EUR-Lex Access to European Union law

Back to EUR-Lex homepage

This document is an excerpt from the EUR-Lex website

Document 32000Y0419(01)

Special Report No 4/2000 on rehabilitation actions for ACP countries as an instrument to prepare for normal development aid, accompanied by the Commission's replies

OJ C 113, 19.4.2000, p. 1–32 (ES, DA, DE, EL, EN, FR, IT, NL, PT, FI, SV)

In force

32000Y0419(01)

Special Report No 4/2000 on rehabilitation actions for ACP countries as an instrument to prepare for normal development aid, accompanied by the Commission's replies

Official Journal C 113 , 19/04/2000 P. 0001 - 0032


Special Report No 4/2000

on rehabilitation actions for ACP countries as an instrument to prepare for normal development aid, accompanied by the Commission's replies

(pursuant to Article 248(4), second subparagraph, EC)

(2000/C 113/01)

CONTENTS

>TABLE>

Summary

In 1993, a concept was established for the rehabilitation of countries after situations of war or natural disaster. Rehabilitation should be the bridge between the humanitarian aid and the normal long term development aid. The facility planned to be temporary became almost permanent with actions taking much more time than expected. At the end of 1998, decisions under the accumulated programmes, financed by the European Development Fund and the budget, amounted to ECU 1065 million. Payments were only ECU 486 million which indicates slow implementation.

Although rehabilitation actions have led to certain physical results, neither the overall impact nor the impact at the level of beneficiary countries is known, either because no comprehensive information was available on what had been achieved or because the objectives of the actions were in many cases not clearly defined.

In general, the programming and planning was not the result of a thorough examination of the needs and possibilities of rehabilitation actions. In a number of countries rehabilitation actions were launched when the condition of relative stability was not yet met. Consequently, the sustainability of actions was not ensured. Often the actions were the continuation of already existing humanitarian aid actions. Whilst such actions had their value, it was not always clear to what extent the rehabilitation actions were different from humanitarian aid.

The coordination of rehabilitation programmes and actions with other donors (Member States, United Nations) was mainly limited to exchange of information. Whereas the content of such coordination is to an extent beyond the Commission's control, the Commission should have created firmer coordination between its own services.

Monitoring and reporting were not adequate. During the implementation of any action, it was almost impossible to follow the progress on the basis of reporting on the activities and the expenditure. It was hence extremely difficult for the Delegations to intervene actively in the process. This was aggravated by the fact that the Delegations were not adequately staffed to cope with the additional workload of the rehabilitation actions.

Despite the limited information available, the Commission should carry out an in-depth evaluation on the experiences with the rehabilitation instrument so far in order to decide whether and in what form rehabilitation programmes should be continued.

Introduction

Background

1. In 1993, in the aftermath of the conflicts in Ethiopia, Eritrea, Mozambique, Angola and Somalia, a special instrument was created to assist in rehabilitating these countries. At a later stage other countries were included in the programme (see paragraph 16).

2. Rehabilitation in the context of development aid has been defined as the intermediary stage between a situation of disaster or emergency and a stable situation where normally planned long-term development activities can take place. According to the Commission, a rehabilitation programme is "a link between humanitarian aid and the resumption of support programmes for long term development aimed at developing countries which are emerging from a crisis or war or whose economy has been seriously affected by other factors"(1).

3. Rehabilitation programmes were also executed by the European Union prior to 1993 under the Lomé Convention, but it was not until 1993 that the concept of rehabilitation was given its own framework, and in 1994 specific provisions were also made under the general budget. The new concept was established under what is known as the "Special Initiative for Africa (SIA)"(2).

4. The programme started on 25 May 1993, the date of the Council conclusions, and is still continuing. In a resolution of 4 November 1993(3), the European Parliament globally approved the Commission's communication. At the end of December 1998, decisions under the accumulated SIA programmes amounted to ECU 1065 million (both European Development Fund (EDF) and budget). Payments to this date are limited to only ECU 486 million, indicating very slow progress compared to the anticipated quick implementation. The financial situation is summarised in Tables I and II.

Table I

Decisions and payments for EDF and budget by countries

>TABLE>

Table II

Decisions and payments for EDF and budget in the period 1993 to 1998

>TABLE>

The Court's audit

5. The audit was carried out with the objective of examining, verifying and assessing:

(a) that there is a clear policy forming a coherent link between humanitarian aid and development aid and that the use of different European Union funding instruments is also coherent;

(b) that the identification of eligible countries and of actions is justified and coordinated with other donors, in particular with the European Union Member States;

(c) whether the procedures and systems ensuring a systematic and permanent monitoring put in place by the Commission are efficient and transparent;

(d) that actions were implemented over a limited time span and accounted for in a transparent manner;

(e) that the results are lasting, improve the local economy and institutions and help in establishing political and social stability and so provides a basis for normal development aid.

6. The audit was carried out at the level of the central services in Brussels and in a number of countries (Ethiopia, Eritrea, Mozambique, Madagascar, Zaire and Haiti). Initially the Court concentrated on the actions financed by the European Development Fund. When it became clear for a number of ACP countries that an important part of the rehabilitation actions was financed through the budget, these actions were also included to be examined in the context of this audit.

The framework

The concept

7. The concept of rehabilitation as laid down in the Commission's communication was further developed by the European Parliament in a resolution of 4 November 1993. The Council, in its conclusions of 2 December 1993, endorsed the continuation of the rehabilitation programme. However, the Commission had not made a synthesis with a view to lay down a clear conceptual framework for rehabilitation.

8. After some years of experience with the rehabilitation programme the Commission produced a document on linking relief, rehabilitation and development (LRRD) as the cornerstone of the SIA concept(4). In this document, which was endorsed by the Council conclusion of 28 May 1996, the Commission set out proposals and recommendations on how the links could be strengthened. A pilot exercise was recommended, together with those Member States interested, to prepare global policy frameworks for a number of countries and regions prone to political or economic risks and tensions or natural disasters. Task forces were to be established both centrally in the services of the Commission and locally in the relevant countries, for the countries/regions chosen for the pilot phase. For that purpose Angola, Liberia, Haiti, Somalia and the Great Lakes region were identified.

9. So far, it has been difficult to identify the real outcome of this effort. In particular the link between emergency actions already undertaken by the European Community Humanitarian Office (ECHO) and rehabilitation under SIA should of course have been established, but the audit only identified three projects in Somalia and some in Haiti which were started by ECHO and taken over under SIA. Although it is clear that some task forces have been working, the Commission has not yet presented any comprehensive description/documentation on the results.

The legal framework

10. As already pointed out in the Court's special report on humanitarian aid(5) in relation to Council Regulation (EC) No 1257/96 of 20 June 1996 on humanitarian aid(6), no overall regulation, be it for humanitarian or rehabilitation aid, has yet been enacted. Decisions and procedures are functions of the origin of the financial resources and the internal organisation of the Commission and not of the nature of the actions.

11. On the basis of the two documents prepared by the Commission and the Council in May 1993 (see paragraph 3), actions managed within the framework of the Lomé Convention were funded from the EDF, although without any specific guidelines on rehabilitation. Actions financed through the budget were initially covered by existing regulations, which did not specifically address the issue of rehabilitation. Much later the Council Regulation (EC) No 2258/96 of 22 November 1996(7) was established for rehabilitation and reconstruction operations in developing countries financed by the general budget. Consequently, this range of sources of funding, each with its specific rules on management and execution, meant that no coherent approach within the budget and between the general budget and the EDF could be achieved (see paragraphs 42 to 46).

Financial resources

12. The financial resources determined by the Council in its conclusions of 25 May 1993 indicated that a minimum of ECU 100 million should be provided from existing EDF funds immediately. This amount of ECU 100 million was then announced as available from within the EDF during an initial planning stage with non-governmental organisations (NGOs) (see paragraph 31). The proposals then made by the NGOs far exceeded ECU 100 million. However, most of the projects which could not be contained in this first tranche were financed at a later stage from existing sources of funds (EDF or general budget). The total commitments (EDF and budget) after five years amounted to just over ECU 1000 million.

13. Although it was originally intended that Member States would make specific funds available to the programme, this was never realised. In certain SIA countries Member States have executed their own bilateral rehabilitation programmes. This situation presents further complications in ensuring overall coordination, planning and execution of programmes in a coherent manner.

14. In conjunction with its decision of 1993 on the ECU 100 million initially made available, the Council also concluded that the Commission should urgently evaluate the resources needed. However, no such evaluation was made. Nevertheless, additional funds were made available for amounts which could not be rapidly absorbed. In particular in Angola and Mozambique many programmes and projects were decided on and contracted for with budgets much higher than necessary for the measures planned, either because of problems in actually implementing some of the actions or because of poor cost estimates. There was a strong tendency for the implementing partners to suggest extensions and, in some cases, instead of closing programmes and projects, new activities were financed from the remaining funds.

15. For most of the countries there was already available a relative abundance of funds within the EDF to be used for rehabilitation actions under SIA. Nevertheless the Commission has continuously added funds from the general budget without first analysing whether the funds under the EDF could be utilised (see Table III). In the national indicative programme (NIP) for Eritrea for example, 20 % or ECU 7 million of EDF funding was allocated to be used for rehabilitation purposes. In practice none of these funds was used for this purpose but a further ECU 12 million was allocated from the budget.

Table III

Addition of budget funds although NIP funds were still available

>TABLE>

Overall programming

Eligibility

16. In an Annex to the Commission's proposal(8) 10 countries, which were by May 1993 likely to be or to become eligible for the programme were listed (Somalia, Ethiopia, Eritrea, Mozambique, Angola, Rwanda, Zaire, Liberia, Haiti and Sudan). The first five of these countries were included in the first part of the programme, while the next four were included later. Sudan has so far not yet been subject of the programme. In addition, Madagascar was included in 1994 after being severely affected by a tornado, while Uganda, Burundi, Malawi, Tanzania (refugees in the Great Lakes region), Sierra Leone and Namibia were added to the programme successively.

17. For each of the 10 countries included in the Annex, background information was given, by which the Commission presented crucial baseline information to be considered in connection with decisions on the eligibility of the countries.

Criteria

18. The Commission has set the following conditions for a country to be eligible for assistance under SIA:

(a) a request has to be made by the authorities;

(b) there must be an ongoing state of economic crisis or serious deterioration of basic infrastructure;

(c) hostilities or upheavals must have ended;

(d) the authorities must be willing to initiate an economic and social transition process which respects democratic principles, human rights and takes steps towards a structural reform.

Moreover, according to the proposal, countries receiving structural adjustment support would not be eligible since being recipients of such financial support indicates that they have already successfully completed the rehabilitation stage.

19. While all countries fulfilled the criterion of "economic crisis" many of them failed to meet one or more of the other criteria. A number of countries were still in a state of emergency. To implement rehabilitation programmes in such situations carry with it the risk of significant delays and even the failure of actions.

Somalia

20. For Somalia, the Addis Ababa agreement of March 1993 between the main factions should have paved the way for a political settlement. The rehabilitation initiative was implemented immediately and has been ongoing since 1993 although the political settlement is still not in place after six years. The security situation has however made operations difficult and no activities (except emergency operations) have been possible in the Mogadishu area. The lack of recognised governments in some of the clan-controlled areas makes transparency and accountability problematic and there are doubts as to the sustainability of the actions undertaken. Commitments under SIA for Somalia amounts to ECU 95 million as at the end of 1998.

21. In November 1997 a series of incidents in Middle Shebell region in Somalia led to a decision to make the region no longer eligible for rehabilitation activities. Expatriates left the area and all investments in infrastructures and supplies (for which ECU 440000 had already been paid) stopped. There were temporary suspensions of five projects in the health sector over the period of June 1994 to March 1998. One of the projects was terminated and the NGO withdrew permanently since the regional "ministry" was not cooperating as envisaged (ECU 322000 paid). In the other four cases the suspension was used "as a tactic in negotiating with Somalia and in strengthening the negotiating power of the outside agencies"(9).

Angola

22. Although it was hoped in 1993 that new negotiations would create a peaceful situation, the environment for rehabilitation in Angola has continuously been problematic since, and is still far from satisfactory. All the programmes, the old ones as well as the new ones under SIA (amounting to some ECU 150 million), have suffered from severe delays and suspensions. Instead of embarking on SIA, humanitarian aid should have been continued.

