DFID in 209-10 and the Resource Accounts 2009-10

Written evidence submitted by Adam Smith International

DFID’s approach to assessing the effectiveness of aid and the role of DFID’s new aid watchdog

Overview

1. Evaluation of DFID’s own projects and programmes is fairly thorough and was strengthened in 2009. In general there is a clear logframe produced before the project or programme starts, and progress against that logframe is monitored and evaluated as it progresses, with a final assessment made at the end.

2. Those performing the monitoring and evaluation (M&E) are usually independent advisers appointed for the purpose. Programmes are often altered as a result of M&E recommendations, demonstrating flexibility and the ability to learn in real time. In addition DFID’s evaluation department conducts backward-looking evaluations a few years after programmes have concluded, allowing the passage of time to provide a clearer perspective.

3. The central problem with evaluation of DFID spending does not relate to the programmes and projects conducted by DFID itself, but to the funds given to other organisations to spend.

4. Creating an external evaluation body will lend greater credibility to the evaluation process but it will have little impact unless it can also deal with funds given to other, third party organizations.

5. Spending by third party organizations is generally of a much poorer quality that that conducted by DFID directly, with extremely limited opportunities for DFID to evaluate results. The problem is in this area, not in DFID’s direct spending. Thus creating an external watchdog which focuses only on DFID’s direct spending would largely miss the opportunity to deal with the area of greatest ineffectiveness.

Evaluation of budget support

6. Budget support, in which funds are paid directly to developing country governments, is extremely difficult to monitor or evaluate. As DFID says: "the aid is spent using the government’s own financial management and procurement systems and is accounted for using the government’s own accountability systems." These are in every case wholly or partially defective. To be effective budget support relies on:

• A strong, effective and independent audit function – to ensure that funds are spent properly and effectively

• Strong public financial management systems

• A strong, transparent budget system that is open to public scrutiny

• Clear evidence that Ministers, MPs and the rest of the country’s elite are paying their fair share of taxes

• Effective parliamentary oversight of the budget process and public spending, with parliamentarians able to analyse budget proposals, monitor expenditures and evaluate outcomes

• Effective civil society oversight of the budget process and spending

• Effective independent judicial and anti-corruption institutions

• A strong, effective and independent statistics function – to provide accurate data on state expenditure

• Effective delivery of public services and means of qualitative analysis of such services

• A free media, so corruption and other abuses can be investigated and publicised

7. Countries in receipt of UK budget support lack most of these attributes, in some cases all. Of all the countries in receipt of UK budget support only India can be said to have an effective independent audit institution. Tanzania provides a pretty typical picture of a country in receipt of UK budget support:

• The Auditor-General estimated that 20% of the Government budget is lost due to corruption

• A World Bank survey identified major leakage from a Primary Education Development Programme, but the Government orchestrated a cover-up

• Massive corruption was revealed in natural resource aid programmes

• Very poor budget transparency – minimal information is provided to the public

• The Supreme Audit body is not independent and not able to fulfil its assigned tasks due to lack of finances, infrastructure and human capacity

• There is excessive expenditure on a large & inefficient civil service

If we take Pakistan as another example, the picture is similar:

• Endemic corruption: The 2009 Transparency International corruption perceptions report says corruption has grown 400% in the last 4 years. A 2009 World Bank report states corruption is ‘a serious and growing obstacle to the investment climate’

• The anti-corruption National Accountability Bureau has been neutered. It estimates that over US$1.8 billion has been lost due to immunities granted to politicians and bureaucrats and the restrictions placed on the anti-corruption body

• The subordinate judiciary is seen as notoriously inefficient and corrupt. It lacks independence and contributes to a general culture of impunity

• The Auditor’s Department is not independent and is frequently accused of collusion with those it audits

• The Public Accounts Committee is ineffective and largely ignored

8. In the absence of reliable accountability systems within the countries to which we give budget support it is difficult to understand how the independent aid watchdog will be able to scrutinize the expenditure. One could seek to have special agreements in place that would allow the watchdog special access to Government systems, but in the absence of reliable data and systems the task would be a gargantuan and extremely expensive one.

