House of Commons - Department for International Development's Annual Report and Accounts 2011-12

Department for International Development's Annual Report and Accounts 2011-12 - International Development Committee Contents

2 Expenditure

The UK's Official Development Assistance: key developments

Figure 1: The Department's spending in 2011-12

4. Total UK Official Development Assistance is increasing. It grew in cash terms by 2.1% in 2011 to £8,629 million, with the Department accounting for £7,716 million (89% of the total).[2] In 2013 ODA is due to reach 0.7% of GNI, which is currently estimated to be £11.3 billion (a reduction of £254 million from the Autumn Statement 2011). DFID's budget in 2013-14 will be £10.765 million. DFID's expenditure is split between its bilateral programme , mainly spent through DFID's country offices and multilateral organisations such as the World Bank and the EU (see Figure 1).

5. The NAO analysis of DFID's expenditure produced a number of findings which suggest that the challenge of spending an increasing budget was affecting DFID's ability to achieve value for money: notably, that the Department had to reschedule planned spending from future years and that the proportion of the bilateral budget which is being spent through multilateral organisations was increasing.

6. The NAO analysis also shows that DFID has switched spending from low to middle income countries and that there were great discrepancies between planned and actual expenditure in key areas such as the MDGs, Global Partnerships, climate change and wealth creation.[3]

Increased spending at the end of 2011

RESCHEDULING PAYMENTS FROM 2012 INTO 2011

7. The Department manages its spending so that planned levels of UK ODA can be achieved, but during the last quarter of 2011, to increase UK ODA to the planned level of 0.56 per cent of gross national income, the Department rescheduled into 2011 payments planned for 2012 worth £450 million, for example rescheduling a payment to the World Bank's International Development Association planned for April 2012 to 2011,[4] and increased lifetime spending on other projects by £130 million.

8. Officials explained that events arose 'that mean we have to rephrase or juggle when programmes come on stream or the rate at which we implement them'.[5] They added:

The key to doing this in a way that balances value for money and annual spending limits and targets is about having enough contingency and enough activity in the programme, so that when some activities fall away, which they always do every year-one thing we know about every year's spend is some activities you thought would happen at the start of the year will not happen- enough other things are on stream that represent good value for money and deliver good results.'[6]

9. We were assured that DFID would

not take decisions that we believe are poor value for money simply to hit a target. If we are surprised and the level of our contingency planning is insufficient to deal with changes in the real world, the casualty of that, in terms of the advice I provide, will be the target rather than the quality of investment.'[7]

SPENDING THE INCREASE IN BUDGET IN A COST-EFFECTIVE MANNER

10. DFID argued that it was able to spend the large increase in its 2013-14 budget efficiently and effectively. It had built a strong 'pipeline' of projects and programmes to deliver in 2013 and 2014. It had recruited 500 people over the last eighteen months 'to provide technical expertise on the front-line'. The Department therefore had the time and the right people to develop an effective 'pipeline' of projects. DFID had ensured the pipeline was of good quality by improving its 'business processes to drive value for money'.

11. When the NAO conducted its research for us in autumn 2012, the Department had projects in place which it estimated would account for around 75% of its 2013-14 programme budget and 60% of its 2014-15 programme budget. In addition the Department was developing other potential projects.[8] The Department subsequently provided a paper in [December] showing that the value of approved projects and projects it was currently designing (if subsequently approved) would account for nearly 100% of DFID's budget for the next two years (98.2% programmed for 2012-13; 98.9% for 2013-14; and 97.0% for 2014-15).[9]

THE ESTABLISHMENT OF A DEVELOPMENT BANK

12. We questioned officials as to whether a development bank, along the lines of the Agence Francaise de Développement, would ensure that DFID was not so constrained by the need to ensure that cash was spent by the end of the calendar year. The bank would also, more importantly, provide DFID with mechanisms which would put it in a better position to meet future challenges. such as offering concessional loans alongside grants. The bank could follow the model used by the French development agency whereby a 'development bank' operates within the organisation.