Democratic Republic of the Congo (formerly Zaire)

23. For the Democratic Republic of the Congo, the Commission proposal stated that the administration has been paralysed since 1990 and that the country is totally in decay. The information note(10) ends: "However, before a rehabilitation operation can begin several conditions must be satisfied and in particular there must be a political compromise between the President and the opposition, so that the process of democratisation can continue and an emergency programme of economic recovery can be launched".

24. Despite this the Commission decided to initiate actions for the Congo, which involve commitments of more than ECU 200 million. The environment as confirmed by the Court during its audit visit in October 1996 was definitely not conducive to SIA at that time. All activities in eastern Congo had to be stopped by the end of 1996 because of security problems. A large programme, started in 1996, had also to be abandoned by the end of 1996, when the entire budget of ECU 9 million for technical assistance had been used without the preparation being completed.

25. The Commission immediately after the take-over by President Kabila restarted the existing programmes under circumstances when it was completely unclear whether criteria such as sustainability or cost-effectiveness could be met.

26. A new programme of ECU 30 million was to support forthcoming elections in the Congo after the change of power. When the proposal was brought to the EDF Committee in March 1997, several Member State representatives expressed doubts and questioned the situation in respect of democratisation. At the same time the Commission stated the need to have these funds available to be used immediately when they were needed. As war is still going on, elections are unlikely to take place soon. The relevance of the proposed programme and its urgency at the time of the EDF Committee meeting has to be questioned. This was later realised by the Commission and the programme was closed in August 1999.

Rwanda/Liberia

27. For Rwanda, for which an amount of ECU 153 million has been decided, the conditions for rehabilitation were not met, i.e. a minimum of security was not assured and no administration capable of realising the actions was in place.

28. Despite the difficult situation in Liberia, the actions undertaken by the Commission since 1994 have been rather successful. This is to a great extent the result of very special organisational measures taken at the level of the headquarters and in Liberia to set up and implement actions. By creating one strong focal point in Brussels and one strong focal point in Liberia, quick reactions were possible in each of the stages of the peace process. The Liberia experience is rightly considered by the Commission as a pilot for rehabilitation programmes.

Ethiopia

29. For Ethiopia a huge reconstruction programme backed by all possible donors had been in operation since 1991 under emergency aid. The Commission concluded in its proposal that Ethiopia's requirements for rehabilitation and reconstruction were thus already covered by that programme. Furthermore Ethiopia became eligible for structural adjustment aid in 1992. Despite the above factors, Ethiopia received aid under the SIA amounting to some ECU 25 million.

Mozambique/Haiti

30. Mozambique should not have qualified for SIA programmes (over ECU 170 million) as it continued to receive structural adjustment aid. Although Haiti had initially no structural adjustment support, it started to receive it in 1995. Despite this Haiti has continued to receive SIA support ever since.

Considerations over the programming process

31. Almost immediately after the Council's approval of the SIA (mid-June 1993), the Commission held a meeting with representatives of the European NGOs to inform them about the initiative and ask them to come up with ideas and inputs on a country basis. Following that, meetings were held individually for the five countries chosen initially (Angola, Mozambique, Somalia, Ethiopia and Eritrea), during the first half of July. In these meetings a restricted number of NGOs (chosen by the NGO Liaison Committee) took part. They had to indicate priority sectors and give a description of the overall situation in the country as well as stating a deadline for submitting project proposals. Further, a note regarding the process was drawn up. This note had the character of a "regulation" which gave guidelines for the identification and initial planning of programmes in the five countries. The note did not include anything about coordination with the Member States in spite of the importance of such coordination, as indicated in the Council conclusions of May 1993.

32. Once project files had been prepared by the NGOs and received by the Commission, the NGO community had to designate one or two NGOs in each country to act as the focal point for the EC Delegations, the governments and the NGOs. This coordination never materialised as intended. In Mozambique, the tasks of the so-called focal NGO were limited to collecting proposals for one programme. Apart from this the NGO did not play any further role. For the other countries reviewed by the Court there was no indication of any coordination activities performed by an NGO or any documents outlining the role of coordinating NGOs.

33. In Eritrea NGOs complained that they were not given enough time to prepare their projects because of disbursement pressure from the Commission. At the end of 1993 they were requested by the Delegation to forward project proposals in less than one week.

34. It would have been useful if the Commission had prepared a document for the EDF Committee outlining the overall rehabilitation programme for each country. Such overall programmes have never been prepared. Instead programmes or projects have been presented along the normal lines, i.e. in the form of individual financing proposals, or information notes for projects costing up to ECU 2 million.

35. After the initial stage described above, for the original five countries as well as for the new countries, programmes and projects were identified in a more ad hoc manner. In some cases the European Union took a share of a programme prepared by a United Nations (UN) organisation or the World Bank. In other cases no such coordinated actions have taken place. In one case the country itself had presented to the donors a nation-wide programme for its rehabilitation (Eritrea). It was initially planned that the European Union should participate in the programme but this idea was abandoned. In Haiti the final part of the identification of actions took place five months after the financial agreement had been signed.

36. In Mozambique, two contracts with the United Nations High Commission for Refugees (UNHCR) were signed at a time when the SIA programme should actually have been finalised. The actions were to a significant extent agreed and even started under contracts between UNHCR and a number of NGOs and local Mozambican authorities. These subcontracts were not communicated to the Delegation, hence the Commission accepted actions without any proper appraisal. The actions were all emergency ones and as such they did not qualify for assistance under SIA. The same problem with sub-agreements under UNHCR contracts was also pointed at in the Court's special report on reconstruction in former Yugoslavia(11).

Coordination

Donor coordination

37. Coordination of donor-funded actions is of vital importance. The necessity of a strong donor coordination was stressed by both the initial Commission proposal and the Council Conclusions. It was mentioned that coordination has to involve the local authorities of the country, all the donors (Member States, UN bodies and others) as well as the implementing partners (NGOs are mentioned specifically). Although the Commission's services stated for all countries covered by the audit that coordination had been respected systematically and carefully, the Court has found this difficult to verify. This is due to the lack of documentation and the non-systematic way of following-up and reporting on most aspects of the programme (see also paragraphs 102 to 107 on monitoring).

38. For Somalia, a Somalia aid coordination body (SACB) was established in Nairobi in 1994. For a long time it was chaired by the Commission. However, the donor interest has declined since, the reason being that all donors operate through implementing agencies that possess their own specific operating methodology.

39. Even for the so-called pilot countries for EU coordination(12), where the Commission has been given responsibility to coordinate its own and the Member States' activities, there is a lack of documentation. In one of those countries (Mozambique), Member States actually complained that the coordination had been far from effective. The Court also commented earlier on the lack of coordination with Member States, in its Special Report No 2/97 on humanitarian aid.

40. The coordination efforts were mainly aimed at information sharing. Genuine coordination, such as joint operations and procedures, evaluations, controls and monitoring, never takes place. Under the circumstances it was not possible to assess the impact of the combined actions supported by the donor community in any of the SIA countries.

41. In some of the countries supported by SIA, the UNHCR played a leading role during the emergency phase and also continued to play a dominant role for some time thereafter, coordinating and raising funds for rehabilitation activities. Such coordination included establishing trust funds where other donors had brought in their contribution, handing over implementation and reporting to the UNHCR. The policy of the Commission is not to hand over financial responsibilities to anybody else, for which reason the Community is never part of any trust fund facility. There is no justification for such an absence of coordinating measures.

Coordination within the Commission

42. In practice, for most of the countries and programmes there was a separation within the central services of responsibilities for planning, implementation and monitoring depending on whether the activities were funded from the budget or from the EDF. For separate budget lines within the general budget (e.g. co-financing and rehabilitation) the responsibilities were also split, reflecting the allocation within the organisation of the responsibility for different budget lines. Although information was exchanged between the various services, a stronger coordination would have been justified. Even the Delegations sometimes lacked information on certain activities funded under, for example, co-financing since the central services had not delegated any management responsibilities to them.

43. In Haiti a technical management unit was established for EDF projects. There were no technical assistants for the budget line in 1995 and 1996 (commitments for these years amounted to ECU 12,8 million). As control on the spot of these projects was weak, the head of the unit also tried as far as possible to follow some of these projects. In 1997, a technical unit was established, which also covered the general budget expenditure.

44. As mentioned in the Commission's evaluation of Angola(13), DG VIII and ECHO are acting more as separate donors than as complementary units within the same organisation.

45. The problem of internal coordination or lack of coherence is described in a Commission-financed evaluation of rehabilitation actions under Article 255 of the Lomé Convention (aid to refugees)(14). It is stated in this report that: "There is no reason for rehabilitation to be dealt with separately by different services of the EC. The present situation is due to the overlap of objectives of the food security strategy, certain ECHO projects, Community budget lines and Article 255. In part because often the partners are the same, policy discussions about Article 255 decisions are similar to those concerning other funding instruments."

46. The table below shows the financial instruments used for SIA activities over the years since 1993 and indicates the problem of coordination of responsibilities for the different departments/services of the Commission.

The lack of coordination within the services of the Commission was reported earlier in the Court's special report on humanitarian aid in a general sense and more specifically for Angola(15). By the end of 1998 the responsibilities within the newly created common service for external relations (SCR) were not yet clear.

>TABLE>

Human resources

47. The European Parliament, in a resolution of 4 November 1993, doubted whether the Commission was able to administer such extensive programmes in so many countries with no additional staff and with very tight deadlines. It therefore stressed the need for extra staff and expertise to be deployed to enable the policy to be properly coordinated and to guarantee the sustainability of the measures. In reality, no extra permanent staff have been deployed, or even requested, by the services of the Commission. The need for extra human resources for SIA has been met by setting up technical support units (TSU) with technical assistants. They have often done a good job but the expansion of the programme and its complexity would have justified setting aside (even if temporarily) specific resources inside the services of the Commission in order to secure coherence and coordination of planning, monitoring, evaluation and reporting. The lack of human resources, especially in the field, was mentioned by the Court earlier, e.g. in its Annual Report for 1997(16).

48. In the Delegation in Mozambique the situation was particularly bad due to vacancies and a general lack of resources, which made it impossible to deal with the volume of funds and projects. Furthermore, the administration of this Delegation was generally poor. Responsibilities were not clear and were frequently reshuffled, in a manner detrimental to any order or continuity. It was therefore virtually impossible to find documents and to account properly for the actions executed.

Feedback to the budgetary authorities

49. Information available on the way reporting was done to the Council on the progress made has been scarce for both actions financed by the EDF or by the budget. Annual Reports are requested by Regulation (EC) No 2258/96 for actions financed by the budget. Only one report was submitted, in 1997.

50. However, this report is not evaluative as stipulated in Article 9 of the Regulation. It merely describes and lists for each country the actions undertaken the previous year and briefly mentions proposed actions for the coming year. No mention is made of experience gained, lessons learned or problems encountered, be it from the few evaluations made in respect of individual projects or from the Commission's own monitoring. Apart from this report, the Commission has prepared a list summarising all decisions and payments from the EDF and the specific budget lines for rehabilitation. However, this list does not cover all rehabilitation activities of SIA, but does include actions which are purely food aid. The operational purposes of those lists remain unclear.

51. The Parliament, in its resolution of November 1993, called on the Commission to report back to it on implementation with an assessment of the results for each country receiving rehabilitation aid. This was not done in a systematic manner. Before the programme was expanded the requested report should have been presented.

52. The requested assessment of the results could not possibly be made without information available as a result of careful monitoring. However monitoring activities were almost exclusively carried out by technical assistants or by the implementing agencies themselves.

53. The Commission should control the monitoring process and document what has been done at Delegation level as well as keeping central services informed. In particular, in the light of the weaknesses and the often poor quality of the "first level monitoring" performed by the implementing agents (as documented in progress and financial reports), the Commission's monitoring is essential. However, for most countries reviewed there was no documentation available on the monitoring performed by the Commission's staff and many projects had not been visited at all by them. Unfortunately, not even checks or reviews of the monitoring reports prepared by them were documented. This means that there was generally no regular feedback within the services of the Commission on the ongoing SIA programmes. Admittedly, in some cases such as Somalia, Angola and the Congo, reports were often not available as the security situation made visits difficult and sometimes even impossible.