Evaluation of funds given to third party organisations

9. Funds given by DFID to UN organisations are also difficult to monitor. This is partially because the activities are often run badly and partially because they can hide behind UN immunities when it comes to investigations. For example the USAID Inspector General identified gross mismanagement of a UNDP $25.6m USAID-funded quick impact project [1] . The UNDP misspent millions of dollars and delivered shoddy work, including a bridge at risk of collapsing. $7m was withdrawn by UNDP after project closure and diverted to other countries. Despite a court order the UNDP refused to answer questions, citing UN immunities.

10. Adam Smith International undertook an evaluation of UNDP programmes in Somalia. We found that there was insufficient data to monitor outcomes. Our report concluded:

"UNDP Somalia is obliged by UNDP headquarters to use a web-based management information system called ATLAS. UNDP kindly gave us a demonstration of the system. Quite apart from its complexity it seems to take several minutes to load each page. From our brief review we noted that much data was missing from it – presumably because staff find the system too burdensome to use. It is clear that this system is so complicated and difficult to use that it is not a viable monitoring or management tool. We put this point to UNDP Somalia senior management who agreed with us."

"The weakness of M&E (systems and operation) has been a fundamental flaw. Quite simply one does not know whether activities and outputs are effective, lasting and of good quality, and how much progress is being made relative to intended outcomes. [2] "

11. Other evaluations have come to similar conclusions. For example the UNDP evaluation entitled "Assessment of Development Results in Afghanistan" [3] stated that:

"Monitoring and evaluation of UNDP programmes for the period 2002 to 2008 was limited by the following:

· Constant redefinition and repackaging of expected outcomes in keeping with changing requirements under corporate outcome indi cators contained in documents such as the Multi-year Funding Framework and the Corporate Strategy

· Failure to develop clear criteria and indi cators for monitoring the achievement of programme outcomes on a regular basis

· Failure to establish a baseline and to collect data to monitor results at the outcome level (activities and outputs are faithfully monitored and reported on)

· Duration of the country programme and projects, which is too short for the attainment of results at the level of outcomes.

· Tendency to state expected outcomes in broad and sometimes grandiose terms that tend to render them less useful

· Absence of a facility independent of project funding that is able to monitor performance beyond the limited duration of projects and at the level of outcomes and impact.

" Monitoring remains heavily output and activity based, and these are meticulously recorded. Outcome-oriented, results-based monitoring and evaluation systems are not yet in place, and no baseline data has been collected to establish the basis for such monitoring."

"Because of the rigidity of ATLAS, it has also become difficult to obtain definitive, reliable and timely figures on expenditures. This is disturbing for an organization that predicates its existence on the efficient management of funds."

12. UNDP’s own evaluation systems cannot be relied upon by DFID. Internal M&E is very poor as discussed above, and (with some exceptions) most external evaluations of UNDP activities are done by those who are close to the organization and are likely less critical than is merited. It is largely those evaluations that are done by external agents appointed by another donor that have been the most critical. [4]

13. Again it is difficult to see how the UK’s independent aid watchdog, even if the UN agreed to waive its immunities and allow independent investigation, could realistically undertake effective evaluation of UNDP activities in the absence of base data.

14. Multi-donor trust funds to which DFID contributes would also be very difficult to evaluate. These are often very slow and inefficient. Because of their multiple parentage there tends to be a lack of accountability. For example the Financial Times earlier this year reported that after more than four years of $524m committed to the South Sudan Multi-Donor Trust Fund (MDTF) $323m is unspent. ODI commented on the Sudan MDTFs that: "these World Bank-administered funds were supposed to ensure coordinated external donor support through to 2011. In practice, they have failed to achieve visible impact (Scanteam, 2007). The rate of disbursement has been excruciatingly slow, with most projects failing to deliver ‘tangible goods’ to the public even by year two (Fenton 2008). Bureaucratic World Bank procedures, staffing problems and protracted negotiations on who does what hindered initial implementation (Pantuliano et al., 2007). The government’s inability to cope with the bureaucratic requirements caused serious delays and inefficiencies." [5] Evaluation of such an inefficient entity would not only require special permission but would also be very difficult.