13. We received a very positive response from the Permanent Secretary:

the development bank concept-this is just a personal comment-that is a really interesting idea for us to look at in general terms. We have a narrower suite of instruments, ways of providing support for development, than some other donors do. From time to time, it is worth asking if we should explore a wider range of instruments. Next year, in a way, is not a bad year to be having a look at that, because the goals for international development with the post-MDG Panel, the G8 and so on will evolve after 2015.[10]

14. From 2013 the UK will be increasing ODA to 0.7%. It is vital that this money is spent cost-effectively. At the end of 2011 the Department took action to boost calendar year spending by £580 million and thus hit the UK's ODA target. The Permanent Secretary told us that the Department would not take decisions which did not represent good Value For Money. DFID should miss ODA calendar year spending targets where there are delays or cancellations to its planned projects and it does not have alternatives which provide good value for money.

15. The establishment of a Development Bank would not only mean that DFID was not so constrained by the need to ensure that cash was spent by the end of the year, but also, more importantly, would provide DFID with a mechanism to offer concessional loans and put it in a better position to meet future challenges, including its engagement with middle income countries.

Increase in spending on multilaterals and reduction in aid to recipient governments

Figure 2: The Department's bilateral expenditure by type of aid, 2007-8 to 2011-12

NOTES

1. Financial aid to recipient governments in 2011-12 comprised £242 million of General Budget Support, £294 million of Sector Budget Support and £545 million of other financial aid, which includes projects and programmes not classified as poverty reduction budget support.

Source: NAO presentation of data from the Department for International Development, Statistics On International Development 2007-08 - 2011-12, October 2012

16. The Department's overall level of expenditure in 2011-12 was £7,682 million, which was similar in cash terms to 2010-11 levels. The proportions of the Department's spending going as core funding to multilateral organisations (i.e. multilateral aid) increased from 38% in 2007-08 to 42% in 2011-12. There has also been a major switch in how bilateral aid is spent since 2007-08 with a considerable increase in the amount going through multilateral organisations. (see Figure 2). This is now the largest element of the Department's bilateral programme.[11] Funding to multilateral organisations totalled £4,900 million in 2011-12 with £1,650 million from the bilateral programme and £3,250 million in core funding.[12] Thus almost two-thirds of DFID's spending goes through multilateral organisations.

THE DECLINE IN BUDGET SUPPORT

17. Financial aid to recipient governments used to be the largest element of the Department's bilateral programme, but it is now at its lowest level for five years due to reductions in general budget support.[13]

Figure 3:

Changes in the value of general and sector budget support

Country 2010-11 Budget Support (£ million) 2011-12 Budget Support (£ million) % Change

General budget support

Pakistan 30.0 - -100%

Malawi 19.0 - -100%

Ghana 36.0 12.3 -66%

Zambia 32.8 12.5 -62%

Uganda 27.2 20.0 -26%

Tanzania 103.5 80.0 -23%

Mozambique 48.2 48.0 0%

Vietnam 20.0 20.0 0%

Rwanda 35.8 37.0 3%

Sierra Leone 8.0 12.5 56%

Total General budget support 360.5 242.3 -32.8%

Sector budget support

India 46.0 - -100%

Vietnam 9.8 - -100%

Malawi 26.7 14.0 -47%

Mozambique 28.6 21.4 -25%

Pakistan 32.5 38.0 17%

Ethiopia 94.7 132.8 40%

Ghana 25.0 46.0 84%

Nepal 7,0 14.0 100%

Rwanda 10.5 23.2 120%

Uganda 5.0 N/A

Total Sector budget support 283.3 294.4
4%

Source: NAO presentation of Departmental data

18. UKAN/BOND were concerned that

since 2008-09 DFID has been moving away from 'aid modalities' (ie. the way in which aid is provided by donors) that promote country ownership - a strong determinant of aid effectiveness and the first Busan principle. ...Large reductions in general budget support (GBS), have only partly been compensated for by increases in sector budget support.