Implementation of actions

Introduction

54. Following the Commission's communication of May 1993, the programmes should be quick to implement and take a maximum of two years to complete. They should concentrate on repairing basic infrastructure quickly without any new investment. In reality, implementation within two years has been achieved only as an exception and has often taken four years or even more. One reason for the slow rate of implementation is that the planning of programmes and projects was (at least initially) not properly done (see paragraphs 55 to 63).

Planning and budgeting

55. The majority of SIA projects had been prepared and proposed by NGOs which were interested in taking part in the rehabilitation programmes in the different countries. The relative abundance of funds available and the marketing of the new concept resulted in hundreds of proposals right from the beginning for the then five eligible countries.

56. The project proposals were reviewed by the Commission, particularly by the Delegations in the countries concerned. The Delegations discussed the NGOs' proposed activities, their locations and the related budgets with the local authorities. However, due to insufficient human resources, most of the Delegations and the governments were not in a position to deal with this huge task adequately. It also usually took a long time to recruit and install technical assistance to deal with appraisal and monitoring of programmes.

57. In the meantime there was great political pressure to get programmes started, so many project proposals were accepted more or less as presented by the NGOs without sufficient critical appraisal. The technical and financial specifications for such projects were often not well based on the facts on the ground. This has frequently necessitated changes in the nature of actions, their locations, the target groups and the budgets. There were also instances where projects were interrupted from time to time and even completely stopped. A few projects did not start at all, although already contracted.

58. A project in Eritrea on malaria control included a budget for drugs amounting to ECU 125000. After the start of the project, it was discovered that the World Health Organisation (WHO) had included enough drugs in their separate programme. Although the files did not contain any request to use the ECU 125000 for something else, some ECU 80000 was in fact used for other purposes. Another project, also in Eritrea, aimed at the resettlement of refugees. After the contract for ECU 1,7 million had been signed, it was discovered that the land identified for the resettlement was already occupied by others. By then expatriate staff had already been recruited and some vehicles and other equipment purchased. The project had to be terminated after an amount of ECU 287000 had been spent. There is still an outstanding claim from the NGO to be settled. In the Congo, a social infrastructure project included a significant element for rural actions. As it was found subsequently that priority should be given to chemicals for purification of the water supply in Kinshasa, no funds remained available for rural actions.

59. In Angola, large sums were allocated to NGO projects which did not have clear objectives. This gave the NGOs considerable autonomy and changes were subsequently made without substantive justification by the Delegation, or any changes to the budgets. One project was reduced from two municipalities to one because of security problems and the number of beneficiaries was reduced from 23000 to 8000 families. Although this change was reported to the Delegation, no changes were made to the budget of ECU 810000.

60. In Mozambique, an amount of ECU 1,95 million was decided on for "reinsertion of young people in three provinces" in August 1994. The financial agreement was signed in June 1995. The project's target was to find employment for unemployed youngsters. To date no funds have been committed. The Government no longer finds the project relevant, but no proposals for reorienting the project have been made. The commitment should be cancelled.

61. Although many NGOs performed satisfactorily, a significant number of them did not have enough resources or experience for the sectors in which they were to operate to prepare proper project proposals or to easily recruit the required manpower, be it expatriate or local expert. They were also often too much oriented towards emergency aid with little experience in the field of development. As a result low quality projects were undertaken based on wrong assumptions about grass root involvement and the local administrations' capacities to take over the operations. Such projects are unlikely to prove sustainable.

62. In other cases the planning and appraisal were of good quality. These were often the less sophisticated projects, such as water points with hand pumps which followed on as continuations or extensions of projects which had been started/implemented before SIA under already existing rehabilitation programmes and co-financing. The mid-term evaluation for Haiti described the selection of actions as "exemplary".

63. Due to the use of many NGOs, the existence of several donors and the lack of proper coordination in all the countries, the system of planning and location of projects was rather fragmented and unintegrated. This led to parallel implementing structures, procurement systems and reporting systems at national level. Furthermore, projects were often too scattered to allow a country-wide impact.

Involvement of beneficiaries

64. The local (grass-roots) involvement of the beneficiaries in decisionmaking and implementation is a precondition for most aid programmes. Such involvement takes, and should be allowed to take time to have a chance of success and sustainability. Unfortunately the pressure to spend money and to produce quick results, both from the donor, the Member States and the recipient country, has been so strong that these aspects have often been overlooked or neglected. In Ethiopia and Eritrea, the governments and local authorities have been more involved in both the implementation and the monitoring processes. Government involvement has sometimes been considered too time consuming by the Delegations.

65. The generally acknowledged role of NGOs working in developing countries is to act as a facilitator during the implementation of projects. Implementation itself should be the responsibility of the beneficiaries themselves, as much as possible, with the additional input of local experts as required. Instead, many NGOs assumed the implementing role themselves by extensive use of expatriate staff. This makes successful transfer of knowledge and sustainability of the actions less likely. It has also created tensions with local authorities who normally are in favour of using as few expatriates as possible. The problem illustrates the tension between "quick implementation" and "sustainability of actions".

66. In Haiti beneficiaries of some agricultural projects were asked to participate in projects of which they did not know the final aim, the expected results, management modalities and financial arrangements. Involvement of the population in decision-making and reception/evaluation of the results has not been established and NGOs' project proposals are too often vague on this subject(17).

67. Whatever the reasons, proper monitoring and understanding followed up by enough support or corrective measures can normally compensate for shortcomings and should be appreciated as part of the efforts to rehabilitate the country. Almost any local involvement should therefore be regarded as a positive factor under the condition that it is transparent and accountable.

Sustainability

68. The involvement of the local authorities should include from the very beginning of a project an agreement that they will take over the structures and responsibility to secure operationality of schools, clinics etc. This responsibility should be stated in all FAs and contracts where applicable. Too often handing-over certificates had not been issued and the continued financing of infrastructures was not clear at the time of finalising projects. In Mozambique, for example, the Ministry of Health did not accept the structures and responsibility for their operation since it had not been involved in the planning. It was found that buildings did not meet Government standards and/or that they were not located where the ministry would have liked them to be. As a result, it is stated in two evaluation reports(18) that several clinics and health posts were closed or, when they continued in existence, they were operating below standard.

69. It is particularly difficult to secure sustainability in innovative projects and in areas where NGOs and others have little or no experience. There has been a trend, for instance, to create credit schemes and revolving funds under so-called income generating projects, such as farming, livestock, handicrafts and other manufacturing of which the target groups have no previous experience. Cost-sharing schemes for drugs, school equipment and teachers' salaries as well as water charges to make maintenance possible are other examples where beneficiaries are often not used to paying and have very few resources to do so. These projects have often more or less failed, or their successful implementation must be doubted. Projects which always involved the beneficiaries themselves from the early stages of the project have a greater chance of being successful, as was the case in a microhydrological project in Madagascar. In this project, local enterprises were also successfully involved.

70. After taking over completed projects, such as schools and health clinics, the staff will normally be paid salaries according to the established schemes for public servants in the respective country. NGOs and other implementing partners always pay salaries to local staff which are higher than those offered by governmental authorities. This makes it difficult to retain the staff after local take-over. Either the staff manage to find a new job, e.g. with an international agency or the private sector, or they might stay, but become demotivated.

71. In Angola monthly salaries were USD 1000 for a secretary working for a NGO and USD 300 for a driver. At the same time a doctor working for the Ministry of Health earned USD 10 to 20. The salaries offered under contracts financed under SIA were completely out of line with local conditions and not sustainable once Government took over the projects.

72. Such situations can partly be explained by the lack or scarcity of human resources. Not only did it make the recruitment of experts by NGOs a lengthy process but as a result of competition between NGOs high salaries had to be offered (more or less the same as those offered to "commercial" TAs) and even then people with enough experience and qualifications for the tasks were not always available. Sometimes people who had been recruited left the country before the contract had expired or did not extend their contract due to difficulties in coping with the local working conditions.

73. In Haiti neither the Government nor the parents have the funds required for the maintenance of schools. Equally there is no system which ensures the payment of charges for drinking water(19). Under such circumstances, the sustainability of the investment depends entirely on continuous outside funding.

Execution and results of programmes

74. The Court acknowledges achievements under the programmes in the various countries in the form of provision/restoring of schools, health facilities, water points, irrigation schemes, etc. which were operational and well used by the target groups. Examples of successful actions are also given in various evaluation reports.

75. However, because of inadequate reporting (see paragraphs 88 to 99), insufficient information is available on the overall results of the actions and on the extent to which the objectives have been met. Nevertheless, it was found in a number of cases that projects had been poorly executed and that some were not at all cost-effective.

76. As stated in the Commission proposal and the Council conclusions the immediate goals for the SIA initiative were to rehabilitate the social sectors and to (re)activate food and income generating activities as quickly as possible while at the same time paving the way for sustainable long-term development.

77. The start of many projects was severely delayed as it took a long time to disburse the initial funds. The lengthy procedures within the Commission services to transfer the funds to the implementing agents forced them to delay implementation for several months after contracts had been signed. Successive payments were also frequently delayed. In a few cases NGOs advanced their own funds to start or continue the actions.

78. There were also long delays because the local authorities had to approve financial reports from the NGOs and other agents, which were often unclear and of poor quality.

79. In Somalia, the cost of technical assistance under the first rehabilitation programme was very high. Around ECU 9 million of the total amount of ECU 38 million was used for technical assistance and consultancies (studies). If transport (cost of flights: ECU 1,1 million), administrative costs and other overheads in all projects are added to the technical assistance, not more than half of the total budget was spent on services or works directly for the beneficiaries.

80. Administrative costs could be included if justified, but not on a flat-rate basis that in many cases the Commission actually accepted. In Mozambique it was a rule rather than an exception that a 6 % flat rate of administration cost was added to all other budget items. Although the treatment of administrative costs was clearly stated in the guidelines (see paragraph 31), this was completely overlooked by its Delegation.

81. Initially it was stipulated that only existing structures should be rehabilitated and no new investments should be made. As it was felt that the restriction, imposed to promote a speedy implementation, was counterproductive, nevertheless such investments were made in the form of new water installations, schools and health facilities in most of the countries. The stipulation against new investments was therefore not included (except for the purchase of buildings) in Regulation (EC) No 2258/96.

82. The standard of equipment and constructions should not be excessive compared to the normal standard of the country to avoid higher construction and maintenance costs. In Mozambique the government complained of excessively high construction costs for clinics and water pumps in a project evaluated by independent consultants. They said they cost so much that they would be difficult to maintain and replace when worn out. In Haiti, on the other hand, for all road contracts local firms were chosen and they performed adequately.

83. In Somalia, support was given to new structures that would generate revenue (e.g. import duties), the most important being the port in Berbera. Although the Commission supported the introduction of financial control systems, there are still doubts about transparency and accountability. The Commission should have been more cautious before committing itself to revenue-generating projects under the present conditions as Somaliland, where the port is located, has not been internationally recognised as an autonomous State.

Simplified procedures

84. Simplified procedures for implementing the programmes were to be used for contracting, similar to those used by the Commission for emergency aid. As a result direct contracting was often used although this was not enough to ensure the required rapid execution. Instead, both implementing agencies and local authorities frequently complained about what they considered were excessively lengthy and bureaucratic Commission procedures. This related in particular to tender and procurement procedures, for which it took a long time to obtain the approval of Brussels.

85. A project in Mozambique covered the rehabilitation of roads, for an amount of ECU 30 million, which was part of a larger road rehabilitation programme funded by several donors. At the time of the audit visit in October 1996, almost three years had elapsed since the signing of the Financial Agreement and most other donor-funded roads were already finalised. At that time the contractors for the project in question had just started work. Several reasons were mentioned for the severe delay, such as poor quantity surveys, insufficient knowledge of EDF procedures by both the consultant used and the Ministerial Department responsible. These problems led to disqualification of tenders for administrative reasons. Eventually tender evaluation took place in Brussels, which did not help to increase the speed. The project continued to exhibit problems after it had started and, as at the end of December 1998, five years after the project had been decided, it had still not been finalised (only ECU 20,4 million of the contracted ECU 25,7 million had been paid). In the meantime urgent repairs to the same roads had to be made.

