Background information on the SPFS objectives, implementation strategy and content

145. - The main objective of the SPFS was to help LIFDCs improve their national and household food security on an economically and environmentally sustainable basis through:

• rapid increases in productivity and food production;
  
  
• reduction in the year by year variability in production; and
  
  
• access to food.

The underlying assumption was that in most LIFDCs viable and sustainable means of increased food availability existed but were not realised because of a range of constraints that prevented farmers from responding to needs and opportunities. By working with farmers and other stakeholders to identify and resolve such constraints and to demonstrate ways of increasing production, the SPFS was meant to open the way for improved productivity and broader food access.

146. - A programme content and implementation strategy had been designed by FAO to be used as a framework. The SPFS activities in any given country, which started with a country's request to participate in the programme, unfolded as follows:

• Upon receipt of a request, FAO designated a country core team and made a preliminary assessment of national food security strategies and of the possible components of the programme, drawing on in-house country knowledge and relevant technical experience.
  
  
• Formulation of the programme was a national responsibility, supported as necessary by FAO, either through the FAO Representation (FAOR) or though an exploratory mission. The formulation was usually carried out by a team of Government staff or national consultants guided by a Steering Committee, bringing together representatives of the public and private sectors, farmers' organisations and, in some countries, Non-Governmental Organisations (NGOs) active in the rural sector.
  
  
• The output of the formulation process consisted of:
  
  • A National Programme Document that placed the Special Programme in the context of national food security strategies, related it to other complementary initiatives and indicated how successfully piloted activities could be replicated on a larger scale;
  • A Plan of Operations for the Pilot Phase (or Phase I) of the SPFS.

147. - Phase I (done under farm conditions) provided an opportunity to demonstrate and refine proven and promising technologies and approaches to increasing agricultural production and rural incomes, before applying these on a much larger scale. Phase I usually consisted of the following four main inter-related and complementary areas of activity:

• water control through small-scale water harvesting; irrigation and drainage systems; simple lifting devices, such as the treadle pump; and land development and tillage systems that offer greater resilience to climatic variations;
  
  
• crop intensification with the introduction of widely accessible technologies (effective use of high yielding varieties, improved cultural practices, integrated pest management, etc.) combined with functional marketing and credit schemes and improved support service delivery systems;
  
  
• diversification of production systems, which included aquaculture, artisanal fisheries, small animal husbandry (poultry, sheep, goats, pigs, etc.) and tree crops; and
  
  
• constraints analysis and resolutions to identify any problems or obstacles that prevented farmers from increasing their productivity or output and made it impossible or difficult for them to adopt new technologies and management practices.

The balance between these four depended on country circumstances and the results of diagnostic work.

148. - As resources became available from external and national sources, Phase I could be extended in two ways:

• by increasing the number of activities at existing sites, and
  
  
• by adding to the number of sites, to create a maximum of 30 sites covering different ecological and administrative areas of the country.

The outcome of Phase I (or its extension) determined the scope of the last phase (Phase II), which was supposed to build on the achievements of Phase I to create the conditions for large scale replication of development approaches that have proved successful. Since all the countries visited were still in Phase I at the time of the review, the following paragraphs only concern this phase and the preparatory stages.

Management arrangements of the Programme at the country's level

149. - One of the distinguishing features of the SPFS was that it aimed to link practical demonstration work conducted by farmers with changes in the institutional and policy environment that addressed constraints to the wider replication of successful innovations. Such an objective was met through the issuance of the SPFS National Programme Document, and the creation of a national institutional framework and organisational structures. Arrangements for the latter had to be made at the local level to ensure a high degree of farmer participation and at the national level to bring people who were influential in the processes of policy and institutional change together. There was also a clear need for strong and competent technical leadership.

150. - In this regard the review my staff conducted showed that the degree to which these requirements were met depended on country circumstances. Their main observations were as follows:

• Although the formulation of the programme through the SPFS National Programme Document was of capital importance, it was still under negotiation in several countries when the first SPFS activities were implemented.
  
  
• The advantage of national execution was that the skills learnt during implementation of the SPFS remained within a country's structures. However, as the budgetary constraints of countries undergoing structural adjustment had resulted in a reduction of local administrative staffing, national personnel was sometimes recruited to handle the additional workload related to the SPFS but only for short-term missions. The probability that the skills acquired would be put to subsequent use within the governmental structures for the benefit of the country's agriculture might, therefore, be lower.
  
  
• The National Technical Committees, whose mission was to follow up the implementation of the activities at national level, often did not report on their activities. However, when they operated efficiently, they played an important role.
  
  
• The National Steering Committees, which were intended to provide a link between technical committee findings and policy makers, had not entirely fulfilled their role. Few decisions had been taken at national level.
  
  
• The key organisational issue for SPFS implementation proved to be the designation of the National Coordinator (also named Project Manager in some cases). Since he was paid by FAO, the Organization usually managed to select a suitable candidate from among the applications presented by the participating country. However, in one case FAO was not given a choice and in another it was faced with excessive turnover (four Nationals Coordinators successively appointed in three years). Furthermore, the source of funding also proved important. In one case, the position of the incumbent was rendered precarious because the budget allocation (under TCP funds) permitted his salary to be paid for only 12 months.

Management arrangements of the Programme at FAO

151. - When the SPFS was initiated, the core teams appointed to support the countries were made up of FAO staff based at Headquarters. The selection of core team members reflected the concern for multidisciplinary although, initially, many members were drafted from the Agriculture Department. The decentralisation that was initiated in 1996 dispersed these teams, with some members assigned to the Regional Offices and others remaining at Headquarters. Liaison between team members was made more difficult, especially as travel funds were limited, and relied on modern communication facilities, i.e. tele-conferencing and e-mails.

Implementation of the Pilot Phase (or Phase I)

152. - The methodology used for Phase I consisted of carrying out demonstrations of improved technologies on the plots of volunteer farmers and comparing the results with those of traditional practices. The whole process was meant to be a participatory one. Local farmers were supposed to be involved in deciding which crops would be grown on the demonstration plots and, above all, made well aware of which factors would condition the technical and economic results. Individually or in groups, they were to be encouraged to identify and analyse the constraints to adopting the improved techniques, and to seek appropriate solutions.

153. - The extent to which the implementation approach had been indeed participatory depended sometimes on availability of funding. For example, my staff noted that in one case, the process was initially participatory with the involvement of the local co-operatives. However, because of lack of funds, only individual farmers, who were able to pay the travel expenses of the technician, were eventually concerned. In the six other cases, my staff noted that the technical personnel were trained in the participatory approach, which was a positive step that indicated a willingness to accommodate new methods. In the same countries, the farmers themselves were also trained to form groups capable of working together or assuming collective responsibility. However, in one case the training was stopped because of a lack of funds. It should be noted that training had not been identified as a requirement from the very start, but subsequently proved indispensable if the intended approach was to be applied.