15. DFID gives considerable funds to the World Bank. The World Bank does have a robust internal evaluation department that conducts thorough backward-looking evaluations. But these are not performed on all programmes by any means. The vast bulk of WB programme spending is handled by developing country governments. Special programme management units are established to handle the funds. It takes a very, very long time for anything to happen. DFID might contribute funds but it could take several years for the activities to start and the funds to be spent. Monitoring and evaluation of projects as they progress is often either very thin or absent. Moreover, although many World Bank staff members are capable and smart they are spread far too thinly to be able to supervise programmes effectively. Programme management units are often very inefficient and administration is of poor quality and bureaucratic. Again this does not create a very propitious environment for external evaluation by the UK aid watchdog.

Conclusion

16. In conclusion we stress again that the need for robust independent evaluation is much greater in relation to the roughly 50% of DFID funds that are given to other organisations to spend. Evaluation of this spending will be much more difficult to achieve. If it cannot be achieved in respect of some of that funding then it is questionable whether the funding should continue. There is a danger of perpetuating and entrenching two separate classes of DFID spending: First Class, involving well-supervised and monitored directly committed funds that are subject to robust independent evaluation, and Second or Third Class, poorly supervised and poorly monitored indirectly spent funds that are not subject to robust independent evaluation.

Reduction in DFID’s administration costs

17. It is clearly preferable that DFID’s administration costs should be no higher that is necessary to carry out DFID’s functions effectively. If those costs can be reduced through greater efficiencies while still maintaining effective delivery of DFID’s functions then so much the better. However, if administration costs are reduced to such an extent that the quality of DFID’s work is threatened, then this would be a serious error. Further the debate on DFID’s administrative costs should be undertaken with sight of the cuts and saving already made over the last 3 years at a time when DFID’s budget was expanding rapidly. In 1997 the DFID budget was £2b with 1,600 staff. In 2004 it was £4b with 3,000 staff. Now it is almost £8b with 2,400 staff.

18. Since the total amount of DFID’s budget is not being reduced but rather ring-fenced, then reduction of administration costs does not help to cut the budget deficit. Rather it means that a slightly larger quantity of resources goes into programme spending. While quantity of aid spending is of course significant, quality is paramount. It should be noted that DFID’s aspirations to provide global leadership on a range of development issues, and to ‘punch above its weight’, are dependent on DFID’s ability to retain and further enhance its reputation for quality. It seems evident that the political and policy underpinnings of the decision to ring fence ODA have their basis in maintaining quality. If quality can be improved, so much more can be delivered. There is a distinct danger that the drive to reduce administration costs could have a very negative effect on quality and delivery of effective and efficient development aid. In short cutting administrative costs could very easily be a false economy.

19. Reduction in administrative costs can take various forms. For example, it could mean:

· Better use of existing resources.

· Reduction in DFID permanent staff headcount

· Transfer of funds to other organisations

20. If we look at each of these in turn we can see quite different effects. For example better use of existing resources could be achieved by sharing space overseas with the FCO, reducing telecommunications costs through use of free services such as Skype, or reducing office space usage and leasing out spare offices. Spending on public relations activities with the UK also falls into this category of cost reduction that can occur without harming frontline development activity. Such types of cost reduction do not affect delivery and are clearly beneficial.

21. By contrast reduction in DFID permanent staff headcount can clearly have a very damaging effect. If there is an administrative function, for example ordering stationary and other supplies, that can be carried out just as efficiently either by fewer DFID staff or by contract staff then obviously the reductions should be made or the function contracted out. However, arbitrary targets for overall reductions in DFID headcount can be set as some kind of political virility symbol, irrespective of the effects. This was certainly the case under the last administration.

22. To carry out its functions well DFID requires sufficient numbers of high quality staff, particularly in the field. Good programme and project design and strong supervision is essential to success. Things often go wrong in developing countries. Experienced and politically astute staff are needed to keep activities on track. As providers of technical assistance we would stress that we cannot do it all on our own. We need strong, experienced DFID staff to work with us. We need cover from time to time on certain points and government counterparts need gentle/not so gentle pushes from DFID - acting with the weight of HMG - on others. Further as projects evolve over time and the context changes it is really important to keep for DFID staff to remain engaged on what is happening and that occurs most effectively when there is a strong link between the advisers and DFID. Such strong links are more, rather than less, important in conflict affected contexts.