The organisations recommended that the decline in the use of budget support be reversed.[14]

19. We asked the Permanent Secretary why DFID was reducing budget support and was passing money over to multilateral organisations to deliver what were effectively bilateral programmes: whether it was the most effective way of achieving DFID's objectives, whether it was because administration costs could be off-loaded or whether it was 'a way of just getting the money out' because the Department could not 'spend it fast enough'. In response, we were told that the main reason was that progress had been made on the UK Government's four objectives for the use of budget support when it was introduced just over a decade previously. The Permanent Secretary told us these objectives were:

i) to help countries stabilise their macro economy

ii) to provide more financing for the recurrent costs of developing countries' budgets, "so they could pay the teachers and health workers needed to achieve those MDGs in countries where the tax base is very small"

iii) to try to promote improvement in public financial management and public expenditure management

iv) to reduce the transaction costs between governments and the donors.

20. He reported "very good progress" on the first two and "some progress" on the second two, and said that "because we have made that progress, there is a smaller need for general budget support".[15] We pressed him on why in countries like Ghana, Zambia and Tanzania, where budget support had worked quite well, DFID had reduced it. We were told that:

in Ghana, the economy has grown substantially over the last 10 years and the tax base has increased, but also they have oil coming on stream, so they are less reliant for funding their recurrent cost budget on general budget support. …

The economy in Zambia has grown very substantially, …but that has not translated by the degree everybody would have wanted in either strengthening the budgetary position or reducing poverty. We also, like you, have a number of concerns about the composition of the budget in Zambia, especially over the maize subsidy. … We … need to engage in a continuing way with the Zambians on building the tax base and improving the use of their own tax resources.

Tanzania is again a different case. The economy has grown very substantially…. The tax base has grown pari passu with that. They do finance a lot more of their own recurrent budget now than used to be the case.… Accordingly, we do less general budget support and sector budget support actually, and we are programming to do less still over the years ahead.[16]

21. The Permanent Secretary said a "subsidiary reason" for reducing the use of budget support was the Department's increasing focus on fragile and conflict-affected states, where it was often not sensible to provide budget support.[17]

22. DFID stressed that there were other mechanisms than budget support in fragile and conflict-affected states where Government to Government aid was not appropriate.[18] The Permanent Secretary set out three alternatives to budget support:

The first route is to give more grants to NGOs. The second route is to use private-sector contractors. The third is to work with multilateral organisations, design programmes with them and use them as an implementing channel.[19]

We will discuss the first and third of these alternative options in this section, and the second later on in this report in the section on consultants.

OBSTACLES TO MAKING MORE USE OF NGOS

23. Regarding the first option, we questioned DFID about the obstacles to making more use of NGOs and whether some of the procedures the Department used, such as the Logframe process,[20] were too bureaucratic and difficult for smaller NGOs—even though these organisations might provide the best value for money. Richard Calvert of DFID told us:

for small organisations that may be new to using a Logframe, it can be a bit daunting. Mostly we have tried to offer advice ourselves, provide structured training and also not to necessarily knock back a project just because an organisation does not get it right first time.[21]

DFID'S INCREASING SPENDING THROUGH MULTILATERAL ORGANISATIONS

24. The third method of re-directing money that might previously have been allocated as budget support also interested us. As we said above, almost two-thirds of DFID's budget now goes to multilateral organisations. On our overseas visits we have often been struck by the lack of effectiveness of some multilateral organisations.

25. Multilaterals have higher administrative costs than DFID. The Department provided us with information on multilateral administration costs (see Appendix1) and trends (See Appendix 2). This information is based on a review of annual reports, budgets, audited/unaudited financial statements and dedicated administration cost papers. DFID stressed that

due to a number of caveats, these figures were approximations.

The nature of the work carried out by different organisations varies and this has implications for administration costs. Staff costs are a large proportion of administration costs and as such organisations that do a lot of policy/advocacy work as opposed to programmes/projects tend to have higher administration costs - but this will depend on how the organisation defines admin costs as some organisations will only count the costs associated with the programme/project work that they do. These organisations also tend to require specialist/niche staff input which can be expensive. Other organisations manage funds and do relatively less policy or implementation work and as such require relatively less staff input, which contributes to lower administration costs. [22]

Despite these caveats, DFID believe this is the best information currently available and provides a reasonable indication of the extent of administration costs across these organisations.