154. - The food crop intensification and water control components of the programme were implemented from the beginning in all the countries visited except two of them for which the water control component was implemented later.

• Intensification focused on cereal crops. The most important outcome was the improvement of the technical expertise and, in some cases, economic competence acquired by the farmers concerned, raising awareness among other farmers and the availability of increasingly reliable technical and economic field data.
  
  
• Water control was addressed at village level to improve farmer practices. Real progress has been made in organising water users in the maintenance of small irrigation schemes and water utilisation. Even though the problems linked to the larger systems could not be settled at this level, these were nevertheless discussed, which provoked thought and generated ideas.

155. - The diversification component was only introduced at a later stage in the first countries implementing the SPFS, with the exception of one where it was introduced earlier. The farmers generally appreciated this component, especially when they were really given the freedom of choosing activities. In general, my staff noted that diversification could help to vary and balance the household diet or to generate income from the sale of produce on local markets. It was further noted that this component was the one that brought the women into contact with the SPFS.

156. - Although constraints analysis was a major component of the SPFS, it was only in April 1998 that the SPFS teams were provided with the "guidelines for the conduct of the results and constraints analysis component". The review also disclosed that the different countries had usually appointed experts or consultants to implement this component, particularly to train farmers and technicians, but also to directly carry out the analysis. The constraints identified during the implementation of Phase 1 were not different from those frequently pointed out. However, the fact that they had been analysed under real conditions made a difference. In my opinion, the most noteworthy outcome of this component was to have allowed participating farmers to acquire the intellectual capacities necessary to understand what mechanisms were operating in the development process as well as to become key players involved in this process.

South-South cooperation

157. - One innovative feature of the SPFS had to do with the South-South cooperation. The objective of this initiative was to allow recipient countries to benefit from the experience and expertise of the more advanced developing countries. This was done by providing a considerable number of experts for two to three years to work in the implementation of the SPFS in the recipient countries. The experts worked directly with farmers in rural communities involved in the SPFS. Since FAO launched the scheme at the beginning of 1997, it was only active in one of the countries visited where it had achieved good results, although my staff noted that six of the 44 technicians assigned left prematurely (mostly for health reasons).

Monitoring of the programme in FAO

158. - Along with an Oversight Panel, consisting of eminent persons from outside the FAO who met once a year to review SPFS implementation in detail and provide advice, three ad hoc Committees, one monitoring unit and one special inspection system were put in place at Headquarters.

• The Special Programme Joint Committee, chaired by the Director-General, was composed of representatives at the highest levels of all the structures within FAO. Meetings were held with variable periodicity and the agenda took into account the different problems arising from SPFS implementation.
  
  
• The Special Programme Policy Committee, renamed Programme and Project Review Committee in June 1999, chaired by an Assistant Director-General assigned by the Director-General was established in 1996. It provided policy guidance to the SPFS and reviewed all the new projects to be submitted to donors (bilateral or multilateral) to ensure their harmonisation with the goals of the SPFS.
  
  
• The Special Programme Implementation Committee was in charge of overseeing field operations, from project preparation to project evaluation and assessment. It guided coordination between the SPFS and all segments of the Organization and between FAO and external partners.

159. - The SPFS Monitoring and Coordination Service (TCOS), within the FAO Technical Cooperation Department coordinated all aspects regarding SPFS implementation. The TCOS collected all information about SPFS and, particularly, the information sheets updated every three months by the SPFS National Coordinator in each country. Guidelines for completing these information sheets were provided by Headquarters in 1997 and again in 1999. The information gathered in these sheets was rather factual and focused mainly on practical achievements. It did not provide any appraisals.

160. - The SPFS Field Inspectors were appointed as consultants among retirees with experience in agricultural development and food security or staff members of international or multilateral institutions or agencies. They were under the authority of the Special Programme Implementation Committee and the Programme and Project Review Committee. They proposed an inspection programme for approval by the chairpersons of these committees. My staff noted that the reports resulting from these inspections mainly focused on technical topics and did not raise major policy problems to be discussed by the committees. According to the Organization, this was due to the fact that the activities at the time were implemented on a rather small scale where finding solutions to technical topics had been the most noticeable. As a result, however, the information sent back to the decision-makers was in fact no more enlightening than the information contained in the information sheets.

Budgetary and financial issues

161. - Under the Regular Programme, the funding of SPFS activities was normally to be provided under Programme 256 (entitled "Food Production in Support of Food Security in LIFDCs"). In one of the countries visited, my staff noted that the SPFS was initiated under the TCP funding. This was not an exception. Even though TCP funding only accounted for 8% of total funding, the indicative table drawn up by TCOS showed that 40 of the 74 participating countries at the time would be drawing some funding (29% on average) for initiating the implementation of one or more of the SPFS components under TCP arrangements. As shown in the case of the National Coordinators' salaries, such a situation might be a source of problems for the programme's implementation.

162. - Another drawback stemming from the Organization's rules and regulations was that the budget for SPFS implementation (subject to biennium limitation for the funding provided under Programme 256) in the forthcoming year needed to be submitted for approval on the basis of a plan of action determined by the results of the previous year. In many cases, however, harvesting occurred at the end of the year. The time it took, therefore, for the corresponding data to become available meant that the plans of action submitted in support of the budget arrived too late for funds to be released in time to set up the demonstrations. This situation constrained the national agencies implementing the demonstrations to advance the money. However, the liberalisation of agriculture made this increasingly incompatible with its juridical status (generally a government service) and could result in the demonstrations being suspended and a slowing down of momentum. This was particularly the case during the initial years, when farmers groups were not strong enough to advance the funds themselves.

163. - Other shortcomings noted in the budgetary and financial monitoring of the programme were as follows:

• The TCOS was supposed to maintain an overview of SPFS funding. However, for all the countries visited, the amounts shown in the indicative TCOS tables were only forecasts based on approved budgets for FAO funding and expected other funding. An overview based on effective payments from the donors would have been more accurate.
  
  
• Although budget revisions originated from the FAO Representatives, they had no control of disbursements made by Headquarters. Because of the problems encountered with FAS [paras. 184 to 190], they could not be informed of them on a timely basis. Reciprocally, the FAS problems prevented Headquarters to properly account for expenses incurred in the field. As a result, my staff could not always reconcile amounts disbursed in the field with the Organizations' Books. Furthermore, they could not be provided with an overall view of total disbursements for the SPFS at the time of their review. Eventually the difficulties noted also affected the reporting to donors.