23. One suggestion that is made from time to time to reduce the DFID headcount is to contract out the job of managing groups of programmes and projects to private sector consultants. This can sometimes work where the activities are very straightforward but in other cases there are real difficulties. There is a dilution of DFID’s authority and a lack of clarity as to respective responsibilities of contractor and DFID. When a problem arises, usually with how a counterpart is acting, DFID itself will need to intervene. Such problems occur on even the best-managed projects. DFID’s intervention will likely carry the weight that is required, whereas that of another set of consultants will likely not.

24. Moreover, high quality contractors are interested in implementing programmes and achieving results, rather than fulfilling largely administrative roles. Their reputation stands on the basis of the results they achieve and how successful they are in implementing their projects. They will make an effort to identify problems and resolve them. On the other hand a contractor involved in an administrative role of supervising others can stand back and blame the other contractor if things go wrong. Their stake in success is heavily diluted.

25. If the contract is one in which the lead contractor largely passes through revenues to other contractors it has procured, then there are difficult cash flow issues for the lead contractor as well as the sub-contractors. If DFID pays in roughly 30 days then the lead contractor will have to bear the cost of settling the sub-contractors bills before it gets paid – or alternatively only pay the sub-contractors when it gets paid. If the latter then it would take the sub-contractors likely 50 to 60 days before they are paid. This would place heavy pressure on their cash flow which they would likely seek to offset by not paying the individual consultants until they had been paid. Most good consultants will not work on that basis, as one cannot guarantee a reliable cash flow. Thus one is likely to only to be able to recruit poor quality advisers who cannot get reliable work elsewhere.

26. In reality, when it comes to design and supervision of technical assistance programmes and projects, as well as other DFID overseas activities there is no real substitute for having enough experienced DFID permanent staff to handle the workload. If these are too thinly spread to handle the workload, then quality will undoubtedly suffer. The consequent reduction in impact will more than offset any saving made in staff salaries.

27. By far the worst form of administrative cost reduction is the transfer of resources to other organisations. The pretence here that resources spent on administration are being reduced is entirely false. If DFID writes a cheque to another organisation, whether it is a UN organisation, the World Bank or a developing country government, it is deceit to suggest that all of this goes on programme expenditure and none on administration. The truth is that most of these organisations generally spend much more on administration than DFID and are much less effective at achieving results. In some cases the bulk of the funds are wasted. But since no DFID administrative costs are involved it can be claimed that a reduction in administration has been achieved. This paints an entirely inaccurate picture of the reality.

28. Our recent report ‘Strengthening Pakistan [6] " points out how a presentational drive to reduce administration costs actually leads to a much more ineffective programme in that critical country:

"An increase in the UK Government’s staff presence in Pakistan is required in order to support a [more effective] programme. More analytical firepower is needed on the ground, which has been resisted in the past for presentational reasons in order to keep the ‘administration budget’ low in relation to the ‘programme budget’. Budget support can be done by a very small team, but unfortunately achieves little…. Writing a larger budget support cheque to accommodate additional financing would be very simple – it would look good in keeping the administration budget down, but would be highly unlikely to leverage any change. This type of spin is pointless and indeed very harmful when critical national interests are at stake. The British Government and others need to put in the administrative and analytical resources to make an effective programme work".

29. In summary, the UK Government needs to put in the administrative resources necessary to implement a high quality aid programme. That is both in our national interest and in that of the recipient countries. While some efficiency savings can undoubtedly be made without damaging quality, if the DFID overall budget rises it is critical that sufficient resources are retained in the DFID administration budget to ensure that quality can be maintained and improved in the delivery of frontline services.