86. In Angola technical assistance was contracted in March 1995 for the supervision of a health rehabilitation programme of ECU 15 million which had been decided already in 1993. No real activities started until after June 1995. The programme has been extremely slow in implementation (50 % of allocation committed at the end of 1998 and 25 % disbursed after about four years). The Commission has blamed the national authorising officer (NAO) as one of the reasons for slow progress, i.e. a long time taken for processing project proposals and obtaining approval of reports/payments in respect of the NGOs. At the same time, the Court observed that about ECU 1 million had been spent under a separate programme in the form of technical assistance to the NAO. It seems that the outcome of these efforts was poor and not based on sound appraisal of the feasibility of the support.

Separate bank accounts

87. The contracts usually required that funds from the Commission had to be kept in a separate bank account and that interest should be accounted for separately. The contracts required NGOs to seek agreement on how the interest should be used: either on the same project or to pay it back to the Commission. These stipulations were mostly not respected and it is not known how much interest has been earned as the Commission has not closely monitored this aspect although substantial amounts may have been earned. For example, under a ECU 12 million project in Mozambique about 20 contracts were signed with NGOs. Interest was not separately accounted for in any of the related financial reports, which remained unnoticed by the Delegation.

Reporting

Requirements

88. In general, the reporting requirements for implementing agents were not specified and laid down before or at the start-up of the initiative. That was despite the specific requirement for reporting at all levels, mentioned in the Commission proposals and Council conclusions of May 1993.

89. This slippage caused many problems during the process. In the cases where technical support units (TSU) were gradually created, procedures for reporting by the implementing partners were established eventually. But TSUs were not set up in place in all countries or for all SIA programmes in a specific country. In most of those cases the Delegations generally had to invent, with or without assistance from the central services, the format for the reporting.

90. Reporting requirements for projects financed by the EDF, food aid, co-financing and other budget lines was the responsibility of different units in the Commission with, as a consequence, different formats, both as to content and frequency. This may explain partly why there has been no overall monitoring reporting for SIA until today - neither for all SIA actions nor for a single country.

91. Lacking precise requirements, NGOs either used their own established format, which might well have been accepted for earlier emergency projects financed under EDF and/or the general budget, or they tried to follow the format used in the proposal process. Since there were different reporting requirements for different financing instruments (SIA, food aid, other NGO actions), some NGOs had more than one format to follow. This also complicated the NAO's task, who had to approve the reports for EDF-financed projects before any further payments could be made.

92. In certain cases the Commission required that original invoices and other supporting documents should be attached to the reports, sometimes in the original, sometimes in copies. In other cases the supporting documents did not need to be attached. And sometimes these requirements changed during the duration of a contract. Different requirements have created confusion and unnecessary administrative work, both by the implementing partners and within the Commission.

Conflicting rules

93. In Mozambique, the UNHCR and certain NGOs argued that their own internal regulations on auditing and the keeping of original invoices and other supporting documents prevented them from following the requirements laid down by the Commission. Attempts made to solve these problems were not followed through.

94. For the UNHCR in Mozambique, the same contract format as the one used for direct contracts with NGOs was used. The UNHCR did not report back to the Delegation, as required by the contract, since their own contracts with the subcontractors were not designed accordingly and the accounting procedures used by the UNHCR were not compatible with the contract requirements. The reports from the UNHCR showed great discrepancies between budgets (in total ECU 3 million) and actions agreed between the UNHCR and the Commission and what was actually performed, yet neither the UNHCR nor the implementing agents ever sought Commission approval. The problem was partly due to the fact that some NGOs had mixed up funds from the European Union with funds from the UNHCR and other donors and/or own funds in the same project. By mid-1998, when the final payment was made, a number of questions had still not been clarified by them. Under the circumstances, there is a lack of transparency in both financial and administrative procedures related to these contracts such as:

(a) financial and performance monitoring by both the UNHCR and the Delegation;

(b) reporting to the Delegation and the NAO;

(c) the final approval and payment authorised by the NAO and the Delegation.

The same problems with UNHCR as to contracting, subcontracting, reporting and monitoring have been mentioned in the Court's special report on reconstruction in former Yugoslavia(20).

95. Approval procedures for contracts and payments also differ since funding from both the EDF and the budget was used. This has negatively affected the transparency for the recipient country, since the NAO's involvement in those procedures is not obligatory for actions funded from the budget. Sometimes, but not always, the NAO was given all information anyway.

Timeliness of reporting

96. The reports were seldom received at the required time and frequently reminders had to be sent to obtain them. However, in the case of rehabilitation actions funded under food aid and co-financing arrangements the reporting was even worse for both financial and progress reports. The files reviewed by the Court for these instruments for some of the countries did not contain any such reports and frequently did not even contain any signs of reminders having been sent out.

97. In Angola, an action was to be finalised in March 1996. A final report was only presented by the NGO in October 1996 but not accepted. The corrected report was not resubmitted until March 1997, six months later. Final payment was then made in the second half of 1997.

98. The Commission decided to finance a water supply project (ECU 750000) in July 1994. The first report was submitted in December 1995 after having already been subject of a time extension. A further extension was requested and approved for six months in April 1996. The second report was submitted in July 1996 and the final report in March 1997. The financial report was not accepted and returned. One year later (i.e. March 1998), the revised report had still not been submitted.

99. In certain countries the national or local authorities were directly involved in the implementation, either on their own or in cooperation with NGOs. Accordingly, funds were received by or channelled through them. It is of course equally important that there should be transparency and accountability in these instances. In those countries it often took a longer time and needed more efforts by the Delegations to get the reports, if they got them at all.

Audits

100. In a few cases where irregularities were suspected audits were undertaken by external auditors and the cases were handed over to UCLAF in 1996. These were still open by the end of 1998. One case concerned an NGO, which signed a contract with the Commission in 1994. After the first instalment of ECU 183000 had been paid, the "owner" disappeared before any activities had started. The contract is still open in the accounts, but the Court is not aware of any recent measures taken to recover the amount which had been claimed and paid. Otherwise there have been very few audits, although the regular undertaking of audits was mentioned from the start of SIA as important and a provision to this effect was included in Regulation (EC) No 2258/96. No overall record or knowledge in the Commission could be found of the number of its own audits within the SIA programmes. Even less is known regarding audits of NGOs and other implementing agents. In one case the implementing NGO mentioned in a report that it had initiated an audit itself. The Delegation in question had not asked for the report and could not provide a copy after the Court's request. It would anyhow have been in line with normal sound financial procedures to include audits in each or most of the programmes as a standard procedure.

101. There were also a few instances in Eritrea and Somalia where it had been noted by a cooperating NGO or the Delegation that funds had not been used for the intended purposes. Although it has to be acknowledged that a follow-up was difficult. Attempts to clear these cases on the basis of adequate evidence should be made.

Monitoring

102. The main responsibility for monitoring the programmes rests with the Commission, i.e. normally the Delegations.

103. The Delegations initially were not well prepared or equipped to monitor SIA actions. Procedures for monitoring were gradually developed in the SIA countries only after the programmes had started. In the absence of sufficient Commission staff, the monitoring was performed by TAs, mostly in separate TSUs but sometimes integrated into the Delegation. In Angola, NGOs complained that they had not had visits from the Delegation in the 18 months since the activities had started.

104. The Commission has rarely provided for ex post follow-up on projects as a standard procedure. The exception is Eritrea, for which all contracts with NGOs contain a stipulation that the NGO has to present to the Commission an "operational report" within one or two years.

105. For Somalia, the monitoring has been the responsibility of the technical assistants at the Somalia unit attached to the Delegation in Nairobi. Their reports can hardly be considered proper monitoring reports. No reference is made to the specific objectives of the projects. Only brief mention of problems discussed or solved during missions are made in the reports.

106. In Mozambique the Commission planned to establish a special unit to coordinate and monitor all of the rehabilitation operations financed by the Commission. This was mentioned in the Commission's proposal to the Council and Parliament in May 1993. The unit was never established. Instead several TAs were contracted to perform monitoring for the huge SIA programme which amounted to some ECU 180 million as at the end of 1998. Their monitoring was never documented at all and several projects had never been visited at the time of the Court's mission in September 1996. The permanent staff of the Delegation were not involved at all in the monitoring, apart from the local accounting staff, who checked the financial reports. The Commission commented that there was no need to document the monitoring since the TAs were stationed within the Delegation itself. For Mozambique, the Commission did not follow the progress of EDF-funded projects but concentrated on those funded under the general budget.

107. In Eritrea, the Commission's Delegation had documented its regular and rather frequent monitoring activities, including visits to most of the project sites. At the same time proper monitoring reports were regularly received from a TA specifically recruited for the programme. However, the monitoring reports prepared by the Delegation were not communicated to central services.

Evaluations

108. Evaluation of the impact of any development action is an essential element in ensuring sound financial management, particularly in helping define future programmes. If the programmes undertaken are conceptually new or inventive, as in the case of SIA, it is even more important that evaluations start as soon as possible and that decisions are taken to correct any shortcomings or problems which are observed during the evaluation. This importance was acknowledged by the Council and the Commission from the beginning.

109. The lack of measurable objectives in many of the programme documents is a factor that makes evaluations difficult. In the beginning, most project proposals and related financing agreements did not contain a clear logical framework of objectives, intended results and activities to achieve them. The situation improved during the later years.

110. A number of evaluations have been prepared but several evaluations of individual projects were made by the implementing agencies themselves or by local consultants who merely quoted from the progress reports, adding no further value. Comprehensive, overall evaluations of the rehabilitation actions undertaken by the Commission have only been prepared for Liberia and Mozambique. According to the Commission the budget allocated for evaluations has not been big enough to allow such evaluations. The Commission should give priority to such evaluations and should seek the funds for it.

111. Evaluations or midterm reviews are sometimes, but not always stipulated in financing agreements. But even when they are required they are often not carried out. Furthermore, the evaluations made did not normally cover a programme in total, but only one or a few of the projects or subcomponents of a programme, i.e. individual contracts and not the whole programme as presented in the financing agreement. In these cases it is very difficult to draw any conclusions as to whether the objectives have been met since the performance indicators and the goals are normally set for the overall programme.

112. For example in Angola, for three EDF programmes, evaluations were supposed to be made by early 1998. So far, nothing has been submitted. Two evaluations stipulated for programmes in Ethiopia will not take place, according to the Delegation. In Mozambique, apart from the overall evaluation (see paragraph 110) no specific evaluation of any project of the ECU 180 million programme has been executed.

113. A proper evaluation must be made by an independent body other than the implementing agent or the Commission. When asking for evaluation reports, the Court's auditors, for some projects, received reports prepared by the implementing agents themselves. These were generally of low quality and not problem oriented.

114. Sometimes evaluations were contracted out to local consultants. Ideally, local consultants, provided they have sufficient capabilities, should be used as much as possible since they are less expensive and could be expected to have a better knowledge of the local environment. In Ethiopia, a local consultancy firm was used to evaluate a number of projects under one programme. These evaluations, however, merely repeated conclusions drawn by the implementing agents themselves in their final implementation reports. Consequently they did not contribute to any future improvements of the performance of similar projects or the implementing agent under evaluation.

Conclusion

Concept

115. Rehabilitation activities under the SIA concept are first and foremost intended to be a strong link in the chain between emergency actions and long term development activities. However, no clear criteria has been established to define what should be considered "rehabilitation aid", as distinct from emergency aid and no management structure appropriate for this type of aid had been established (see paragraphs 1 to 4).

116. In practice a great deal of rehabilitation actions did not distinguish from humanitarian aid and in many cases emergency aid actions were continued under the name of rehabilitation programmes. Given the fact that within the Commission other services became responsible for the rehabilitation actions, a comprehensive documentation on the previous actions would have been useful. However, for none of the countries could such documentation be provided (see paragraph 9).

117. The financing structure should be adapted to the needs of the programme. There is no coherence between the different sources of funding, i.e. the general budget and the EDF. Different basic legislation and financial regulations apply and the beneficiaries' powers are much more important in the EDF context than in the general budget. Rules about contracting and other procedures such as reporting vary widely and the organisational structures within the Commission are different. Under such circumstances a mix of financing by the budget and the EDF is not effective (see paragraphs 10 and 11).

Identification and coordination

118. Countries, which did not meet the criteria of a minimum of stability, security and the existence of a government with a legitimate power in the country were nevertheless considered eligible by the Commission for rehabilitation aid (e.g. Angola, the Congo, Somalia). In such situations there was no basis for rehabilitation activities (see paragraphs 16 to 30).