Conclusion

164. - Overall, the review conducted by my staff led to positive achievements of the SPFS Phase I as far as they were able to observe in the countries visited. One of the lessons to be learned though was that the commitment of the participating country was a fundamental key success factor. Although it would inevitably vary from country to country, FAO should always make sure that the recipient country was clearly aware of its responsibilities. As a rule, I recommend that SPFS activities never start before the finalisation of all negotiations and the signing of the SPFS National Programme Document. The tendency, noted in some cases to start some activities before the finalisation of the NPD in order not to miss the agricultural season, should be avoided.

165. - Apart from the degree of commitment of the recipient country, other key success factors proved to be related to the level of the participatory involvement of farmers (individually and collectively) and to the mobilisation of field technicians able to convince farmers to participate, as well as to support them. Because of the importance of these factors, I recommend that a panel of experts be established to evaluate them in the first 15 countries where the SPFS has been initially launched. At the time of writing this report, my staff was informed of the Director-General's decision, in line with the request of Governing Bodies, to carry out an evaluation of the SPFS using a team of independent experts.

166. - The review my staff conducted also raised the following financial issues:

• As the Organization's policy was to implement the programme in all the countries that requested it, resources were, therefore, spread over a large number of countries, whereas they could have been concentrated on a more selective group. Furthermore, because of the financial constraints of the Organization's budget, resources could not be increased in line with the number of countries concerned. The amount budgeted under programme 256, which was US$ 9.2 millions for the 1998-99 biennium, was only increased to US$ 10.3 million for the 2000-01 one. At the time of writing this report, a total of over 84 requests had been received from the Member States to participate in the SPFS. The Programme was currently operational in 62 countries and under formulation in another 14 countries.
  
  
• Pressure, therefore, existed from the start on the recipient countries to mobilise funds from different donors to ensure the continuation of the activities. If actual mobilisation varied again from country to country, donors had been quite responsive in general. The support provided by the donor countries, the recipient countries, the international and regional financial institutions and banks, had been indeed essential in extending the SPFS coverage. This support should not be compromised by the difficulties noted in the reporting.

Decentralisation

Background

167. - The policy of progressive decentralisation away from Headquarters was approved by the Council, acting on behalf of the Conference with delegated power, at its 106^th Session in May/June 1994. As outlined in the Director-General's proposals, the objectives were to "enhance the use of national, sub-regional and regional capacities, achieve substantial economies in implementation modalities and shorten the time-lag between the expression of needs of Member Nations and their satisfaction by the Organization". The process involved the redeployment of staff in order to strengthen the Regional Offices, the abolition of the Joint Divisions, and establishment of Sub-regional Offices as well as the replacement of internationally recruited Programme Officers in FAORs with qualified National Professional Officers. Furthermore, the Council stressed the following:

• "the need to keep unity of purpose and to maintain effective interaction between the two sets of [normative and operational] functions";
  
  
• the fact that "staff resources and responsibilities should be judiciously distributed among the geographical levels of the organizational structure, allowing a sufficient critical mass to remain at Headquarters";
  
  
• "decentralisation should be accompanied by adequate delegation of technical and administrative policies and adjusted staff policies".

168. - The policy for decentralising operations to Regional and Sub-regional Offices was implemented in 1996 and was further developed throughout the 1998-99 biennium. In my previous report I stressed the following points:

• New delineation of authority between Headquarters, Assistant Director General/Regional Representatives and FAO Representatives was still in discussion at the time of writing the report.
  
  
• Redeployment and transfer of staff from Headquarters to the Regional Offices had not taken place in a satisfactory manner, while the vacancy rates in the Regional Offices were a source of concern.
  
  
• The linkage of financial and management information systems between Headquarters and the Regional Offices was inadequate.

169. - As a follow-up to previous reviews carried out in 1998, my staff visited two of the five Regional Offices between November 1999 and February 2000 in order to assess the state of implementation of the decentralisation of operations. These missions took place during or a few months after the implementation of Oracle at Headquarters and in the Regional Offices. Complementary information was collected through replies given to questionnaires sent to the other Regional Offices and examination of the documents that were provided to the Finance and Programme Committees.

Organisational arrangements

170. - Since 1996 budgets delegated to the Regional Offices have risen significantly, showing a clear transfer of budget authority to the Regional Offices. For example, in the two Regional Offices visited, budgetary allotments rose by 41% and 55% from 1996 to 1999, respectively. Nevertheless, the ability of the Regional Representative to become the focal point for the coordination of FAO's policies and programmes in his region did not increase in the same proportion. As a confirmation of this my staff observed that, in the Regional Offices they visited, the Regional Representatives:

• had not organised any meetings of the FAO Representatives of their region since 1996;
  
  
• had no say in the budgetary allocations issued to the FAO Representatives and did not check or even monitor Country Office field transactions accounted through FAS; and
  
  
• were not in a position to know about any duty travel carried out by Headquarters staff members in their region.

In fact, regional meetings between Regional Representatives and FAO Representatives had not been organised for a long time for budgetary reasons. However, my staff was informed that, in 2000, the 22 FAO Representatives that assumed operational responsibilities participated in a one-week course at the respective Regional Offices.

171. - For the rest, it should be noted that it has never been the Organization's intention to put the Regional Representative in a position to be the only hierarchical level between Headquarters and the FAO Representatives in his region. As defined by the Office for Coordination of Normative, Operational and Decentralized Activities (OCD) circular on responsibilities and relationships initially issued on 18 August 1998 and updated on 27 September 1999:

• FAO Representatives, as the senior representatives of the entire Organization in the countries of assignment, continued to report to the Director-General through the Director, OCD. Furthermore, they continued to have direct access to technical and operations units for day to day matters. They were instructed to copy their annual reports to the Regional and Sub-regional Representatives.
  
  
• Regional Representatives were not given any hierarchical responsibilities towards FAO Representatives. They were made responsible for identifying priority areas of action for the Organization and the overall planning of FAO activities in their region as well as the provision of managerial and administrative leadership and support for the implementation of approved programmes.

172. - As mentioned above, part of the decentralisation policy was implemented by outposting staff from Headquarters to Regional Offices in order to improve the efficiency and the effectiveness of the field action of the Organization. This concerned not only administrative and operational staff, but also technical officers. However, they were not all made to report to the Regional Representative only. In fact only regional Management Support Units (MSUs) clearly did. It is my opinion that as long as Operational Branches do not report directly to the Regional Representative, and technical officers still maintain direct links with Headquarters technical divisions, the powers of the Regional Representative will remain limited.

