events.”55
33.DFID has taken some key steps towards improving the way it learns through contractors, though concerns remain about whether this knowledge is effectively extracted and internalised. The assumption that contractors are keen to share learning for the greater good overlooks the fact that, as for-profit companies, there is a strong commercial incentive to withhold information from DFID and potential competitors. We stress the importance of independent evaluations, and note that there appears to be little clear strategy to determine the degree of independence in evaluations across various programmes. A number of other donors use very different contracting models which are likely to have their own advantages and disadvantages.
34.As part of DFID’s Key Supplier Management (KSM) system, it should apply a metric on the contractor’s contribution to DFID learning, including how cooperative they are with other contractors. Other innovative approaches should also be explored in the Supplier Review. DFID should use guidelines on best practice to set out clear regulations in its updated Smart Rules on when programme evaluations should be done completely independently of both DFID and contractor staff. It should also foster closer relationships with the procurement departments of other donors as well as multilateral agencies to try to learn about what works in different procurement models.
8 UK Government, UK aid: tackling global challenges in the national interest (November 2015)
12 US Government, US Agency for International Development: Total Funds Awarded by Type - FY 2016 (accessed 10 March 2017)
13 DFAT, Australian Engagement with Developing Countries Part 2: Official Sector Statistical Summary 2014–15 (April 2016)
14 National Audit Office, Managing government suppliers (November 2013)
15 DFID, DFID Smart Rules: Version VI (last updated 3 October 2016) p57
16 ICAI, DFID’s approach to delivering impact (June 2015) para 2.47
19 ICAI, DFID’s Use of Contractors to Deliver Aid Programmes (May 2013) para 2.13
20 ICAI, DFID’s Use of Contractors to Deliver Aid Programmes (May 2013) para 2.147
24 International Development Committee, Third Report of Session 2010–11, Department for International Development Annual Report & Resource Accounts 2009–10 (February 2011) p4
25 International Development Committee, Seventh Report of Session 2016–17, UK aid: allocation of resources (March 2017) para 94
29 International Development Committee, Sixth Report of Session 2014–15, Recovery and Development in Sierra Leone and Liberia (October 2014) para 50
31 Q40 (See Annex 1)
32 International Development Committee, Seventh Special Report of Session 2014–15, Government Response to the Committee’s Sixth Report of Session 2014–15 (December 2014) p8
34 UK Government, UK aid: tackling global challenges in the national interest (November 2015) p4
35 OECD, OECD Development Co-operation Peer Reviews: United Kingdom (November 2014)
38 Stabilisation Unit, Outsourcing conflict, justice, security and stabilisation interventions (March 2016) p17
40 International Development Committee, Seventh Report of Session 2016–17, UK aid: allocation of resources (March 2017)
41 ICAI, DFID’s approach to delivering impact (June 2015) para 4.41
42 An Amber-Red assessment means: The programme performs relatively poorly overall against ICAI’s criteria for effectiveness and value for money. Significant improvements should be made.
43 ICAI, DFID’s Use of Contractors to Deliver Aid Programmes (May 2013) para 2.140 and 2.147
44 DFID, Management Response to the ICAI recommendations on: DFID’s Use of Contractors for Aid Delivery, May 2013 (June 2013) p3
49 DFID 6600 Service Provider for Strengthening Government Social Protection Systems for the Poor (SGSP) in Bangladesh
50 By DFID’s rating system, a ‘B’ rating means that ‘Outputs moderately did not meet expectations’
51 DFID, Strengthening Government Social Protection Systems for the Poor (SGSP) in Bangladesh: Annual Review (November 2016)
53 Confidential written evidence (DUC0004) para 22
6 April 2017