119. The political pressure to start projects and disbursements quickly reduced the effectiveness of the planning stage to such extent that it became even counterproductive in terms of the goal of quick implementation. The Commission should have maintained emergency-aid actions until the environment was more favourable (see paragraphs 31 to 36).

120. Coordination between the Commission and other donors was limited to exchanges of information (see paragraphs 37 to 41). Whereas this problem is, to an extent, beyond the Commission's control, the Commission should have created firm coordination between its own services. In practice, each service within the Commission involved in rehabilitation action concentrated only on the actions for which it had direct responsibility. (see paragraphs 42 to 46).

Planning and monitoring

121. The Commission's initial overall planning was not transparent and not based on a thorough analysis of needs. Contracts with the existing partners were frequently extended, without a proper assessment of the financial situation and of the need to extend the activities (see paragraphs 12 to 15).

122. In particular for projects adopted at the beginning, it is not possible to verify that programmes have been based on assessments of the real rehabilitation requirements. Projects were often based on existing activities, e.g. NGOs already active in a region in a particular sector. Under the circumstances, the result is likely to be scattered and incoherent (see paragraphs 55 to 63).

123. Time pressure to present project proposals virtually prevented the involvement of local authorities and beneficiaries. This led to investments of which the sustainability was at risk (see paragraphs 56 to 73).

124. Monitoring and reporting were not adequate. In particular no regular comprehensive reports, covering both aid channelled through the budget and EDF, were made to the European Parliament and the Council by the Commission. In general, on the execution of the actions there was a lack of information which made adequate monitoring possible to a limited extent only (see paragraphs 90 to 95). Moreover, Delegations were not adequately staffed to monitor the activities (see paragraphs 102 to 107).

125. The lack of instructions on reporting and the lack of monitoring by the Commission created room for a number of NGOs to present actions without clear objectives or clear target groups (see paragraphs 59 to 61). The reporting by the NGOs was generally late and often of poor quality to an extent that is was extremely difficult to find out what had been achieved at what cost (see paragraphs 96 to 98).

Implementation and accountability

126. The rehabilitation programme was planned to be a temporary facility and the actions should have been executed rapidly. In practice the programme has become almost permanent and the actions took much more time than expected. It was often not possible to distinguish programmes from traditional development ones.

127. Some actions were delayed because of difficulties to recruit adequate local staff (see paragraph 72). The execution of actions was slow, partly because of long disbursement procedures and time-consuming decision-making via the Commission in Brussels (see paragraphs 77 to 78, 84).

128. Financial reporting was poor and there was no permanent financial monitoring. The Commission could not present a comprehensive view of the financial situation of the rehabilitation actions financed through the EDF and the general budget (see paragraphs 12 to 15).

Results

129. A number of rehabilitation actions have had results such as construction/rehabilitation of roads, schools, health centres and water supply. But no information is available on the overall results of the rehabilitation activities, be it on the programme or at national level, and often it was not possible to find out to what extent the goals of the individual actions had been met (see paragraphs 74 to 75).

130. For the beneficiaries, the minimum requirement for any sustainability is the existence of an effective institutional framework. Institution and capacity building should in a rehabilitation situation have a higher priority than what has been given so far.

131. On the basis of the experiences with the rehabilitation programme, the Commission should make an in-depth assessment of it. Such an assessment should include all phases of the programme, i.e. planning, execution, monitoring and evaluation, as well as the coordination of the actions, taking account of the positive experiences of the programme in Liberia. It would be useful to reconsider the concept of rehabilitation as a programme and to replace it by comprehensive individual country programmes on the basis of in-depth analyses of the specific situation of each country after a crisis situation.

This Report was adopted by the Court of Auditors in Luxembourg at its meeting of 26 and 27 January 2000.

For the Court of Auditors

Jan O. Karlsson

President

(1) Source: COM (93) 204 final.

(2) SIA was established by the Council in its conclusions of 25 May, 1993 following a proposal made by the Commission in its communication of 12 May 1993; COM (93) 204.

(3) PE 205.689 final.

(4) COM (96) 153, 30.4.1996.

(5) Special Report No 2/97, paragraph 2.4 (OJ C 143, 12.5.1997).

(6) OJ L 163, 2.7.1996.

(7) OJ L 306, 28.11.1996.

(8) COM (93) 204.

(9) The Commission's answer (13.5.1998) on the Court's desk review FD 3009 of 16 March 1998.

(10) COM (93) 204.

(11) Special Report No 5/98, paragraph 3.55 (OJ C 241, 31.7.1998).

(12) Council resolution of 3 December 1993 following which six countries were selected by the Council in May 1994 (Ivory Coast, Mozambique, Ethiopia, Peru, Costa Rica and Bangladesh).

(13) COWI Report, Angola Country Case Study, July 1997.

(14) Council Resolution of 3 December 1993 following which six countries were selected by the Council in May 1994 (Ivory Coast, Mozambique, Ethiopia, Peru, Costa Rica and Bangladesh).

(15) Special Report No 2/97, paragraphs 3.8 to 3.12 (OJ C 143, 12.5.1997).

(16) OJ C 349, 17.11.1998, paragraphs 5.28 to 5.31.

(17) Evaluation report on Haïti, DRN, September 1997.

(18) Evaluation, Article 255 Mozambique, COWI, July 1997; Phase II Country Report Mozambique, APT Consult, September 1998.

(19) OJ C 349, 17.11.1998, paragraphs 5.28 to 5.31.

(20) Special Report No 5/98, paragraphs 3.55 and 5.7 (OJ C 241, 31.7.1998).

The Commission's Replies

Summary

The Commission interprets the 1993 communication (COM(93) 204 final) as an important point in the evolving experience of policy and actions in this area, which emphasised flexibility and realism; it did not set up a "facility" of a given amount, nor did it fix a list of countries to be covered (it simply gave some examples).

The disbursement rate after five years of activity may at first glance appear relatively low, but it should be viewed in the context of the unusually difficult operating conditions in countries emerging from prolonged and profound crises. Account should also be taken of the rate of commitments and their targeting on certain "at-risk" countries, the upward trend in disbursement rates and the disbursement rates achieved for development instruments as a whole.

Concerning "actions were launched when the condition of relative stability was not yet met", this is true, but normal. It has been stated in the 1995 LRRD (linking relief, rehabilitation and development) communication, and in evaluations by the Commission and other donors, that the continuum is not linear, it is a contiguum; one country may have different regions in different stages of this contiguum, and the political and military stability of these regions may change. So the Commission's approach has been to decide to implement a project if there is "relative stability" in a region. Rehabilitation funds are there to consolidate peace, but there is no guarantee that peace, once established, will continue; it might be war which restarts. The Commission is not of the opinion that emergency aid should have been given, rather than rehabilitation - particularly considering that emergency aid is a relatively expensive form of aid to deliver.

The sums allocated for budget-financed rehabilitation in the ACP countries average EUR 40 million a year, which was obviously too little to cover all the needs of the communities affected. In line with the Council's conclusions of 2 December 1993, the programming of budget resources for certain countries at the start of the year has, since January 1995, sought to respond to needs in priority sectors, taking account of the other sources of financing available (national indicative programmes, Article 255 of the Fourth Lomé Convention). The adoption of Council Regulation (EC) No 2258/96 of 22 November 1996(1) formalised an established practice. Article 8 requires general guidelines for the operations to be carried out in the year ahead to be submitted to the Member States at a Joint Committee meeting. The general principles governing rehabilitation, the countries eligible and the priority sectors have all figured on the agenda. Due consideration has been given to rehabilitation's specific function: operations must increasingly take over from humanitarian action, taking account wherever possible of the existence of a minimal level of security. There has therefore been regular programming, coordination and definition of the scope of budget-financed rehabilitation(2).

The implementation of budget-financed rehabilitation has been the subject of close coordination between headquarters departments. Since October 1995, one unit's activities have included programming and preparing decisions on the basis of proposals from operational partners forwarded by the desk officers and Delegations. The pivotal role played by the desk officers at headquarters and the Delegations in programming and forwarding projects and programmes shows the coordination between the various sources of funding available. There is also very close coordination with ECHO (European Community Humanitarian Office).

Almost all budget-financed rehabilitation operations have been carried out by NGOs, whose projects included a detailed budget and clear objectives. The contracts for such operations specified the arrangements for reporting, especially when requesting payment of further instalments or the balance of funding. Special technical assistance teams have been employed in several countries receiving rehabilitation funding to monitor the implementation of projects and programmes on the ground and provide regular reports. The Commission nevertheless shares the Court's opinion that staff shortages, especially at the Delegations, have made the monitoring of operations more difficult.

The Commission has set up a task force to study the results achieved since 1996 in the matter of the relationship between humanitarian aid, rehabilitation and development. The aim is to work out the details of an enhanced integrated approach to handling emergencies and post-emergency situations.

Introduction

Background

4. The developing countries' capacity to absorb aid is recognised to be a general problem. In countries undergoing rehabilitation it is exacerbated by the destruction and the breakdown of structures. The operating conditions in such countries are therefore particularly difficult: not only are the needs very great and sometimes urgent, but the solutions are harder to implement. The reality experienced by the Commission's partners on the ground, which is often reflected by a longer implementing period, affects the financial indicators. That being said, compared with overall disbursement rates for aid, the financial performance of rehabilitation operations is not bad. What is more, the period concerned has seen disbursements steadily pick up speed. The implementation by NGOs of budget-financed rehabilitation activities addressing the priority basic needs of the communities concerned (health, water, education) has speeded up even more. This attests to a real and focused effort on the part of those implementing such assistance.

The framework

The concept

7. The Commission's communication of 12 May 1993 to the European Parliament and the Council,(3), the Parliament resolution mentioned by the Court and the Council's conclusions of 2 December 1993 on rehabilitation aid provide a sound conceptual framework for such assistance.

9. The task forces set up for countries and/or sectors have enabled joint analysis of the complex and changing situations in the countries concerned, thereby ensuring closer programming and coordination within a broad approach. The swapping of projects and programmes between ECHO and the Development Directorate-General is just one aspect of their work together.

The legal framework

10. The fact that different regulations apply to humanitarian assistance and rehabilitation does not impede an efficient implementation of the aid provided there is close coordination. Moreover, under the new Commission, both areas have been brought under the responsibility of a single Commissioner.

11. The Commission would like to describe in greater detail the ongoing process which, at its initiative, saw the different Community institutions flesh out the concept of rehabilitation and establish procedures for its application in the period May 1993 to November 1996. This process took account of the various sources of funding that the Commission was required to implement side by side.

Article 257 of the Fourth Lomé Convention (1990 to 1995) provides for ACP countries to be granted rehabilitation aid from a variety of instruments covered by the First Financial Protocol. Such projects and programmes are governed by the principles and rules of the European Development Fund (EDF).

Since there is also a considerable need for rehabilitation aid outside the ACP area, the Commission submitted a communication setting out a coherent approach to the Council and the European Parliament in May1993.

With the European Parliament resolution, the Council's conclusions of December 1993 provided a framework for the implementation of the first rehabilitation operations financed under Item B7 - 5076 of the general budget of the European Communities for 1994, supplementing the existing framework for operations financed by the EDF.

In mid-1995, the Commission submitted a proposal for a Council regulation on rehabilitation and reconstruction operations in developing countries(4). The European Parliament delivered its first opinion on 15 December 1995 and a common position was adopted on 29 January 1996. The European Parliament delivered a second opinion on 21 May 1996 and the Regulation was finally adopted on 22 November 1996.

Meanwhile, to foster a comprehensive approach, the Commission submitted a communication to the Council and the European Parliament on linking relief, rehabilitation and development (LRRD)(5), which led, inter alia, to the Council's conclusions of 28 May 1996 and the Parliament's opinion of 21 February 1997. One significant result was the setting-up of the task forces.

Against this background, it is difficult to see how anyone could say that there was no basis for a coherent approach to rehabilitation embracing both EDF and budget resources, whose complementary nature has been mentioned several times. What is, however, true since the special initiative for Africa was launched is that rehabilitation has not been viewed as a self-contained new instrument but as the flexible and coordinated use of a series of existing instruments, with the development over time of a consistent approach in the manner described above.