173. - Although, the quantitative and qualitative consequences of decentralisation were duly addressed in two documents (entitled "Report on decentralisation" and "Decentralisation of operational responsibilities to the country level", respectively) prepared for the Joint Sessions of the Programme and Finance Committee in May 1999 and September 2000, the intended study of actual cost and benefits was not carried out. An estimation of the savings arising from the process was indeed included in the former document, but the figure obtained (US$ 14 million per biennium) was calculated by identifying the staffing structure and cost before and after the decentralisation. Other actions generating either savings or additional costs were not included. For example, savings in travel costs were achieved. On the contrary, my staff noticed that the structural vacancy rates and training needs, especially due to Oracle implementation in 1999, caused the Organization to outpost, on a temporary basis, Headquarters staff members, hence balancing the savings expected by local staff appointments.

174. - While I do acknowledge that decentralisation is also a cultural process which takes time to take root and produce all the desired results, I still recommend that a thorough evaluation be promptly carried out to provide a better view of the main effects and results so far. This evaluation should take the full view of all FAO's decentralised structure and, as much as possible, measure progress against all the objectives adopted by the Council at its 106^th Session, as follows:

• faster response to Government requests and more effective delivery of services;
  
  
• regional, subregional and national capacity building;
  
  
• enhancement of technical excellence.
  
•  
• cost reduction.

175. - The Organization is currently setting up new arrangements for the field programme that will, inter alia, lead to the merger of the Regional Operational Branches (ROB) with the MSU in the Regional Office concerned and the transfer of operational responsibility of country level projects from the ROBs to FAO Representatives and regional technical officers. Whereas it is too early to assess the savings that might be generated by these developments, it is certain that the rapid succession of changes of responsibilities, roles and functions between the different levels of the decentralised structure of the Organization is a factor of destabilisation and demotivation for the staff and contributes to disorganise normal work. In such circumstances, cost efficiency of permanent reforms is doubtful. This gives one more reason to conduct the evaluation I recommend above before taking any new steps in the decentralisation process.

New financial system implementation

Background

176. - From 1985 to 1991, the FAO developed bespoke financial and personnel management systems, but it soon became apparent that these systems, known as Finsys/Persys, could not meet some of the basic requirements of the Organization. In its 78^th Session on 21-29 April 1994, the Finance Committee agreed with the need to replace them and discussed alternative options to determine an appropriate solution. On 13 June 1995 the Director-General imposed, as fundamental conditions, the fact that the new software must operate on the existing hardware, be a tried and tested software and have been in use for an acceptable number of years within organisations of a similar stature as FAO. At the time, the new system was expected to be operational by 1 January 1998 at a cost estimated at US$ 8 million (exclusive of FAO staff costs).

177. - As already mentioned in my previous report, both the deadline and the estimate eventually proved to be unrealistic. If the delays and additional costs encountered were in part inherent to a project of this magnitude and complexity, in the case of FAO though, two major factors also contributed to these developments:

• One was the hurried manner in which the more important and complex contracts were awarded at the end of 1995 in order not to let available budget authorisations lapse. Oracle applications were purchased on the basis of insufficient or inadequate definition of the Organization's requirements.
  
  
• The other cause of the difficulties was that the initial management structure proved inadequate and had to be thoroughly reorganised in February 1997. The former Steering Committee (SC) was replaced by a restricted Oracle Project Management Committee (OPMC) chaired by the Director-General. Staff members were assigned to the project full time and a new contract for consultancy services was signed with another consulting firm.

178. - I had mentioned in my previous report that a partial implementation of the original plan was expected to be achieved at the end of 1999 at a cost of US$ 17 million, including equipment, software and implementation costs. In fact, the system went live at the end of May 1999. Development costs incurred since inception of the project amounted to US$ 19.5 million as at 31 December 1999 (exclusive of FAO staff costs), while maintenance costs incurred since implementation amounted to US$ 1.5 million. Furthermore, the first year and a half of the new system in operation were hampered by major implementation problems. The following paragraphs analyse how this came about with a view to identifying the causes for such developments.

Audit scope

179. - The review conducted was based on:

• the examination of all preparatory reports submitted to the former SC and to the OPMC, their summary records since their first meetings (on 28 March 1996 and 30 April 1998, respectively) as well as all the documents related to the procurement aspect, and the ones provided to the Joint Sessions of the Programme and Finance Committees in May and September 2000;
  
  
• the questionnaires and interviews with staff of the Finance Division (AFF) and of the Information Systems and Technology Division (AFI);
  
  
• the findings of the Office of the Inspector-General (AUD) teams and of my staff since last May 1999 when visiting FAORs and Regional or Sub-regional Offices; and
  
  
• the results of the review conducted by AUD on the data migration, and the ones of the review conducted by an information management system consultant, whose services I engaged on the basis of Financial Regulation 12.8.

Implementation problems

180. - As already mentioned, major implementation problems were encountered during the first year and a half of operation. According to the two documents provided to the Joint Sessions of the Programme and Finance Committees in May and September 2000, the problems were identified in the following five areas:

• Field Accounting System (FAS);
  
  
• Data conversion;
  
  
• Clearing account management;
  
  

• Interface problems;
  
  
• System Support.

I commend the approach taken by the Organization in duly and honestly acknowledging these problems. While some of the difficulties encountered were inherent to such projects in any given organisation, I am of the opinion that, besides the overall issue of the functionality gap between Oracle applications and FAO requirements, two decisions contributed significantly in rendering the whole implementation much more complicated and risky. The first one had to do with the system used for the field transactions, while the second one was related to the date chosen for the go-live.

System architecture

181. - As already mentioned, Oracle applications (Oracle Financials, Human Resources, Payroll and Data Warehouse) were purchased on the basis of insufficient or inadequate definition of the Organization's requirements. These shortcomings had adverse consequences on the adaptation phase of the Oracle applications acquired. Any integrated commercial package is composed of a set of standard modules. To function, it must be adapted, i.e. configured, by consultants trained and certified by the software editor. The degree to which this adaptation can be performed through configuration depends on the flexibility of the specific application and the business requirements of the organisation concerned. In the case of FAO, the gap existing between what Oracle could achieve and what the Organization required led to substituting some of the functionality of the original package with other solutions and postponing the deadline for such implementation.

182. - In fact, only the core modules of Oracle Financials, together with a limited portion of Oracle Human Resources (HR) were in fact implemented at the end of May 1999 (Phase I). They were four major functional areas where the Organization did not use Oracle functionality and selected other products or bespoke solutions.