DFID’s use of technical assistance and consultants and whether reducing this expense could be an appropriate target for the new Government

Overview

30. Adam Smith International provides technical assistance, some of which is financed by DFID and some financed by other donors, including the World Bank, Asian Development Bank, Inter-American Development Bank, the European Union, the Swedish International Development Agency, Danida, CIDA, AUSAID etc.. We also provide advice and assistance to a wide range of governments who pay directly for our services. We have conducted evaluations of technical assistance programmes financed by a variety of donors. We are thus familiar with technical assistance (TA) as well as with other aid modalities.

31. Our view is that technical assistance, when provided effectively, can have a hugely positive effect. It is a form of aid that can have a major impact out of all proportion to its cost and help achieve transformational change. As examples we cite below demonstrate, the return on investment in well-designed and well-delivered TA can be very high indeed.

32. A distinction needs to be drawn between consultants who work within DFID to help it carry out or improve its functions and those who work as part of DFID technical assistance to developing countries. Consultancy work that is internal to the department is similar to that performed in other departments of state, some of which has attracted unfavourable comment because of its cost. Consultancy that forms part of technical assistance to developing countries is entirely different and should not be confused with the former, internal consultancy. Technical assistance – the provision of policy and programme support, the transfer of expertise, and the building of capacity - is one of the main vehicles by which DFID policies and programmes are delivered to countries that receive overseas development assistance (ODA).

33. Questions relating to the desirable level of consultants working within DFID are best answered as a subset of the issue of how to handle DFID’s administration costs. Issues that would need to be addressed would include:

· the cost of consultants versus the full cost of permanent members of staff

· the extent to which those consultants bring specialist skills that cannot be found within DFID’s permanent staff, and or fill temporary roles

· the effectiveness of the consultancy services provided

34. In this note we address the question of technical assistance and whether DFID should provide more or less of it. We think it is most useful to describe a number of cases of DFID-financed technical assistance that we have provided. Below we set out examples from Afghanistan, Iraq, Zimbabwe, Nigeria and Sierra Leone.

35. One also needs to consider the fact that any decision on levels of TA cannot be taken in isolation but should be taken in light of the efficacy of DFID’s other main instruments for delivering ODA and, indeed, in light of new policy thinking on the aims and objectives of ODA.

36. On the former:

· DFID has a well developed system for managing and monitoring performance, securing vfm and evaluating impact of TA;

· DFID TA is contracted after a rigorous, merit-based process of public procurement;

· It can be seen from recent evidence that such systems are not in place and/or consistently applied for direct budget support or the funding of many third party organisations.

37. On the latter it should be noted that TA is often used in higher volumes in conflict affected countries both as a means of building and strengthening institutions, supplementing weak capacity, providing civilian stabilisation and reconstruction support and providing advisory services directly to counterparts in government.

Case Studies [7]

38. The Strengthening National Provincial Tax Administration in Afghanistan project, now in its second phase, originally started in 2004. It is credited with playing a major role in reforming Afghanistan’s tax administration system and providing a reliable and expanding source of revenue for the Afghan Government.

39. Key achievements include:

· Major increases in revenue collection, with all IMF targets in 2009 being exceeded. Revenue for the first six months of 2009 increased by 60% compared with same period in the previous year. Income tax collection rose from US$13.7 million in 2003 to US$267.9 in 2008 – a twenty-fold increase

· A new tax policy has been put in place, modern tax laws and regulations passed, tax administration staff recruited and trained, and Taxpayer Offices established.

· These reforms, while being consolidated in Kabul, are also being rolled out to priority provinces in key economic centres around Afghanistan.

40. By helping the Afghan Government increase substantially the revenue it raises the project is contributing considerably to both creating a viable Afghan state and the conditions in which the international community can greatly reduce its involvement.

41. In Iraq we have been helping strengthen centre of government institutions and address key finance issues since 2004. A number of witnesses have described the impact of that work to the Chilcot Inquiry. Tim Foy was head of DFID Baghdad:

TIM FOY In terms of the quality of consultants which we were able to engage to work for us in terms of the machinery of the centre of government, the Prime Minister's office, to work within the finance department, I think we would be hard pushed to have got better people. They were people that gave us fantastic leverage with the Americans, and where we punched genuinely above our weight, it was the quality of expertise that we were able to bring in. The Americans might have outnumbered us, but in terms of quality, I think there was a great deal of difference, and that brought us an awful lot of kudos. It brought us that access that I spoke about, the ability to engage at the highest level. I am amazed that it was the UK that basically got the standby agreement with the IMF, resolves Iraq's debt problems. It wasn't the United States. It was about half a dozen UK consultants that did it.