Financial resources

14. The need for very rapid action was specifically recognised in point 7 of the Council's conclusions of 2 December 1993. Quite apart from the scale of the needs, the sheer difficulty of implementation on the ground can cause delays and unforeseen expenses, just as sudden changes in the course of events can make it necessary to adapt operations.

15. ACP countries in the process of rehabilitation have had access to special sources of funding. The source used depends on the needs of the target populations, for instance, Article 255 of Lomé IV covers refugees, displaced persons or returnees. The budgetary authority has also made available additional funding from the general budget, in line with the terms of the European Parliament's resolution of 16 November 1993(6), which stressed the scale of the need for rehabilitation and called for a special framework invested with substantial financial resources to address such needs.

The resources of the national indicative programmes are subject to different criteria and implementing procedures from budget resources, which are administered more autonomously. From the standpoint of programming resources, the nature of the planned operations has played a key role. Budget resources have chiefly been channelled through NGOs and used for rapid response operations in priority sectors for the target groups. This has allowed any NIP funds still available to be used where possible, especially from the political standpoint, for other types of projects and programmes, and in particular projects and programmes aimed at getting development back under way.

Overall programming

Criteria

19. The Commission believes that the criteria governing a country's eligibility for rehabilitation funding are set out in two main documents: point 6 of the Council's conclusions of 2 December 1993 and Council Regulation (EC) No 2258/96 of 22 November 1996. Both cite "a minimum level of security" as a criterion, but qualify it with "all the necessary flexibility" and "wherever possible". They thereby strike a balance between the need for rapid operations to take over from humanitarian operations and the recognition that the complex process of emerging from crisis, rehabilitation and the resumption of development moves at different speeds in different places. Some periods or regions are stable while others are turbulent, hence the call in the texts for flexibility to meet the priority needs of the target communities.

Moreover, the eligibility of countries for rehabilitation and the operations to be financed have been established in close liaison with the Council via the Joint Committee, the EDF Committee and the prior information process.

Lastly - and this applies particularly to budget-financed rehabilitation - implementation by NGOs in extremely difficult and sometime dangerous conditions has served to maintain support for devastated communities. While some projects, albeit relatively few, have indeed been suspended or interrupted, to stand by and do nothing would have been morally indefensible.

Somalia

20. The Addis Ababa Agreement of March 1993 between the Somali faction leaders was followed by the Addis Ababa Declaration of 1 December 1993 to which the whole international aid community (governments, UN agencies and international non-governmental organisations) as well as representatives from the different regions of Somalia subscribed. This Declaration mapped out a strategy for reconstruction and rehabilitation activities for Somalia, based on the existing situation, i.e. absence of a central government, later on, this strategy became known as the peace dividend approach. The Community has been following this approach since mid-1994 when the first rehabilitation programme started. The international aid community clearly feels that there is presently no alternative to this strategy if Somalia is not to be abandoned.

21. Community-funded activities in Middle Shebell had considerably increased when the Somalia Aid Coordination Body (SACB) declared the region eligible for rehabilitation. The degradation of the situation in the region due to intra-clan conflicts that took place later on was not foreseeable. Independently of the problems that led to the stop of infrastructure projects, the investments already made can be considered as basically useful. If in the health sector five projects had to be suspended during the period 1994 to 1998, this happened in reaction to the changing security situation. In fact, as mentioned, only one project was prematurely closed whereas in the other cases the suspension was lifted after improvements had been introduced.

Angola

22. Notwithstanding the difficulties of implementing November 1994's Lusaka Protocol, Angola went through a period of phoney war until 1998, during which the international community supported the country with a view to peace and national reconciliation. The Commission continued granting the country emergency aid. At the same time the international community, at 1995's Brussels Round Table, pledged its backing for the country's reconstruction in support of the reconciliation process. The Commission's rehabilitation aid is part of this support.

Democratic Republic of the Congo (formerly Zaire)

23 to 24. The Commission cannot agree with the Court's observations concerning the advisability of launching rehabilitation programmes in the Democratic Republic of the Congo. The truth is that the Commission, in COM(93) 204, advocated a political settlement and the relaunch of the transition process. This document bears out the analysis that led the Community and its Member States to suspend all but humanitarian aid in January 1992. In its statement of 27 July 1994 the European Union took note of the turning point represented by the inauguration of Mr Kengo Wa Dondo's Government and the new assembly (High Council of the Republic - transitional parliament).

Pending the restoration of the conditions for conventional development cooperation, the Commission financed programmes aimed at improving the population's living conditions. This process, which faithfully reflects the "humanitarian aid - rehabilitation - development cooperation" sequence, has been partly interrupted by the conflict that flared up in the east of the country in late 1996 and spread in 1997. The conflict which brought President Laurent Kabila to power was unforeseen because it was unforeseeable.

25. The decision taken was to complete and launch those programmes already decided. With the arrival of a new government, the possibility of a transition could not be dismissed from the outset, nor had there been any change in the needs of the people or the conditions for the implementation of rehabilitation projects. To have abandoned the projects would have been far more reprehensible.

26. The election support programme (PAPE) to which the Court refers occasioned no expenditure and the commitment was closed in full.

The programme would have involved contributing to a census, a prerequisite for preparing not only the elections planned by the Kengo Government but larger-scale rehabilitation and development programmes reflecting the priority needs of a population whose present numbers and location can only be estimated.

When it became clear that elections were becoming a distant prospect under the new authorities the commitment was closed without a single euro being spent, as was the commitment in the CFSP budget heading providing for the establishment of a European team to monitor and observe the preparation and conduct of the elections. The Commission would point out that the elections were not specially planned to take place after the handover but to assist a change of government according to the rules of democracy: the programme was decided at the time of the Kengo Government, which had itself budgeted about USD 150 million for the purpose and was expecting the United Nations to contribute. The Commission programme supplemented these commitments and was kept open because the new President had promised elections within two years of coming to power.

The Commission feels that complementarity between the different sources of funding available (EDF and budget) and close coordination with other donors is more necessary than ever for operations of this kind.

Rwanda/Liberia

27. The Commission takes the view that rehabilitation aid was wholly appropriate in Rwanda's case.

While maintaining humanitarian aid, the Commission felt that more was needed to meet the needs of a people traumatised by the genocide. The Commission considered it necessary to get operations to rebuild infrastructure and relaunch economic activity under way as quickly as possible in the interests of national reconciliation.

Operations of this kind cannot be carried out with humanitarian aid, which cannot, for example, finance major infrastructure.

The reasoning behind the necessary link between the emergency, rehabilitation and development phases does not preclude the coexistence of all three types of operation in a given country.

28. The success of the operations in Liberia has also been recently confirmed by two independent evaluation reports, which came, inter alia, to the following conclusions:

- the objectives, results and impact of the rehabilitation activities have been highly coherent with EU goals, and improved the complementarity of all EU and Commission initiatives(7),

- the European Commission has proved that it is much more operationally responsive than generally believed(8),

- the Community has had significant influence both in the peace process and in the early stages of recovery from the civil war(9),

- since this evaluation was carried out the Commission has continued to play the key donor role in Liberia, through the implementation of the second rehabilitation programme. This role is widely recognised by the donor community(10).

Ethiopia

29 to 30. The Commission proposed this limit in its communication, but the Council chose not to take it on board in its conclusions of 2 December 1993. Nor does the limit figure in the regulation on budget-financed rehabilitation. The remark made on the Commission support towards the SIA has to be seen in its proper context. Already as from 1991 onwards the Commission financed a large number of "emergency actions" in favour of victims of war, drought and famine. The "social rehabilitation project" which started in 1994, following the emergency assistance previously given was part of the "continuum". One also has to remember that after the coming into statehood of Eritrea, EUR 20 million of Ethiopian EDF funds were transferred to Eritrea to finance similar activities in regions which previously belonged to Ethiopia. Available data show that an amount of EUR 14834739 has been used to finance projects in the context of the social rehabilitation programme for Ethiopia. This amount is composed as follows: EDF: EUR 2,7 million, Article 255: EUR 4,1 million and from B7 - 6410/5076 EUR 8,0 million.

In the period 1994 to 1999, Haiti was a country in political and institutional crisis. In such a situation it was just as important to support post-embargo rehabilitation efforts as to keep a firm grip on the economy to prevent the political and social crisis from being compounded by an unstable and worsening economic and financial climate. Financing adjustment programmes alongside rehabilitation projects and programmes was part of this strategy.

Emergency, rehabilitation and development aid are not so much successive as complementary instruments. In Haiti's case in particular, the development strategy necessarily included must continue to include emergency and rehabilitation operations. The objective is to achieve consistency between the strengthening and decentralisation of government through budget aid and rehabilitation schemes and measures to organise civil society.

Considerations over the programming process

33. It can happen that the budgetary planning of the Commission which has to apply the existing regulation does not coincide with the rhythm of project preparation by NGOs. As a rule NGOs are made aware of deadlines well in time, but do not always respect these. Wherever possible, a last chance is given to them to meet a given deadline they are about to miss, but this can sometimes just be a matter of days.

34. The Commission would stress that Regulation (EC) No 2258/96 provides for the Member States to discuss general guidelines for operations in the year ahead in a special joint committee. It would also point out that any rehabilitation operation over EUR 2 million entails the submission of a detailed financing proposal to the EDF Committee, which reviews the project in terms of the rehabilitation programme.

36. UNHCR was well established in Mozambique with a large team, good logistics and a presence in the districts. There were far better placed than the Delegation to select and monitor subprojects. UNHCR ran their operation through autonomous provincial suboffices. These were allocated funds according to the availability and the need, and contracted NGO partners accordingly. Their administrative systems were far more flexible than those of the Community, allowing for the rapid disbursement of funds and modifications to project outputs.

Given those comparative advantages, it was worth participating in the funding of UNHCR's global programme, without any precise earmarking of the Community funds on individual projects. Other Member States had a similar relationship with UNHCR. This had also the advantage of freeing up the Delegation's management for other tasks.

In this framework, UNHCR used the Community's funds as a contribution to their global workplan.

From October 1995 onwards UNHCR identified projects for specific Community funding under the Article 255 programme (7 ACP 080). However the programme was only approved in March 1996, with the result that these projects were either cancelled or other donors found.

Coordination

Donor coordination

37. The Commission has, for budget-financed rehabilitation in particular, employed a variety of means to coordinate its activities at headquarters and in the Delegations with its partners in the rehabilitation process, especially NGOs. The same goes for ECHO, which has received official notification of all planned operations since 1994, and for the Member States, which are consulted via the Joint Committee set up by Regulation (EC) No 2258/96 on the general guidelines and receive reports before and after all projects. Task forces have also been set up for certain countries and/or sectors of activity.

The Commission feels that there has been sufficient coordination and that this has helped prevent duplication in the use of funds.

38. There are no indications of declining interest of donors in the Somalia Aid Coordination Body (SACB), whose particular task it is to assure a maximum of harmonisation of operating systems and methodologies used by implementation agencies which are themselves virtually all SACB members, together with donors and UN agencies. On the other hand, there has been a certain decline of donor interest in Somalia as a reaction to the lack of progress in the re-establishment of peace and order in the country.

39 to 40. The Commission takes note of the Court's remarks, but would qualify them with the introductory observation: "Whereas the content of such coordination is to an extent beyond the Commission's control...".

41. The arrangements for financing the UN organisations have long been a matter of debate, especially with regard to the auditing of funds by independent bodies. During the period in question the Commission used the budget heading for rehabilitation and above all Article 255 of Lomé IV to help the UNHCR finance a number of projects or programmes with specific objectives and established budgets, sometimes taking over operations when the UNHCR's involvement reached its scheduled end.

Coordination within the Commission

42. Coordination at the level of the Delegations during the programming of budget-financed rehabilitation has involved taking account of other rehabilitation and humanitarian aid operations planned.

There is also coordination at headquarters via the desk officers responsible for the overall implementation of development policy.

This coordination comes into play when priority sectors are being chosen and projects appraised.

As for coordination with ECHO, all budget-financed rehabilitation projects require the Office's endorsement.

Since 1996, the task forces have put such coordination on a systematic footing.

43. From the outset and in agreement with the EU Delegation, technical monitoring has been carried out by the EDF rehabilitation unit and financial monitoring by headquarters and the Delegation. The recruitment of a special technical assistant was first mentioned in the 1995 to 1996 Annual Report, but the recruitment procedures have taken longer than expected.