• The Oracle payroll package could not deal fully with the requirements of the UN common system of compensation. The study, which was carried out to determine the extent of the modifications required, concluded that at least US$ 3 million had to be spent to meet basic requirements. Once it was decided to maintain the legacy systems Finsys/Persys to process staff remuneration, the implementation of the remainder of Oracle HR together with Oracle Payroll was put on hold for a subsequent phase of the project (Phase II).
  
  
• To support the Oracle application in the area of field accounting, FAO lacked the communications infrastructure. It was estimated that Oracle could not be implemented without an investment of US$ 8 to 10 million to put the connectivity required in place. Furthermore, the Organization doubted that it could train and sustain the staffing required to operate such a system in many small Country Offices. As outlined in the paragraphs below, an existing software was adapted to cater for field accounting requirements.
  
  
• Oracle had neither the functionality to handle FAO's travel requirements (Daily Subsistence Allowance calculations, itinerary management, etc) nor the ability to support the information needs created by the volume of travel expenditures. Other options were considered and eventually, the Atlas system already used by another UN agency was chosen. The Procurement Committee granted a waiver of competition on 14 July 1998 to acquire the licence for US$ 250,000. Another contract was concluded on 17 February 1999 for system enhancement (US$ 54,000) and extended support (US$ 26,000).
  
  
• To cater for the budget functions (budget formulation, revision and forecasting), which were not within the scope of the Oracle applications purchased, a bespoke system (Budget Maintenance Module - BMM) was developed in-house.

183. - As a result of the inadequacy of Oracle applications in terms of payroll, travel and budget functions, the system architecture was rendered complex. Instead of a single commercial integrated package as initially envisaged (Oracle applications), several interfaces had to be built to link several sub-systems with Oracle Financials. The substantial difficulties met in the functioning of these various interfaces, which were mentioned in the documents provided to the Programme and Finance Committees, were in part the consequence of this complex architecture. As acknowledged in these documents, they were also due to insufficient testing, which resulted in the software bugs, found to be responsible for the interface problems, being discovered only after the go-live date.

Field Accounting System (FAS)

184. - Once the decision was taken to look for another solution, an initial survey of the requirements of the field accounting application was launched in January 1997. The main subsequent phases of the decision-making process regarding the FAS eventually implemented were as follows:

• During the SC 97/7 meeting on 11 September 1997, it was announced that a module from the World Bank (WB), the Field Headquarters Information System (FHIS), was being reviewed. An FAO mission to the WB had, in fact been organised on 22/24 July 1997.
  
  
• As reported to the SC 98/1 meeting on 29 January 1998, the results of the review of the FHIS versus FAO requirements were found to be encouraging. It was pointed out, however, that "its implementation by FAO [would] require changes to both the software and the procedures currently in place in the decentralised offices", and that the system was not going to be enhanced by the WB beyond the present level of functionality.
  
  
• During the SC 98/2 meeting on 26 March 1998, the range of US$ 250,000 - 300,000 was given as a preliminary estimate for the system modifications and interfaces with Oracle Financials. The matter was also discussed during the first OPMC meeting on 30 April 1998. It was indicated at the time that a proposal from the consulting firm, which had assisted the WB in the initial design and development, had been received.
  
  
• Following the decision of the OPMC to choose FHIS, a request for a waiver of competition was submitted to the Procurement Committee and granted on 12 May 1998. At the time, the Procurement Committee was informed that "the WB's experience with the system was positive with respect to the pilot implementation, which included two field offices. The system [was] the successor to another system used by the WB throughout the 1980s. In the mid-90's the WB decided to upgrade the system and completed about 90% of the work and testing. However, they had recently made a determination to put FHIS on hold following the decision to migrate to SAP software".

185. - The contract, which was concluded on 29 May 1998 for a total amount of US$ 576,500, stipulated that the consulting firm would deliver an application fully integrated with FAO's hardware and software environment by the end of the year. Its failure to do so led to the signing of an amendment to the original contract on 11 February 1999, for an additional amount of US$ 294,200, whereby the consulting firm was asked to:

• provide "maintenance and fixing of defects identified in the current release after 15 January and subsequent releases of the software for the duration of the contract".
  
  
• deliver "changes to functionality of Phase I (stemming from refinements of functional specs in the original contract)".
  
  
• give "assistance in transferring information technology know-how and hand-over of system administration to FAO staff".

A new schedule for deliveries was agreed: end of March 1999 for the new functionality and end of June 1999 for the system hand-over and the final version.

186. - The delivery of the final software version did not resolve the problems encountered. On the contrary, after four months of operation, use of that version had to be suspended in November 1999 because of numerous failures and defects. For one, the communication technology strategy for moving data from the FAORs to Headquarters, and reciprocally, failed for several reasons. The training programme organised proved to be insufficient and inadequate. The combination of all the failures created not only difficult technical problems, but also constituted a failed system from an accounting point of view. Ten thousands transactions that had been incorrectly accounted for and/or not processed had to be, individually, manually reviewed. It had taken AFF four months to sort out these items, putting a heavy toll on the division whose staff had to be diverted to confront this issue.

187. - According to the Organization, the strategy to adapt the WB's software was a good one, in part driven by the need to minimise the investment in development and justified by the fact that there were few packages on the market tailored to the operations of a UN-type field operations. Still according to the FAO, the problems encountered with FAS originated from the fact that the WB package (FHIS) was extended beyond its design scope, therefore, creating a more complex set of requirements. In my opinion, the strategy to choose FHIS in the first place, without seriously examining any other alternatives, was questionable because the software was not yet operational and was not going to be maintained for long by the WB.

188. - All these elements of risk and uncertainty were in contradiction to the Director-General's instruction, mentioned above, to acquire a tried and tested system in use for an acceptable number of years. Also, my staff noted that the technical justifications used to support the decision to choose and adapt FHIS were only based on "a preliminary analysis of the FAO requirements vis-�-vis the WB system functionality" as mentioned during the first OPMC meeting. Regarding the selection of the consultant it was presented as closely associated with the choice of FHIS so no competition was organised. The firm was selected only on the assumption of its knowledge of the software to be customised and implemented.

189. - In conclusion, I am of the opinion that the many difficulties experienced by the Organization originated, in the first place, from the WB's choice of software. Then, the lack of serious analysis of the technical environment and the integration with Oracle applications, combined with insufficient testing and training, further compromised the application's implementation in May 1999. In the end, FAS, at least in its first version, proved to be costly and ineffective (US$ 870,700 in development costs, training and hardware related costs excluded).

190. - It was only in June 2000 that a revised interim version was released. Based on the documents my staff was provided with, the major changes introduced were related to a more limited but manageable data import/export facility between the field and Headquarters, and control edit mechanisms to limit data entry errors. My staff will make their own assessment of the situation during the 2001 field audit missions.