THE CHAIRMAN: And that is a very big thing, in terms of the numbers and significance.

TIM FOY: It was a big thing, and we should be quite proud of that.

42. Mark Lowcock Director of DFID's Bilateral Programmes, described ASI’s work in a personal assessment at the end of his evidence to the Chilcot Inquiry:

MR MARK LOWCOCK: I can give you a personal assessment of what are the biggest impact things we have done.

THE CHAIRMAN: Yes, we would like that.

MR MARK LOWCOCK: My personal assessment is that getting the macro-economy right and enabling Iraq to manage its growing budget effectively and enabling Iraqis to run their own affairs by better co-ordination at the centre of government level, more effective process in the Council of Ministers, all those things, in my assessment, are, you know, perhaps the most important thing to do for the long-term goal of building a capable state in Iraq able to, you know, look after itself.

THE CHAIRMAN: Thank you.

43. Similarly, when asked by the Inquiry to nominate an example of success, Christopher Prentice, who was Ambassador from 2007 to October 2009, cited ASI’s work at the centre of government:

SIR MARTIN GILBERT: "Were you able to see specific achievements by November when you left?

MR CHRISTOPHER PRENTICE: Well, certainly the longer-term projects at the centre, in particular, as was mentioned this morning, the capacity building in the Cabinet Secretariat was really beginning to show results.

This was a very quiet project, which I think was not widely known amongst the Iraqi politicians whose interests it was serving, and all the better for being below the surface because it was so central to the government machine, and it was one which was very much hands-off. It was DFID's working through Adam Smith International, who were providing consultancy for the Iraqis and that was confidential to the Iraqis and it was - it could have been a delicate matter, but actually was handled quietly and successfully and – by the time I left, the Cabinet Secretary had got to the point of being able to -- already had started transition planning for the handover of government after the elections, now due in March, pulling together deputy ministerial level representatives across their equivalent of Whitehall, to draw some lessons from this period of government in terms of the structure of government for presentation to the new Prime Minister, when elected. That's a fairly sophisticated operation and was really, I think, an example of success." [8]

44. In Zimbabwe ASI is working to support the Prime Minister’s Office. The scope of the work is described in an International Development Committee report of 26 March 2010:

"DFID’s support was designed to enable the Office of the Prime Minister to perform the normal functions that a head of government’s office would usually undertake, including oversight of the budget, ensuring ministries followed through on the Government’s agreed work plan and helping to resolve disputes between government departments."

45. The Prime Minister has found our support very useful, commenting in a letter to DFID that "The assistance we have received from Adam Smith International has been extremely valuable."

46. The International Development Committee report came to the same conclusion, stating:

"DFID’s support for the Office of the Prime Minister has clearly been valuable and has been much appreciated by him. We recommend that DFID extend this support for a further period. This will enable progress achieved to date to continue, in terms of strengthening the capacity of the Office to oversee the smooth running of government departments and implementation of policy, and to lead on executive business in Parliament."

47. Our work to strengthen infrastructure in Nigeria is another example of the ability of technical assistance to achieve transformational change. The Nigeria Infrastructure Advisory Facility (NIAF) has also received favourable mention in a report by the International Development Committee. [9] In addition to its work to improve transport in Lagos another important success has been the creation of a policy and institutional framework for PPP. Mansur Ahmed, Director General of the Infrastructure Concession Regulatory Commission said "without NIAF’s pragmatic, flexible and high quality assistance Nigeria would not be nearly so advanced in its PPP programme".

48. However, the main project focus is now on electricity sector reform. Absence of adequate power supplies is probably the most significant barrier to economic growth in Nigeria. Our recommendations of how to tackle the problems have been accepted by the Nigerian government and power sector reform is now the Nigerian President’s highest priority. The potential gains are huge. The Nigerian Government estimates the cost of unserved energy to be (in terms of lost GDP) 20 trillion Naira per year (circa USD$ 130bn). This should be contrasted with DFID’s total 2009-10 budget of $9.9bn. The impact in terns of poverty reduction even if our project is only partially successful will be immense.