44. The evaluation report in question concerns Article 255 of the Lomé Convention in general, with Angola as a case study. Coordination and complementarity between DG VIII and ECHO has mainly taken the form of joint programming of funding for NGO projects in the health sector (a field primarily supported by ECHO). A study was carried out for this purpose towards the end of 1996. Coordination and complementarity are also guaranteed by the Luanda-based LRRD task force and by coordination meetings at headquarters.

45. Since 1999, an interdepartmental task force drawn from the various External Relations DGs has sought to rationalise the budget headings and simplify practices.

Human resources

48. Human resources and volume of projects and funding are only two of the elements of the equation. Other crucial variables are degree of experience and familiarisation with the concept and financial instrument, capacity and level of ownership of the national administration, organisation of the civil society.

In the situation prevailing during the rehabilitation period in Mozambique, the instrument and concept were new, local administration and civil society non-existent, and available resources abundant. In this context, the Delegation has consistently acknowledged that the effectiveness of the Commission's rehabilitation programme was in large part compromised by the lack of human resources with which to manage a complex and diffuse programme.

One adviser managed this programme, with the assistance of, variously, either one or two expatriate TAs. And given that the concept of rehabilitation was new, few staff could draw on prior experience, or be expected to have such a wide range of professional knowledge. In the prevailing situation, this imbalance between the magnitude of the financial resources and the staff to manage was a contributing factor in the programme becoming demand-driven, with no clearly articulated strategy.

Under these circumstances, it was not easy for the Delegation to undertake a more rigorous institutional and situational analysis for the whole intervention and for every project and every partner. As a result it took longer for "lessons learned" to be incorporated into the Delegation's strategy.

However, based on internal and external evaluations, the Delegation has been able to incorporate lessons learned into decision-making.

- There is now a far better understanding of the capacity of partners, and what activities are likely to take root. Using guidelines developed in Brussels, it has been possible to follow the best practices for interventions such as microfinance.

- The rural development adviser is now supported by two TAs and a field officer based in Maputo. A TA will soon be based in the northern province of Zambezia. This contrasts with the single TA who was managing a portfolio of projects between 1994 to 1996. In the forthcoming rural development programme, there is provision for experts in monitoring and evaluation, poverty and gender and economic analysis.

- The financial administration has two specialised local staff. The Court of Auditor's recommendations have been institutionalised.

- The decrease in the amount of funds has allowed a more considered approach to rural development. New initiatives planned under the eighth EDF are sectorally and geographically concentrated. Institutional absorptive capacity and financial analysis are routinely considered.

- The process of project selection is now systematic and transparent. The Delegation has pioneered the use of the call for proposals in order to better programme Commission resources.

- The Delegations rural development strategy places an emphasis on synergies within the Community, Member States and other donors' programmes.

Feedback to the budgetary authorities

49. Since the special initiative for Africa got under way the Council has received progress reports at least once every six months on the overall rehabilitation effort in terms of decisions and disbursements. It appears to be satisfied with the information received.

50. In line with Article 9 of Regulation (EC) No 2258/96 of 22 November 1996, the Commission drew up a report on the Regulation's implementation in the 1997 financial year, the first following the Regulation's adoption. With regard to the ACP countries, it contained all the information required by Article 9 concerning the implementation of the rehabilitation budget headings. The format of the report complies with the requirements of the Regulation.

51. The Commission would stress that the ongoing discussion of rehabilitation documents (the draft Council Regulation, the communication on the links between emergency aid, rehabilitation and development (LRRD)) permitted the various Community institutions to be kept abreast of the progress of the overall rehabilitation effort. Furthermore, the budgetary authority has each year approved appropriations for the rehabilitation budget headings on the basis of the report on the previous budget year's activities.

52 to 53. The Court has already remarked on the impact of the lack of human resources, especially in the Delegations, on the evaluation of operations. The arrangements for having operations monitored by technical assistance staff as well as by the operational partners themselves, usually highly experienced NGOs in the case of budget-financed rehabilitation operations, have enabled a number of individual projects to be evaluated. Two major overall evaluations have also been carried out by the independent consultants COWI and APT Consult. Note that these evaluations were carried out in 1996/1997 and 1997/1998 as part of the Commission's evaluation policy and that they are available to the public.

Implementation of actions

Introduction

54. The Council's conclusions of 2 December 1993 mention neither of the restrictions cited by the Court, namely a two-year time limit and no new investment. Nor do they figure in the Regulation on budget-financed rehabilitation. The acts in question refer simply to the limited duration of operations.

These restrictions did, however, figure in the Commission communication. The Commission concurs with the Court's finding that the implementation of rehabilitation projects and programmes has often taken longer, sometimes much longer, than initially planned. While not dismissing the explanation put forward by the Court, it must emphasise that rehabilitation operations far more often fall behind schedule as a result of the very difficult context, circumstances and conditions in which they are carried out, and even then they are usually completed within an acceptable period.

Planning and budgeting

56. The difficult conditions in which rehabilitation activities are carried out, and in particular a shortage of human resources in the Delegations, are recognised.

57. The fact that a relatively limited number of operations have had to be changed, interrupted or stopped cannot be attributed to inadequate preparation in every single case. It is primarily a result of the complex and sometimes uncertain situation reigning in the countries emerging from serious crises that are eligible for rehabilitation aid.

58. The outstanding claim of the consulting firm, relating to a cancelled project on resettlement, is currently under investigation by the Commission.

59. The fluctuating security situation has often led to the need to suspend actions in certain areas. At the time of suspension, it is generally not known how long the insecurity will last. This difficulty is not unique to Angola.

60. During the discussions held in March 1999 with the NAO and during the country review of Mozambique, it was decided that this project was not a priority for Mozambique and that it was not going to be implemented. On 28 September 1999 and based on the lack of consensus between the Parties involved concerning the redefinition and formulation of the primary objectives, the Delegation requested its cancellation. A recent reaction of the NAO to this issue (15 October 1999) was that the project should be implemented as it is, as the Financing Agreement was already signed and for this particular reason was not included in the list of projects (priorities) agreed between the NAO and the Delegation in March 1999. The Commission will propose that the project be closed and funds decommitted.

61. The Commission took account of the NGOs' experience, particularly in crises, with a view to ensuring the continuity of activities in priority sectors for the groups concerned. The scale of the needs, the diversity of the situations and the size of the countries concerned has sometimes led to the diversification and multiplication of operations through a number of different partners. This is consistent with point 5 of the Council's conclusions of 2 December 1993: it is to be hoped that the NGOs will play an active role in the implementation and, where appropriate, the drawing-up of Community rehabilitation aid programmes. However, in the interests of consistency, priority sectors and areas for budget-financed rehabilitation have been identified in the light of such factors as operations financed by other instruments or donors ever since the 1995 programming exercise. This approach has been continued and indeed stepped up as budget-financed rehabilitation has been implemented.

Involvement of beneficiaries

66. Projects have been chosen on the basis of the knowledge, experience and reputation of NGOs working on the spot, often on projects financed by the EU. Putting relations with NGOs on a contractual basis has improved the transparency and monitoring of activities. The difficult political situation following the embargo and the dictatorship has to be borne in mind, as must the urgency of the moment. While involving the beneficiaries at every level is desirable, it is undoubtedly less important in a rehabilitation operation aimed at restoring basic services than it is in a longer-term development project. Moreover, some NGOs continued to operate as they had during the embargo, with an emphasis on urgency and meeting the most pressing needs.

Sustainability

68. As for the Commission, firm declarations from the counterparts agreeing to the infrastructure investment have always been requested, even if this can not be considered sufficient to justify the sustainability of the action.

On the other hand, in the absence of a longer-term planning by the Mozambican administration, there was often no choice but to invest in rehabilitating the infrastructure network with little or no sustainability guarantee or to abandon the population. At present, almost no infrastructure is still abandoned.

69. The Commission will make sure that NGOs operating on the ground have the technical capacities and human resources needed to work in specific fields, avoiding subcontracting wherever possible. Note that a project's sustainability also requires that the relevant ministries be involved from the outset so that the national budget can cover spending when Community support ends.

70 to 72. The problems of low government salaries, and demotivation of staff, are of course a common feature of operating in developing countries. The Commission's general approach is two-fold:

(i) to seek to address human resource issues in social sector reform programmes; and

(ii) to provide fringe benefits in ways which maximise social gains such as financing for schools, improved housing etc.

73. In certain cases, the criterion of sustainability can have a lower priority than the immediate needs of the population.

Execution and results of programmes

75 to 76. In the cases referred to by the Court, the Commission wishes to emphasise that the continuous supply of basic services to the affected populations in a post-crisis context was a particularly difficult task to implement.

"The projects have in many cases contributed towards the resumption of minimal services in the social sector and the restoration of infrastructure in priority areas. Project impact has significantly contributed to peace building in Mozambique in early years and more specifically prevented the continuation of violence in Liberia.(11)"

77. The Commission has pledged to clear the backlog and has embarked on a campaign to speed up payments and make up for lost time. In many cases it is the NGOs which breach their contractual obligations by delaying the execution of payments (and in particular, failing to declare interest, failing to submit reports in good time, failing to report in sufficient detail, altering the initial budget, etc.).

78. The Commission accepts that reports from NGOs may not have been sufficiently clear, but has taken the appropriate measures to improve the financial reporting.

79. The difficulties of recruiting TAs to serve in unstable or crisis countries are evident and have already been explained in paragraph 72 of the report.

Simplified procedures

84. It should be noted that the Lomé Convention provides the necessary simplified procedures, resulting in shortened tender and procurement procedures, and rapid execution. As the report explains, other reasons for delay, such as inadequate project preparation by local authorities, have played a role.

85. Project 7. ACP. MOZ.47 funded the emergency opening of roads (EOR) in two provinces of the centre-north of Mozambique: Zambezia and Sofala.

Preparation of the tender documents for works and supervision in both provinces took place in 1993 and the respective tenders were launched in the third quarter of 1994. Beside the reasons pointed out, the mine-clearance component of the project, an area in which there was no previous experience in relation to road construction in Mozambique, caused additional delays.

Supervision contracts for both provinces were signed at the end of 1995 while the two works contracts were signed in July 1996. However, given the time elapsed since the Financing Agreement was signed, there was a need to improve the technical standard initially adopted, so that increased durability of project roads could be achieved. The new technical standard led to the concept of improved EOR, applied during project execution.

The initial date of completion of the EOR project in the province of Sofala was 25 March 1998. Due to the changes approved by the department of roads and the Community, to the need for assistance in the repairs of the sections washed out following exceptional rains, and to delays by the contractor, the project finished on 2 March 1999.

Overall, and despite the fact that two road sections could not be finished within the first contract, the differences between the road conditions now observed and those existing before the rehabilitation are substantial. The time needed to travel from Save to Inchope has now been halved.

The initial date of completion of the EOR project in Zambezia was 25 February 1998. Due to the extensions approved by the department of roads and the Commission, the project finished on 31 December 1998. Despite initial delays, the quality of the roads rehabilitated was good and, in fact, they were considered to be the best of all roads under EOR projects in Mozambique.

Despite the delays in project commencement, there is no information on repairs carried out by NGOs in any of the project roads in the province of Sofala.

In the province of Zambezia, due to project delays and the very critical conditions of the roads, localised emergency repairs were carried out before commencement of the works by NGOs on any of the project roads in the province of Sofala.

86. The Financing Agreement for the ECU 15 million post-emergency health programme was signed in early 1994. The project was indeed very slow to get under way (in terms of disbursements). Though expenditure was way behind schedule, much support and training was provided in the field. The project was evaluated and reinforced in 1997/1998. Since then the main expenditure has been committed.

Separate bank accounts

87. When a project is closed the Commission sees that NGOs declare the interest received on advances paid by the Commission and detail the expenses covered by such interest.

Reporting

Requirements

88 to 99. The Commission is introducing (from 1 January 2000) simplified contract procedures for grants. In the standard contract, reporting requirements are clearly set out, both in relation to technical and to the financial aspects of the contract. The report must be laid out in such a way as to allow comparison of the objectives, the means envisaged or employed, the results expected and obtained, and the budget details for the operation.

In order to standardise procedures invoices remain in the hands of the NGOs for the period laid down by law. They are examined on the basis of sampling or in the course of official audits.