Premature go-live date

191. - Apart from the testing already mentioned, the other areas, which made the May 1999 go-live premature because of insufficient preparedness, were connected to implementation related activities, which encompassed the following hardware improvement or replacement; cleaning and migration of the old system data to the new one; training of the new system users and issuance of new manuals and procedures; and set up of the new organisation. If hardware improvement or replacement was duly carried out, the other areas had not been, in my opinion, sufficiently covered.

Data cleaning and migration

192. - This was particularly evident in the case of data cleaning and migration, which required reconciling old and new data structures and operating procedures often incompatible. Being one of the most complex aspects of any integrated system implementation, it would have required extensive cleaning-up of the existing data prior to the conversion and a detailed written strategy on such. While I do acknowledge that cleaning had indeed been carried out, I am of the opinion that it was not sufficient, especially in the field of advances in particular, and in the reconciliation of all Finsys sub-ledgers with Finsys General Ledger in general.

193. - As far as the actual conversion was concerned, my staff was indeed provided post factum with a comprehensive document reconciling Finsys closing balances with Oracle opening ones. However, this document was only provided to them on 21 February 2000 because of the long migration process outlined below. Furthermore, they had never been provided, beforehand, with a formal document, setting out the plan for the migration, which would have detailed the table of correspondence between the old system and the new one; the nature of the data to be migrated; the programmes to be eventually built to perform the work; and the reconciliation procedures to be carried out to ensure data integrity and accuracy.

194. - The reason why my staff could not be provided with such a document was simply due to the fact that, a formally documented overall strategy for the conversion had never being put together, in spite of significant amount of staff time dedicated to the issue and numerous working papers produced. Because of insufficient preparation, data migration, therefore, became a troublesome implementation problem. The legacy system (Finsys) was closed down on 25 May 1999, and all current transactions from 26 May 1999 onwards were posted in Oracle Financials. However, the conversion process of the closing balances of Finsys lasted several months from May 1999 to March 2000. Because of that, it was not possible to close the books on a monthly basis starting June 1999. When the books were eventually closed, a number of clearing accounts had several thousands of transactions that were not properly matched. Furthermore, the intended AUD's review on the conversion process in the third quarter of 1999 (in order for the implementation of recommendations before the biennial closure) had to be postponed since Oracle opening balances were only finalised on 21 February 2000. The review, only completed on July 2000, became a post factum one, which undermined its purpose.

Training and issuance of new manuals and procedures

195. - The area of training and issuance of new manuals and procedures also suffered from insufficient preparedness. As already mentioned, training in the field proved to be inadequate for FAS, partly because the staff had neither the accounting knowledge nor the computer skills necessary to use such a system. In Headquarters, training was done selectively and in many cases lost its effectiveness because of the postponement of the go-live date. A training programme had been duly formulated in 1998 but the corresponding resources had not been provided for. As a result, because funding was not available to appoint external trainers, only a limited number of in-house skilled staff was available to prepare training materials and carryout training.

196. - The issuance of new procedures and manuals was also hampered, notably by the lack of resources within AFF to confront the issue. For example, the drafting of new account descriptions was only completed upon my staff's insistence in June 2000. While I commend the creation, on the Intranet, of a special section devoted to Oracle, my staff noticed that several documents would need to be up-dated, together with various sections of the FAO manual.

Organisational changes

197. - As far as the necessary changes to the Organization's set up were concerned, the former MSUs were closed down gradually throughout the year in order to limit disruption. Their revamping into a single Management Support Service (MSS) within the AF Department was only effective in September 2000, more than a year after the Oracle implementation. As far as the restructuring of the Finance Division was concerned, several changes were announced, in July 1997, in order to align its structure better with the one of Oracle applications to be implemented. The plan at the time was to reorganise AFF in three different services, namely "Central Accounting, Reporting and Control Service", "Payables and Receivables Service" and "Treasury Operations Service". However, since most of the changes could not be implemented until Oracle was operational, the restructuring that took place between July 1997 and December 1998 was limited.

198. - The main organisational change was undertaken in January 1999. Following the decision to merge Payables/Receivables and Treasury into one service, the creation of the "Receipts, Payments and Treasury Service (AFFR)" was announced in a memorandum from the AFF Interim Director dated 5 January 1999. In the organisational chart attached thereto, the total number of staff indicated was 93 (27 professional and 66 general service), which was lower than the division's current staffing of 104 (27 professional and 77 general service). In addition to the 97 posts authorised by the Programme of Work and Budget (PWB) for 1998-99, seven temporary posts had been established following the transfer of imprest accounting work and staff from the Field Operations Division (TCO).

199. - In August 1999, a new modification of the AFF re-organisation was implemented. Following the Director-General's concern over controls, a "Disbursement Confirmation and Control Unit" was created within the Controls Unit, and the Controls Unit itself was separated from the AFFC Service to which it belonged, and reported directly to the AFF Interim Director. The proposal was for this Disbursement Confirmation and Control Unit to be staffed with four professional and two general service staff. To date, AFF staffing has not been increased to reflect this decision.

AFF staffing

200. - At the time the new AFF restructure was being put in place, the preparation of the PWB for 2000-01 was also under way. The first draft of this document in Spring 1999 was based on a reduction of six posts for AFF from the 1998-99 staffing of 104, as a result of the loss of WFP Treasury income. In July 1999, it was proposed to reduce the number of staff by seven, but with a different grade mix in order to increase professional capacity and strengthen financial control (37 professional and 60 general service). It was anticipated that the new posts, together with the eight vacancies existing at the time, be partly filled using the internal staff from the MSUs. Compared to the PWB 2000-01 proposal, the budgetary cost was estimated at US$ 848,000 per biennium taking into account the loss of WFP treasury income, or US$ 1,478,000 per biennium should WFP income be reduced further. It was, however, pointed out that (i) "the elimination of AFF services to WFP would result in a loss of income which [exceeded] the cost of providing these services, and (ii) with the substantial re-centralisation of processing functions from the MSUs to AFF, there would be more than enough savings in the MSUs as a result of Oracle to cover the above AFF increase".

201. - Further cuts were in fact made to AFF's budget in the final stages of the formulation of the PWB 2000-01 (August 1999). Compared to the 1998-99 authorisation, ten posts were cancelled as a result of the loss of the WFP income (six for treasury income and four for the expected loss of payroll income), while the seven ex-TCO posts, covered in the past by a yearly allotment, were included under the temporary assistance component of the budget. As a consequence, the total staffing proposed for AFF for 2000-01 amounted only to 87 (26 professional and 61 general service). However, as mentioned in the document presented to the Conference (C99/3), further changes in the division were anticipated, since the AFF might "need to undergo further reorganisations to align itself to the new business processes of the Organization, based upon the experience gained during the early stages of implementation of the system".

