Select Committee on International Development Sixth Report


23. PSAs and country planning processes play complementary roles in knitting together the geographical and sectoral axes which make up DFID's organisational matrix, and connecting DFID's internal organisation with its external environment and relationships. The extent to which they achieve these goals is likely to have a large bearing on both the effectiveness of DFID as a development agency, and on its level of accountability. As such, we welcome the fact that DFID seeks to learn and develop as an organisation, and has been re-designing both its PSA and its country planning processes.

24. The 2001-2004 PSA was itself a development from the 1999-2002 PSA. The indicator used for the maternal mortality target was changed from maternal mortality itself to a more easily measurable internationally accepted proxy—births assisted by skilled attendants. In addition two targets on GDP growth were dropped as their measurement was skewed by a few populous countries and made unreliable by the irregularity of data collection.[24] The most striking example of moving the goalposts concerns how access to water is measured. The indicator used to monitor progress as regards access to water changed during the course of the 1999-2002 PSA from 'percentage of population with access to safe water', to 'percentage of population with access to an improved water source'. DFID sought to reassure us that this change was for largely statistical purposes—data is easier to collect and there is international agreement on the use of the new indicator—rather than a deliberate relaxation of the target. We accept this, but do have some concerns about the extent to which targets and objectives might be driven by data availability. We appreciate too that statistics need to be collected and assessments made even whilst data collection is not ideal. But, we do think that an important medium-term response to problems of data collection is to improve the capacity of developing countries to monitor progress towards the MDGs and associated targets and indicators. Future Departmental Reports might usefully include more information about what DFID is doing to build the capacity of developing countries in this regard. We will monitor closely DFID's selection and use of targets, and any changes to measurement methodologies.

The 2003-2006 PSA

25. The new PSA, negotiated and agreed with the Treasury, was announced along with the conclusions of the Spending Review on 15th July 2002. The associated Service Delivery Agreement (SDA) and Technical Note will be published soon. The design of the new PSA is intended to address various shortcomings identified with earlier versions, most particularly to ensure that the MDGs drive DFID's performance through the PSA.[25] DFID accepts the assessment made by the NAO that whilst DFID has internalised the MDGs, the PSA has not enjoyed full ownership within DFID and hence has not been driving performance at the level of delivery.[26] This lack of ownership can be explained in part by the fact that the PSA objectives and targets are sectoral, and that at the level of delivery, DFID—organised into country teams—is of course primarily geographical.

26. The new PSA is also intended to address the problem of attribution; the fact that in an international development partnership it is impossible to prove the extent to which a development outcome has been caused by an individual development agency. This issue was discussed in depth by the NAO in its report on DFID's performance management,[27] and by the Public Accounts Committee in its recent report.[28] We do not intend to revisit this issue in detail, but DFID—along with all development agencies, NGOs, Governments and others involved in international development—struggles to cope with the problem of attribution. It is important that the PSA clearly links DFID's activities to the MDGs, but this must be done in a realistic manner which accepts the impossibility of attributing causal impact to DFID in any simple way. The new PSA, in common with the previous one, seeks to make progress in this direction, in part by focussing attention on a smaller selection of key countries where DFID spends the majority of its resources.

27. DFID told us that the new PSA will set MDG-related outcome targets which "map directly onto our internal organisational structure". It will make senior managers directly accountable for delivering PSA targets, and will enable every member of staff to see the link between their work-plans, the PSA and the MDGs.[29] As Suma Chakrabarti put it: "What we are trying to move towards is having that sort of process whereby the Millennium Development Goals would be reflected in the new Public Service Agreement. Then we would expect each of the country directors—Africa, Asia, the rest of the world—to come up with a strategy for achieving the PSA outcome that they are responsible for."[30] The intention is to create a "ladder of accountability" and a "cascading" system of objectives and targets.