As indicated, many of the problems highlighted by the Court should no longer arise since the procedures for reporting and for payment have been harmonised and simplified.

The Commission sees that payments are made on presentation of intermediate and/or final reports and after obtaining the opinion of the Delegation concerned.

Audits

100. Since July 1999, all financing agreements for EUR 100000 or more have provided for an audit of the operation or the NGO. Furthermore, a guarantee now has to be lodged for the payment of advances of EUR 1 million or more. This guarantee requirement does not apply to internationally recognised NGOs routinely supported by other donors.

All the cases pointed out by the Court will be followed up.

101. The Commission is aware of the risks inherent in the use of aid to organisations and entities in the context of conflict and rehabilitation situations. It is standard practice that the Commission investigates all allegations brought to its attention, although many of these allegations inevitably turn out to be unfounded.

Monitoring

103. For a while the scope for on-the-spot monitoring of NGO projects was indeed limited by a shortage of human resources at the Delegation and logistical problems.

104 to 107. As explained in the reply to paragraphs 88 to 99, the reporting system has now been changed and this should make projects easier to monitor. In addition, the Commission is introducing more systematic and comprehensive monitoring procedures for all its programmes which will ensure a more uniform approach and more intensive coverage.

Evaluations

108. The Commission fully agrees with the importance of evaluation. This is why independent experts have already evaluated most of the country programmes covered by the report from 1997 onwards, as soon as the programme was sufficiently advanced.

109. The Commission accepts that, particularly in the early period of the programme, project preparation was not satisfactory, mainly for the following reasons:

(i) current project-cycle management systems are not always appropriate to rehabilitation actions, given the different types of risk which rehabilitation assistance can address (see APT evaluation report, p. 13);

(ii) social and economic structures have been drastically changed by instability which means that existing data and assumptions may be unreliable.

The Commission is currently considering defining more specific guidelines for rehabilitation projects.

110. Besides the two overall evaluations mentioned by the Court, the Commission has in fact made independent evaluations covering all essential aspects of the programmes in other countries, and considers the evaluation coverage of its rehabilitation actions to be outstanding, particularly in view of the extremely difficult operating conditions which still obtain in most of these countries. All final evaluation reports are available to the public. For NGO projects, because it would not be feasible or indeed cost-effective to evaluate a large number of small projects, samples are evaluated.

112. In the case of Angola, an initial overall evaluation of the budget headings was carried out in 1994 (along with an in-depth study of projects in Angola). Angola was also visited in the course of a second overall evaluation, this time in 1998, which took in all the projects financed from the budget headings. Two evaluations of EDF projects initially scheduled for 1998 were carried out in 1999.

As to Ethiopia, evaluations of completed social rehabilitation programme (SRP) projects by local consultants had been initiated by the technical support unit of the SRP using its own SRP funds. After the closure of the unit the local food security unit (LFSU) took over the remaining projects. Due to the closure of the technical support unit, SRP funds for evaluations of SRP projects were no longer available. However, stocktaking of achievements of completed projects has been carried out by LFSU. (See also the reply on paragraph 114).

Finally, rehabilitation actions in Mozambique have been evaluated as follows: Article 255 (1997), budget lines (1998), overall (1999). A specific evaluation of projects implemented by Handicap International was published in January 1999.

114. The local consultant, to which the Court refers, has done exactly what was expected according to the terms of reference. With 25 to 30 projects under implementation at a given time, an evaluation in the full sense of the term would only make sense if a large number of projects were covered in order to draw lessons for the future. As this was not possible yet at that time it was decided to limit the assessment to a stocktaking of achievements, including recommendations for future actions. Under these circumstances, the term "evaluation" is perhaps somewhat misleading.

Conclusion

Concept

115 to 117. The option of creating a single rehabilitation instrument has been rejected. There is no contradiction between the different acts (Lomé IV and the Regulation) governing rehabilitation, and complementarity between operations is possible. The period covered has seen the flexible and appropriate use of the various instruments available (EDF and the general budget in the case of the ACP countries). The criteria for their use, notably in terms of the concept, have gradually been developed by the Community's institutions. The Council's conclusions of 2 December 1993 and Regulation (EC) No 2258/96 of 22 November 1996 contain specific rules and clauses permitting the requisite flexibility. This flexible approach is amply justified by the key principles that have emerged from a series of discussions on rehabilitation: the process is complex, gradual and non-linear, may give rise to grey areas when a break with the past is needed, cannot simply be reduced to going back to the pre-crisis situation and must therefore be adapted to each situation and potential developments in that situation. In short it requires the flexible and coordinated use of the means available.

There has been a harmonisation in practice: budget funds for rehabilitation have chiefly been used to fund rehabilitation operations proposed and implemented by NGOs. Furthermore, under the new Commission a single Commissioner is responsible for both areas.

Lastly, the creation of the SCR has brought further harmonisation of practices. The introduction of a new standard grant agreement should result in a marked improvement. Disparities in payment methods should also disappear.

Identification and coordination

118. The Commission would point out that the rules call for a flexible approach to examining the eligibility of countries for rehabilitation, and that the Member States should be involved in this process. It emphasises the danger to people and the overall stabilisation process of too hesitant an approach.

Angola has a legitimate government recognised by the international community. The remarks in paragraph 22 should dispel any doubt about the relevance of rehabilitation operations.

119. The need for operations "to be launched as quickly as possible without compromising the quality of assessment" (Article 1 of Regulation (EC) No 2258/96) is a real constraint. Reconciling these conflicting requirements can be difficult in practice. The solution of simply continuing emergency operations would have posed just as many problems from the standpoint of administrative and financial resources and the risk of aid dependency at the expense of stabilisation.

120. Various means of coordination with the Member States, the various donors and other players have been introduced and used at headquarters and Delegation levels. Programming and decision-making on budget-financed rehabilitation operations involve the Member States, various departments at Commission headquarters and, via the Delegations, operational partners in the field. The sequence of events is pre-programming, programming (by country and sector) in the Joint Committee with the Member States and prior and subsequent reporting of decisions taken. The chief instruments are the annual preparation of general guidelines, the regular and systematic exchange of information, the constitution of task forces for countries or sectors. Coordination with ECHO takes the form of prior notification of all budget-financed rehabilitation decisions. EDF and budget resources are coordinated by the Delegations and desk officers, who are involved in programming budget spending and implementing EDF resources.

Planning and monitoring

121. The Commission believes that its programming is transparent and justified by the needs. The funds available each year for budget-financed rehabilitation are approved by the budgetary authority and implemented by the Commission. They are programmed for the countries eligible in accordance with Regulation (EC) No 2258/96. They are implemented on the basis of contracts, which have sometimes required riders to deal with implementing problems. A rider must normally be accompanied by statements of the grounds and the project's financial situation and prospects.

122. The Commission confirms that all rehabilitation operations were programmed with the relevant partners (Member States, specialised agencies and NGOs). Programming took place in the field and at headquarters and involved a variety of coordination methods to achieve the most relevant response to the needs of the target groups.

123. Given that there is often no national or even local administration, the best way to involve the beneficiaries has been to cooperate with operational partners and specialised agencies.

Having the annual budget allocation for rehabilitation implemented mainly by dozens of experienced NGOs from Member States or beneficiary countries, sometimes working in partnership, is an appropriate response to the priority needs of the target populations. Involving local people and their representatives limits the risk that projects will prove unviable.

124. The Council in particular has received regularly reports on the implementation of the rehabilitation effort. See also paragraphs 88 to 99 on reporting within the framework of contracts.

125. Budget-financed projects have normally had quantified objectives (e.g. number of beneficiaries), a detailed budget and a planned implementing period. Reporting, especially on a project's financial implementation, involves both discursive and financial reports. Difficult implementing conditions have appreciably increased the time taken to implement projects. The Commission has introduced changes to improve reporting.

Implementation and accountability

126. According to the acts governing it, rehabilitation operations must be carried out quickly and be of limited duration. Owing to its complexity, the rehabilitation phase may be extended, as indeed may the implementation of projects, and that for the very same reasons. Several countries nevertheless ceased to be eligible for rehabilitation aid during the period in question. Generally speaking, the rehabilitation and development phases can be quite clearly discerned. The transition is nevertheless gradual, and programmes may be called for to smooth the passage from one phase to the next.

127. The difficulties of recruiting staff, whether local or expatriate, in situations of recent or nearby conflict and acute lack of basic facilities are clear, and cannot be attributed to procedural problems in the Commission.

On 10 November 1999, the Commission approved a manual for its departments to simplify, harmonise and render more transparent procedures for awarding contracts across the world. The decision will eventually reduce the number of procedures used for the purposes of different external aid programmes and the EDF from 40 to 8. From 1 January 2000, these new procedures will be phased in for all programmes (MEDA, ALA, Phare, Tacis, Obnova, etc.) and the EDF as the requisite training courses are carried out and the provisions specifically applying to current contracts under the regulations and legal acts governing the various programmes are revised.

These new measures will be a great step forward for European economic operators and the contracting authorities in the partner countries. The most obvious result will be to speed up and improve implementation of external aid programmes and projects while enhancing the security and transparency of contracts and tendering procedures.

In this context, model contracts will also ensure that reporting requirements are standardised and payment procedures are faster.

128. The impact of EU-funded rehabilitation actions has been examined by independent experts who have noted the contribution they have made to the peace-building process and the way in which they can prevent the continuation of violence.

Results

129. Budget-financed rehabilitation has been subject to annual programming, with due regard for the intrinsically transitional nature of rehabilitation aid. In this context, given that the resources fall far short of the needs, establishing priority sectors, including the continuation of activities already under way to prevent sudden breaks, does not seem unreasonable.

All rehabilitation operations, the great majority of which are financed by the EDF, are preceded by an in-depth analysis of the needs and the most suitable means of meeting them. Operations carried out in the field by NGOs have permitted a continuous response to priority needs.

The many evaluations carried out provide an overview of rehabilitation activities. They do not suggest that the Commission's activities in the field, which complement other operations, are inconsistent.

130. Though restoring the institutional fabric has a major influence on the viability of operations, it depends on a number of factors that are far beyond the scope of rehabilitation and require a comprehensive approach. Rehabilitation activities meeting people's basic needs may be part of what is a complex, repetitive and gradual process.

In this context, as far as the institution-building is concerned, the very fluid and unpredictable political and/or military situation has a direct impact on the scope for institution and capacity building.

131. The introduction of rehabilitation aid in 1993 and 1994 differs from the implementation of rehabilitation operations in 1997 and 1998. As part of its ongoing review of the links between humanitarian aid, rehabilitation and development, the Commission will see that account is taken of the Court's pertinent remarks.

The Commission has already embarked on an overall review of rehabilitation activities aimed at drawing practical lessons from experience on the ground and converting these lessons into clear guidelines to harmonise the way different departments approach such situations. The process involves several stages:

(i) On the basis of a methodological study commissioned from a consultant, a task force set up in 1999 is studying rehabilitation operations in the light of the Commission's 1996 communication on the links between emergency aid, rehabilitation and development paper;

(ii) 2000 should see the drafting of a coherent set of operational guidelines. These guidelines will provide a framework for analysing crises and ensure that Community aid before, during and after a crisis is implemented efficiently and coherently under a broad-based and long-term approach. The arrangements for the use of these ground rules by the different departments concerned will also be examined;

(iii) These ground rules must then be formalised in an appropriate manner and supplement the 1996 guidelines.

(1) OJ L 306, 28.11.1996, p. 1.

(2) APT Consult, Phase I, Desk Review Report - Final (4 May 1998), page 21.

(3) COM(93) 204 final of 12 May 1993.

(4) COM(95) 291 final of 26 June 1995.

(5) COM(96) 153 final of 30 April 1996.

(6) OJ C 329, 6.12.1993, p. 77.

(7) Evaluation of the implementation of the budget lines (...) B7 - 6410 "Rehabilitation in all developing countries, ACP section - Financial years 1994, 1995, 1996, 1997 - Country Report Liberia"; APT Consult, UK, p. 5.

(8) Ibid., p. 6.

(9) Evaluation of EU aid, field phase; case study 5: Liberia; Investment Development Consultancy, France; August 1998, p. 1.

(10) Ibid.

(11) APT Consult, Synthesis Report - Final(23 September 1998), p. 13.

Top