202. - One of theses changes occurred in October 1999 when the Director-General decided, as an interim measure, to make US$ 2.2 million available to AFF. Part of it was funded by (i) the reduction of AFF's temporary assistance budget by US$ 360,000; (ii) the reduction of the MSS by US$ 469,000; and (iii) the transfer to AUD of the responsibility for auditing of liaison offices. For the remainder, it was indicated that savings would be sought "from elsewhere in the budget in the hope of finding the remaining shortfall of US$ 1.1 million".

203. - Following a series of meetings and discussions, the Director-General approved the abolition of a further general service post and the establishment of 11 additional professional posts on a temporary basis for a period of up to three years. As reported to the Joint Programme and Finance Committees in September 2000, "a number of factors combine to make it essential that the AFF structure be reviewed and strengthened at the present time. These include:

• the demand for specific technologically advanced skills to take best advantage of Oracle's facilities;
  
  
• the revised work flows arising from the restructuring of the departmental management support units and the creation of a centralised Management Support Service;
  
  
• the increase workload in AFF arising from the demand for enhanced internal control; (e.g. disbursements control, local audit reviews, etc.); and
  
  
• the significant budget cuts to AFF during the recent period of severe budgetary constraints (declining by 27% since 1992-93)".

204. - Although measures were indeed taken to align the structure of the AFF division to the new requirements created by the implementation of Oracle applications, I am of the opinion that the initial assessments made had a lot of shortcomings and tended in general to err on the side of optimism. The centralisation into AFF of many functions previously performed by the Departments' MSU, combined with the necessity for the division to be involved in the system control, assurance and analysis, should have normally led to an increase in the staffing rather than a decrease at least before the actual implementation and in its initial stages. Then, the budgetary process did not provide a viable establishment for AFF for 2000-01.

Consequences of the difficulties encountered in the implementation of the new system

205. - The questions of AFF resources and staffing levels on the new system project had been inextricably linked. Furthermore, they had to be put in the context of the overall resources available to the Organization, whereby FAO had been obliged to operate under a zero nominal growth budget for the past few biennia. The reduced staffing combined with the additional work of coping with the problems encountered with the implementation of the new system put a heavy burden on the division resources. The situation was made even more difficult by the delays encountered in the recruitment process to fill the vacancies, when recruitment was indeed initiated. At the time of writing this report, my staff was informed that 15 posts were still vacant in the Finance Division (10 for the professional and five for the general service staff, respectively, with three posts of the latter being held to use the funding for consultants).

206. - To alleviate part of the burden, the division had an increased use of consultants from 1999 onwards. Apart from the key position of the division Director, which continued to be filled since March 1997 on an interim basis, use of consultants was also noted in the following areas:

• The staffing required for the Control Unit on the professional side has been largely provided so far in the form of consultants under honorarium consultancy contracts or Special Service Agreements (SSAs).
  
  
• A consultant, who was recruited under an SSA in the General Ledger Unit as of November 1999, had his contract repeatedly renewed since.
  
  
• In addition to three retirees, three consultants were given SSAs for a duration of 11 months in July 2000, to assist with the processing of invoices in the Accounts Payable Unit.

207. - Although increased, the use of consultants was, however, not sufficient to bring the level of the division's resources into line with its needs. This was particularly evident in the Controls Unit, which had not been able to put into place the control environment required by the new system because its staff was diverted to other units to deal with critical assignments. Similar evidence of understaffing existed in other areas, most notably in the following:

• With the implementation of Atlas, the number of travel expense claims processed by the Travel Unit trebled in 1999 compared to 1998, but its staffing did not follow the same trend. As a result, controls were neglected in several areas. For example, if the amounts invoiced by the Organization's travel agent were duly checked in line with the travel authorisations issued, the Unit had neither the resources nor the skills to make sure that FAO was indeed getting the best fares. The same remark could be made about the lack of control over shipping and insurance expenses, except in a narrow approval sense.
  
  
• In my previous report, I had commented on the inadequacy of the Treasury Unit organisation and recommended its revamping and upgrading. No steps were taken in this direction during the 1998-99 biennium and the situation even deteriorated in 2000. The two Treasury professional staff members left the Organization in January and April 2000, respectively, and the general service with experience of treasury investments and funds transfer procedures was given another assignment. As a result the consultant, whose services had already been contracted to assist the Organization on investment issues starting February 2000, ended up running the Treasury Unit until the position of Unit Chief was eventually filled by a transfer of an internal staff member in June 2000.

System support

208. - Because of all the above-mentioned problems, system support in general became another problematic area. Faced with a huge amount of work and limited resources, the Organization proved, not only, unable to reorganise itself in line and time with the new system structure, but also failed to take full advantage of the scope and power of the software functionality. While being perfectly aware that improvements could be made, it had not been possible to act on them because of the diversion of both functional and technical staff capacity to other problem areas. Many examples could be given. They range from strengthening the cross-validation rules (creation of inter-dependencies in the segments of the accounting flexfield, restriction and centralised control over the open/closed accounting periods, etc.) to improving the overall system efficiency (analysis of the rejected transaction, automation of the generation of encumbrances for Atlas, etc.).

209. - One specific area of concern had to do with the reporting to donors. Reports not only suffered from issuance delays but also inaccuracy. As pointed out in the documents provided to the Joint Sessions of the Programme and Finance Committees, the issue at stake was complex because of the conjunction of data conversion discrepancies, transaction errors and overall reporting difficulties. The problems not only affected FAO's staff, especially projects' managers who tended to overcome their absence of knowledge on their budget allocations by maintaining parallel manual accounting systems. They also affected the donors themselves who, dissatisfied with the deficient information on the use of their funds, might become more and more reluctant in entrusting FAO with their money. This would undoubtedly affect FAO's image and also make its financial position more difficult.

Overall conclusion

210. - With the implementation of its new financial system, the Organization has embarked on a significant change process. Despite the above mentioned implementation problems, the concepts, structures and foundation to support a more effective and efficient FAO have been put in place. The current system is functioning and most of the troublesome areas had been addressed at the time of writing this report. The agenda for the balance of the 2000-01 biennium would, however, continue to impose a very heavy workload on both AFF and AFI. In addition to the remaining implementation problems, which still need to be solved, a major Oracle upgrade would have to be carried out in order to replace the current version of Oracle Financials (version 10.7), which should not be maintained by the editor beyond 2001, with a more recent one (version 11i). Based on what my staff could observe so far, progress appeared limited at this stage. I therefore urge the Organization to duly take into account the lessons learned from the experience of the 10.7 version to prevent history repeating itself. In particular, the fact that "the project had been understaffed and under-funded from the outset and that the struggle for adequate resources had played a major role in FAO's inability to address and resolve the problems quickly", duly mentioned in the documents provided to the Joint Session of the Programme and Finance Committees in May 2000, should be meditated.