26. DFID is working with other donors to establish a cross-donor consensus on these issues and to encourage a more systematic approach to cost classification and reporting. This is a priority for DFID for the first half of 2013 and it is working with the USA on a joint paper for a Senior Donor Level Meeting.

27. The Permanent Secretary emphasised that reducing the costs and improving the effectiveness of multilaterals was a priority for DFID, adding that there had been improvements, e.g. at UNICEF and UNDP.[23]

28. About two-thirds of DFID's expenditure in 2011-12, including nearly 40% of the Department's bilateral expenditure (spending through country offices), went through multilateral organisations. This represents a major change in recent years and has been accompanied by a decline in general budget support to recipient Governments. DFID should carefully examine its growing multilateral expenditure and ensure that it has thoroughly examined other options such as greater use of local NGOs and sector budget support, especially as multilaterals have high costs and too often limited effectiveness. We recommend that DFID provide an annual report on the progress it is making on reducing the administration costs of multi-lateral agencies through cross-donor consensus. We commend DFID for demonstrating leadership in analysing multilateral costs and seeking to reduce them. We also welcome the Permanent Secretary's willingness to look at making more use of 'sector' budget support.

29. We also recommend that DFID

develop specific strategies for engaging with NGOs and civil society organisations (CSOs) which ensure that the Department's bureaucratic procedures and funding mechanisms facilitate partnerships with effective, smaller NGOs and CSOs;

ensure that there is an effective 'access point' for smaller NGOs and CSOs to make it easier for them to engage with DFID; and

provide clear guidance to country staff regarding its expectations for working with smaller NGOs and CSOs.

The switch of non-humanitarian bilateral aid from low income to middle income countries

30. The proportion of the Department's non-humanitarian bilateral aid going to low income countries fell by 15 percentage points to 65% (£1,709 million) in 2011-12.[24] This reduction has been caused in part because several former low income countries with a large number of poor people have recently become middle-income countries. We were told that DFID:

had decided to focus more … resources on countries that formerly might not have been classified as low income, like Nigeria or Pakistan, but in fact have a massive share of the remaining MDG burden...In northern Nigeria, Pakistan and other countries…there is very extreme poverty, even though they may be getting above some of those thresholds in terms of per capita income. We are … not sure those thresholds are … necessarily the best measure to determine what we should do, … We are not targeting a particular number for the proportion to go to low income countries at the moment, essentially because we think there are other more important measures.[25]

31. However, while spending on some countries with large numbers of poor people such as Nigeria and Pakistan is increasing, grant aid to India, the country with the largest number of poor people, is to end in 2015. We questioned the Permanent Secretary about the potential inconsistencies of approach, who told us:

we need a degree of consistency…, but I do not think we can have a mechanistic formulaic approach to this, because countries are all different and our relationships with countries are different[26]

We do have a view on the kinds of criteria that will be relevant [to when aid ceases]. Progress on the MDGs is one. Growth of the economy is another. The building-up of their own tax base is a third. A fourth one is the country's own preferences and desires. We need to take account of that as well, as importantly we did with India. There is a set of things we need to look at, but we cannot be too mechanistic about it and it will be different in each country.[27]

32. The Permanent Secretary, Mr Lowcock, also stressed that DFID was retaining a development co-operation relationship with India… 'on the private sector, on expertise in capacity development and on our shared roles on global development issues. What we are concluding by 2015 is our finance grant aid relationship'.[28]

33. DFID has switched expenditure from low to middle income countries. This has occurred in part because several countries with a large number of poor people have recently graduated to middle-income status. The Department's policy towards middle income countries is contradictory: it is ending grant aid to India in 2015, but programmes in Nigeria and Pakistan are due to grow rapidly. The Department explains that this is because both countries have a massive share of the remaining MDG burden, yet the same logic applies to India where a different approach is being taken. As expenditure on middle income countries increases, it is even more important that decision-making is rational and consistent; we recommend that the Department establish and make public the criteria it will use to inform decisions of when and how it should cease to provide aid.