49. In Sierra Leone ASI has helped transform the legislative and institutional framework for the minerals sector. The project completion report, written by independent evaluators, stated:

"ASI has carried out an exemplary project in terms of their management, technical competence, engagement, and commitment. GoSL has high regard for the quality of the work done. The outputs of the project are clearly evident in the publication (following full civil service scrutiny through the Attorney Generals Office, and often engagement of Parliamentary draftsmen) of two new Acts, Bills, Regulations and Guidelines totaling more than 500 pages of official text.

Some components of the work, including revisions to the Mines and Minerals Act and work on the critical area of regulation and support for Artisanal and Small Scale Miners should be disseminated as examples of good practice internationally." [10]

50. The mining sector has the capacity to serve as the engine of economic growth if effective and efficient Government regulation of the sector is introduced. A conservative assumption is for extra GDP growth of 1.2% per year if supported by a growing mining sector. This means that by 2020 some 900,000 people could be lifted out of poverty due to mining industry growth. The increases in Govt revenue from the mining sector could fund within 3 years a doubling of the Govt direct financing component of the Development Budget or the employment of an additional 30,000 health care professionals.

Conclusion and recommendations

51. It is clear from the examples above that technical assistance can have a very substantial impact. We could provide many other examples, such as the ASI project cited by the Secretary of State in his evidence to the Committee on 15th July 2010. [11]

52. Yet this does not necessarily mean that all technical assistance is of a good quality. Some is of an atrocious quality. One thinks for example of TA implemented directly by UNDP. We conducted a recent evaluation of technical assistance provided by UNDP in Somalia [12] and found that not only was it poor it was next to impossible to establish whether anything was being achieved. Other evaluations of UNDP TA have come to similar negative conclusions. [13] The reasons why UNDP-implemented TA is ineffective are not hard to find. It involves funds being spent on a non-competitive basis by a relatively unaccountable international bureaucracy which has created lots of complicated internal rules to govern its operations and provides few incentives for success.

53. The factors which make technical assistance more effective can be readily identified:

• A relatively short time taken from production of initial project concept note to start of implementation, so that activities remain relevant and political opportunities can be acted on promptly

• Programmes and projects are designed to include sufficient resources to accomplish the designated tasks, including sufficient resources for project management. Resource levels can be easily and speedily adjusted.

• A high quality of programme and project design

• An efficient procurement process, that is relatively non-bureaucratic, does not impose multiple requirements on bidders and is focused on a value for money rather than merely on cost

• Sufficient competent, capable and adequately remunerated procurement officials to handle procurement and administration to avoid delays or this taking up the time of top officials

• A credible procurement process in which very few calls for assistance are abandoned

• An efficient payment process with payments made on a timely basis to contractors

• A high quality of project supervision, conducted by donor agency staff properly trained in appropriate skills and subject areas, and remunerated sufficiently to enable good staff to be recruited and retained

• Donor staff able to take decisions efficiently and speedily, with sufficient delegated authority to move quickly

• Flexibility to change a programme or project, once started, in light of changed circumstances

• Effective regular monitoring and evaluation

• An effort to encourage and motivate high quality suppliers

54. DFID is currently engaged in upgrading its procurement function. A number of improvements have already been implemented. To build on this we recommend that to achieve greater impact with its funding of technical assistance DFID should:

a) reduce the amount of money it passes to inefficient procurers of technical assistance, and expand the amount it contracts directly, thus getting better value from its own procurement and management staff

b) seek more forcefully to draw in technical assistance funds from other donors, thus again getting better value from its own procurement and management staff by handling a greater volume of funds

c) Look to make further improvements in its procurement and management of technical assistance, conducting a thorough study of other donor practices to identify good ones that could be adapted and developing innovative new approaches

d) Identify more aggressively areas where technical assistance can make a substantial impact – and increase support to those areas.