OTHER MATTERS

Cases of fraud

211. - Pursuant to item 6 (c) (i) of the additional terms of reference governing external audit, on 14 April 2000, my staff requested information pertaining to cases of fraud or presumptive fraud known to the Organization for the 1998-99 biennium. On 29 September 2000, they were provided with a reply listing five cases considered as fraud and presumptive fraud amounting to US$ 103,653.13. The reply mentioned though that the list was prepared without the benefit of a common definition of fraud or presumptive fraud for the United Nations system. At the time of writing this report, most of the misappropriations or overpayments had been recovered, with the notable exception of one case of falsification of cheques amounting to US$ 74,480. The criminal aspects of the matter were under investigation by the local police authorities. My staff did not conduct any investigation on these five cases, which were found to have been duly handled and followed up.

212. - Detailed investigations were, however, conducted for another case not considered legally as a case of fraud by the Organization. According to information communicated to my staff on 22 June 1998, the issue at stake was the acceptance of a gift (a trip to Japan in mid-February 1998) by a staff member from an external source. A disciplinary procedure for misconduct was initiated since it constituted a breach of staff regulation 301.016, aggravated by the fact that the staff member concerned (Director's grade) was involved in the Organization's procurement, and that the gift came from a major supplier. The procedure was, however, abandoned when the staff rendered his resignation on 26 May 1998, which was accepted by the Director-General on 16 June 1998 (effective 31 July 1998). Nevertheless, since additional information came to light during the investigation of the above-mentioned matter, my staff was promised to be kept informed of any other developments.

213. - On 17 December 1998, my staff was informed of the nature of the follow-up investigations. These investigations had been centred around the question as to whether there was a link of a legal or control nature between a company, in which the spouse of the staff member concerned held shares, and companies that had been contracted out by the Organization. My staff was also informed of the fact that the follow up to this specific investigation did not result in any evidence of fraud or presumptive fraud, the elements at the Organization's disposal being insufficient to establish the existence of any of the above-mentioned links. On that basis, the Organization considered this particular matter closed. On 15 January 1999 my staff asked to examine the supporting documents on which this conclusion was based. Their preliminary examination led to further requests. All files and correspondences concerning this case were reviewed. Furthermore, several staff members working in the same service as the staff member concerned were interviewed.

214. - The review conducted by my staff led to a different appreciation of the situation considering the following facts discovered by the Organization in the follow-up investigation:

• Serious and converging elements, which indicated possible collusion between the staff concerned and persons being partners or having some interests in several companies contracted out by the Organization, indeed existed.
  
  
• One case of fraud was discovered by the Organization in its contractual relationship with one specific firm. The staff member concerned had issued two false certificates, which resulted in irregular payments for a total of ItL. 19 million. Upon discovery, this amount was recovered through deductions from subsequent payments due to that firm, and the maintenance agreement with the firm was terminated. Further to this termination, a request for arbitration was notified to the Organization on 16 February 2000.

215. - Regarding the latter, which was still sub judicial at the time of writing this report, my staff was informed of the following:

• In line with a general decision of the Conference accepting the representation and defence by the Avvocatura Generale dello Stato of Italy whenever issues concerning its legal status in Italy were involved, the Organization appointed an arbitrator and a lawyer from the Avvocatura.
  
  
• The Arbitration Tribunal was constituted on 26 May 2000 and issued a partial award on 25 September 2000 on the applicable law. On 15 October 2000 the Arbitration Tribunal heard the Assistant Director-General, AF. On 15 December 2000 it heard several witnesses, including three officers of the service concerned. In addition, as part of its powers of investigation and collection of evidence, it requested the submission of a considerable amount of documents, which were provided by the Organization.
  
  
• The converging elements indicating possible collusion between the staff member and persons being partners, or having some interests in several companies contracted out by FAO, were extensively put forward by the Organization in its pleadings as one of the elements for the termination of the contract, and was researched into by the Arbitration Tribunal.

216. - According to the Organization, since these facts were discovered after the acceptance of the staff member's resignation, it was considered "the best course of action not to reopen his resignation and to embark on a new discipline procedure, with all the legal complication and uncertainties". While I do respect the fact that the discipline is clearly of the Director-General's responsibilities, I cannot concur with such arguments. To start with, the strong suspicions that existed before the acceptance of the staff member's resignation should have justified, in my opinion, investigations at the time. In this absence, the importance of the facts subsequently discovered would have rendered the "reopening" of the case more than appropriate.

217. - Following allegations that the staff member concerned had given his staff instructions to work for the private interest of another staff member, my staff conducted their own investigations into the matter. It was in this context that the above-mentioned interviews took place. They disclosed that several of the FAO's most senior management had in fact benefited from the services of the staff member concerned, or of his subordinates. Most of the services rendered were negligible. As such they only constituted breaches of the standards of service of the International Civil Service, which emphasised the fact that "the conduct of supervisors must be free of intimidation or personal favouritism and that solicitation or acceptance by them of favours, gifts, or loans from their staff must not be practised or even suspected".

218. - In two cases, however, the use of the Organization's resources was demonstrated. Once the facts were uncovered by my staff, steps were taken by the Organization to recover the undue expenses in one case. At the time of writing this report I have been informed that the amount involved (US$ 7,880 corresponding to 20.5 days of work performed by four staff members intermittently over a period of two years) was recovered from the staff member's terminal emoluments. No action was taken in the other case. The person concerned, when the involvement of FAO staff was revealed to him, stated that this had happened without his prior knowledge or consent and, therefore, the Organization considered that reimbursement could not be claimed.

219. - The above-mentioned cases raised, in my opinion, the issue of internal control on the use of the Organization's resources. I considered that the announcement by the Director-General in March 2000 of an internal investigation on the matter to be a necessary step. At the time of writing this report, my staff had been provided with the terms of reference for the investigation and I look forward to receiving the results.

Acknowledgement

220. - I wish to record my appreciation of the cooperation and assistance extended by the Director-General and his staff during the audit.

Pierre JOXE
Premier Pr�sident de la Cour des Comptes
de la R�publique Fran�aise
External Auditor

27 February 2001