28. As DFID explained to us in a written response:

    "At present, the Secretary of State is formally accountable for progress against DFID's 2001-04 PSA and thus progress towards the MDGs. Within DFID, senior management are collectively responsible for DFID's contribution to the MDGs. The performance management system designed around the proposed 2003-06 PSA will formalise this internal responsibility by holding senior management directly accountable for progress against PSA targets and objectives. Senior Management will manage performance through six-monthly reviews drawing on the Directors' annual PSA progress reports, Directors' Statements and the annual Departmental Report. They will also be provided with the quarterly PSA returns to the Treasury. Individual Directors in discussion with advisers and other colleagues will incorporate MDG/PSA objectives and targets directly into their job plans and allocate resources accordingly. Their success will be measured yearly against delivery of outcomes. Successful delivery will require that PSA objectives and targets be 'cascaded' down to departmental, team and individual work plans. This will ensure ownership by all parts of the office of each target and create a 'ladder of accountability'."[31]

29. The new PSA, as expected, incorporates some important changes. DFID's overall aim is to "eliminate poverty in poorer countries, in particular through achievement of the MDGs".[32] There are then five main objectives with associated performance targets. The five objectives are to: reduce poverty in sub-Saharan Africa; reduce poverty in Asia; reduce poverty in Europe, Central Asia, Latin America, the Caribbean, the Middle East and North Africa; increase the impact of key multilateral agencies in reducing poverty and effective response to conflict and other humanitarian crises; and, develop evidence-based, innovative approaches to international development.

30. With the unexplained exception of the third objective, and a sixth concerning value for money, each of these objectives has a series of primarily sectoral outcome targets against which progress will be measured, with 2000 as the baseline year and 2006 as the measurement date. For objective one, the focus is on 16 key countries, and, for objective two, 4 key countries. As such, the new PSA includes a mixture of both geographical and sectoral objectives, targets, and measurement strategies, a mixture which maps more comfortably onto DFID's organisational structure. As ODI put it: "The creation of Directors' Delivery Plans for each Department within DFID brings greater consistency in DFID management structure and the PSA targets. These Plans will articulate how regions (Africa, Asia, etc.) and policy departments will work to deliver the PSA. In this way, DFID has attempted to establish more policy-led, top-down management to complement its traditional, responsive country-led programming and provide it greater coherence."[33]

31. A further noteworthy feature of the 2003-2006 PSA is its recognition of the need for a clearer focus on a limited number of countries. This might achieve multiple goals: the more effective use of resources; an improved ability to address the attribution issue and to demonstrate the links between DFID's work and development outcomes; and, a greater ability to monitor progress in particular countries. This recognition is expressed both in the strategy for measuring progress against targets in selected key countries, and—as we discuss below—in DFID's streamlining of its portfolio of countries.

32. We welcome the new PSA as a serious attempt to develop a set of objectives and we hope, increasingly SMART—specific, measurable, achievable, realistic and timed—targets which will both drive performance and facilitate accountability, and which will map onto DFID's evolving organisational structure.[34] We also welcome the continued efforts to grapple with the issue of attribution. However, we do have some concerns about the focus on a limited number of countries. We are uncertain whether this is for measurement purposes, or whether this is part of a move to greater selectivity in the number of countries which DFID is involved in. There may well be good policy reasons—beyond making the measurement of progress more practicable—for a sharpening of DFID's focus; these should be re-iterated. We, along with DFID, also feel that a three-year time-frame does not provide a fair or realistic time-span within which to assess DFID's contribution to international development, and urge HM Treasury to think seriously about the suitability of such a framework for DFID. We also note with some concern the lack of targets or indicators as regards the effective response of multilateral agencies to humanitarian crises.

Figure 6: DFID's 2003-2006 PSA

33. Finally, the 2003-2006 PSA includes three joint targets. The first—with the Foreign and Commonwealth Office and the Ministry of Defence—concerns conflict prevention and management. The second—with HM Treasury—concerns debt relief and the MDGs. The third—with the Department for Trade and Industry and the Foreign and Commonwealth Office—concerns the reduction of trade barriers. We welcome the inclusion of joint targets as a useful mechanism to encourage the joined up Government which international development so clearly demands.

Country Assistance Plans and Policy Teams

34. Since its inception, DFID has had institutional structures and processes to link its stated policy objectives to its actions. This has been achieved through the production of strategy papers; Target Strategy Papers produced by sectoral advisers' groups, and Country Strategy Papers produced by regional departments and country offices.[35] Both the country planning process and the structure of policy teams are in the process of change, in part as a response to the identification of weaknesses with existing approaches. Howard White of IDS and David Booth of ODI have argued that DFID's country planning processes, and specifically the Country Strategy Papers have had a "missing middle".[36] That is, they state clearly DFID's poverty reduction objectives and DFID's planned spending, but lack analysis of the causes of poverty and strategies to break into these causal relationships. The new Country Assistance Plans are in part an attempt to fill in this "missing middle". As DFID wrote before the publication of the new PSA:

    "DFID has recently undertaken a major review of the Department's country planning processes to ensure that DFID's proposed interventions would be explicitly linked to DFID's PSA targets and thus the MDGs. DFID's new Country Assistance Plans will be required to include an annex reporting on progress at the country level against PSA objectives/targets, as well as MDGs. The Plans represent performance contracts between the country team and the Regional Director. They will be annually reviewed and will feed into each Director's report on progress in their Divisions. In the 2003-06 PSA ... regional targets have been aggregated from targets set out in countries' own poverty strategies. This will ensure close alignment between the MDGs, PSA targets and the Country Assistance Plans' interim outcomes".[37]

35. The Country Assistance Plan (CAP) will be an annual document describing how DFID intends to support a country's own development plans.[38] For those countries which have what DFID considers to be a serious Poverty Reduction Strategy Paper (PRSP), the CAP will build closely on that. In those countries where there is either no PRSP, or a PRSP but no serious commitment to reform, the CAP will outline what actions DFID will take to continue to engage with the Government and NGOs. We applaud the development of CAPs as an attempt to fill the "missing middle", and welcome the integration of CAPs with PRSPs and the implicit recognition that DFID needs to engage with different countries in different ways.[39] As well as being a forward planning document, the CAP will have an evaluation function, providing data on a country's progress against key MDG targets, and providing a country-level basis for DFID's outcome-based monitoring of its programmes. This will also reveal the strengths and limitations of local data collection systems, and provide a basis for making decisions about where and how to best offer capacity building assistance in this regard.[40]

36. In addition, DFID is currently restructuring its policy division. There are now five advisory groups dealing with specific aspects of development which cut across geographical boundaries—education, governance, health and population, infrastructure and urban development, and social development—as well as the primarily regional divisions. The intention is that the structure and remit of these advisory groups, which will be responsible for realising non-geographic goals as well as supporting country-level work, will be more closely aligned with DFID's core work and the MDGs.[41]

24   Ev 26, paragraph 5. Back

25   In his contribution to the NAO report-on page 63-Howard White wrote that in response to his question as to what DFID's response should be to the likelihood that none of the International Development Targets would be met at the global level, virtually all respondents thought that no response was necessary. White's conclusion-that this shows that data on the Targets embodied in the PSA do not yield information of operational significance-would seem to be reasonable. See footnote 8 for web-site. Back

26   Ev 1, paragraph 2. Back

27   National Audit Office Report, Session 2001-2002, Department for International Development: Performance management-helping to reduce world poverty, HC739, page 26. See footnote 8 for web-site. Back

28   Forty-Eighth Report from the Public Accounts Committee, Session 2001-2002, Department for International Development: Performance management-helping to reduce world poverty,HC793, paragraph 13. See footnote 2 for web-site. Back

29   Ev 1, paragraph 2. Back

30   Q3 Back

31   Ev 31, paragraphs 36-38. Back

32   HM Treasury, 2002 Spending Review, Public Service Agreements 2003-2006, CM5571, page 23. See Back

33   The Millennium Development Goals and the IDC: Driving and framing the Committee's work, ODI, July 2002, paragraph 91. Back

34   National Audit Office Report, Session 2001-2002, Department for International Development: Performance management-helping to reduce world poverty, HC739, page 24 and 25. See footnote 8 for web-site. Back

35   The Millennium Development Goals and the IDC: Driving and framing the Committee's work, ODI, July 2002, paragraph 88. Back

36   "Using development goals to design country strategies", chapter 4 in Targeting development: Critical perspectives on the Millennium Development Goals and International Development Targets, edited by Richard Black and Howard White, Routledge, 2002. Back

37   Ev 31, paragraph 34. Back

38   The first two CAPS, for Ghana and Peru, have been produced. See- Back

39   See also, Fifth Report from the International Development Committee, Session 2001-2002, Financing for development: Finding the money to eliminate world poverty, HC785, page 39, figure 12. See footnote 19 for web-site. Back

40   The Millennium Development Goals and the IDC: Driving and framing the Committee's work, ODI, July 2002, paragraph 89. Back

41   See; also see The Millennium Development Goals and the IDC: Driving and framing the Committee's work, ODI, July 2002, paragraph 90. Back

previous page contents next page

House of Commons home page Parliament home page House of Lords home page search page enquiries index

© Parliamentary copyright 2002
Prepared 14 November 2002