34. We note that while DFID has increased expenditure in several middle income countries, its bilateral programme in one of the world's poorest countries, Burundi, has ended. We urge the new Secretary of State to re-instate a bilateral programme in Burundi.

The Department's spending by its development policy priorities - 'pillars'

Figure 4

Variance between the Department's estimate of 2011-12 expenditure and outturn, by pillar

Estimate Outturn Variance (outturn less estimate) Percentage variance

Resource expenditure

Wealth Creation 514.4 421.2 -93.2 -18.1%

Climate Change 237.5 157.8 -79.7 -33.6%

Governance and Security 673.8 720.3 46.5 6.9%

Direct Delivery of Millennium Development Goals 3063.5 2183.4 -880.1 -28.7%

Global Partnerships 576.2 1529.3 953.1 165.4%

Capital expenditure

Wealth Creation 97.5 128 30.5 31.3%

Climate Change 46.6 67.1 20.5 44.0%

Governance and Security 27.3 18.5 -8.8 -32.2%

Direct Delivery of Millennium Development Goals 236.0 117.4 -118.6 -50.3%

Global Partnerships 1242.7 1323.5 80.8 6.5%

Source: NAO presentation of departmental data

35. In 2011 the Department began to move towards allocating and reporting its expenditure by development policy priorities. It established five pillars which reflected the priority areas set out in the Department's business plan. The pillars are: wealth creation; combating climate change; governance and security; the direct delivery of the Millennium Development Goals; and global partnerships.[29] It set out estimates for expenditure under the five pillars.

36. The Department's 2011-12 Annual Report shows some large variations between its estimate and actual spend by pillar.[30] For example, the resource budget for the Direct Delivery of the MDGs was under spent by £880 million (29%) and resource spend on Global Partnerships exceeded budget by £953 million (165%).

37. The Department admitted that the reason for the variations in this first year of reporting by pillars was that 'we got our initial estimates wrong'.[31] DFID added that

We have now this year integrated that reporting into our system, so that when people enter projects on to the system, commit projects, we pick up the pillar analysis at the time. When we first did the estimate for 2011-12, we had only just decided to report on that basis and essentially had to do the estimates off system.[32]

We were told that did not mean that the underlying projects and programmes were not delivered or were changed. It was essentially about the way the projects were categorised against the five pillars. It was hoped that in 2012-13 that 'our final reporting will be much more accurate, because we have integrated pillar reporting into our systems much more effectively this year'.[33] The Permanent Secretary summed up:

we made a mistake in not paying enough attention to this bit of the financial reporting, and we will sort that out.[34]

38. We queried whether the pillar reporting was useful. Officials replied:

This is a high level analysis. It does not fundamentally drive our project decisions.…. We think it is a useful complement to our other reporting mechanisms. It is probably not the most important, but if we can, as we now believe we can, pull this information out from our core project reporting system, it is another way of looking at the programme.[35]

39. In its Annual Report and Accounts, the Department reported some huge discrepancies between actual and planned spend by pillar. DFID admitted that it got its initial estimates wrong, but said that whilst data on spending was useful it was giving more attention to results. The calculation of pillar spending appears to have little relevance to the Department's spending decisions. The Department should either explain its relevance or delete the mechanism altogether.

Research

PRIORITISING AND COMMISSIONING RESEARCH

40. The Department's central research programme increased in cash terms by 126% in the six years to 2011-12 to reach £220 million, and is expected to grow by a further £100 million over the three years to 2014-15. [36]

Figure 5: Research and Evidence Division's expenditure on research, 2008-09 to 2011-12 actuals and 2012-13 to 2014-15 planned

NOTES

All values in cash terms

Values for 2013-14 and 2014-15 are based on provisional budgets which, like those of the Department's other business units, were being revisited in December 2012.

Values include core funding provided by Research and Evidence Division to multilateral organisations such as the Consultative Group on International Agriculture Research.

Source: NAO presentation of departmental data

41. The Department conducts research to understand and identify solutions to the challenges of achieving the Millennium Development Goals and to reducing poverty and the effects of poverty. [37]

42. The Department uses six different routes to obtain research evidence.[38] The largest of the 6 routes which takes 27% of all DFID's research expenditure is spent on providing core funding along with other donors for research institutes such as the Consultative Group on International Agriculture Research (CGIAR). Typically the Department seeks to ensure that this expenditure represents value for money in following ways: it is represented on the boards of these research institutes; it monitors its own performance frameworks which set out what it wants an institution to deliver; and it reviews funding each year. Thus it seems there is no market testing for the largest category of DFID's budget.

43. Moreover, the Department's Research and Evidence Division does not have the staff to directly manage the commissioning and oversight of all smaller research exercises; thus some smaller projects are commissioned through programmes the Department has with the Research Councils. To maintain a balanced portfolio of large and small projects the Department may need to make greater use of the Research Councils and other donors and partners, such as the Gates Foundation, to commission research.[39]

44. We asked DFID officials the rationale for increasing spending on research and were informed:

We spend just under 3% of our budget on research. We invest that money to improve the quality of the impact of the rest of the budget that we spend, but also we are genuinely providing a global public good here, improving the quality and impact of other people's spend, whether it be other donors or developing countries. There are some fantastic examples of where our research, whether it be in flood resistant rice or drought resistant maize, has really delivered very significant returns.[40]

45. Tracking the proportion of the Department's research undertaken in the UK is made difficult by the number and location of organisations that can be involved in undertaking a project.[41] We questioned officials about the impact of DFID's research spending on UK universities and how much it benefits UK universities. The Permanent Secretary responded that he had been to a number of universities in 2012, including Reading, the University of East Anglia, the London School of Economics and Sussex. DFID was trying to strengthen its links and collaboration. He also thought it important that universities in the UK strengthen their links with partner organisations in developing countries. He concluded

Britain has some of the world's greatest higher education institutions and research capability. It is a really smart thing to do, we think, to put that to work for development goals. I expect us to be doing more of it in future.[42]

46. DFID's spending on research is increasing significantly. We are surprised that 27% of research expenditure goes in core funding to major international research institutes and is not subject to market testing, but monitored by DFID representation on the Boards of institutes. DFID's large research expenditure has the potential to benefit UK universities as well as providing the evidence needed to improve the delivery of aid. The Department has limited information on how much of the research it commissions is conducted in the UK. We recommend that the Department review its research portfolio to identify any sectors where UK universities look under represented. It should use this information to inform its on-going engagement with universities.

MAKING GOOD USE OF RESEARCH

47. To encourage the use of research, the Department has increased the attention and funding it has given to communicating the results of research to its staff and to external audiences.[43] The Department is seeking to develop its measures of the performance and results of its research activities: to date quantitative measures have focused on the number of research outputs made public and the number of times they were accessed. [44] In addition, the Department's resource allocation and project approval procedures are placing greater emphasis on the use of evidence.

48. DFID told us

we are using that research well in our decision-making. …we have put a lot of effort into this over the last couple of years. The more we use evidence effectively through the business-case process, the more that will help also to ensure that we commission the right types of research.[45]

However, unfortunately the NAO reported that the Department's staff were not yet making consistently good use of research.[46] In February 2012 the Department's Quality Assurance Unit reported that 17 (59%) of the 29 business cases it examined in terms of the choice, function, judgment of quality, or use of evidence were lacking, inappropriate or poorly employed. We were told that training had been designed which aimed to improve the way the Department's staff use evidence.[47]

49. We are concerned that DFID staff are not yet making consistently good use of the research it commissions. The Department should improve its staff training and put increased resources into its Research and Evidence Division.

2 NAO Briefing, paragraph 1.11 Back

3 NAO briefing Back

4 NAO briefing, paragraph1.12 Back

5 Q24 Back

6 Q24 Back

7 Q24 Back

8 NAO Briefing, paragraph 1.10 Back

9 Ev 24 Back

10 QQ 24-25 Back

11 NAO Briefing, paragraph 1.4 Back

12 NAO Briefing, paragraph 1.4, footnote 7 Back

13 see figure 3 NAO briefing, paragraph 1.5 Back